From: Adam Nemet <anemet@apple.com>
Date: Wed, 30 Jul 2014 16:51:24 +0000 (+0000)
Subject: [AVX512] Add some of the FP cast intrinsics
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=c871ff95f3b2caac6e0b8d73571ba1237479b986;p=platform%2Fupstream%2Fllvm.git

[AVX512] Add some of the FP cast intrinsics

Part of <rdar://problem/17688758>

llvm-svn: 214315
---

diff --git a/clang/lib/Headers/avx512fintrin.h b/clang/lib/Headers/avx512fintrin.h
index a336f6d7..2d3a8bb 100644
--- a/clang/lib/Headers/avx512fintrin.h
+++ b/clang/lib/Headers/avx512fintrin.h
@@ -117,6 +117,33 @@ _mm512_set1_epi64(long long __d)
   return (__m512i)(__v8di){ __d, __d, __d, __d, __d, __d, __d, __d };
 }
 
+/* Cast between vector types */
+
+static __inline __m512d __attribute__((__always_inline__, __nodebug__))
+_mm512_castpd256_pd512(__m256d __a)
+{
+  return __builtin_shufflevector(__a, __a, 0, 1, 2, 3, -1, -1, -1, -1);
+}
+
+static __inline __m512 __attribute__((__always_inline__, __nodebug__))
+_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 __attribute__((__always_inline__, __nodebug__))
+_mm512_castpd512_pd128(__m512d __a)
+{
+  return __builtin_shufflevector(__a, __a, 0, 1);
+}
+
+static __inline __m128 __attribute__((__always_inline__, __nodebug__))
+_mm512_castps512_ps128(__m512 __a)
+{
+  return __builtin_shufflevector(__a, __a, 0, 1, 2, 3);
+}
+
 /* Arithmetic */
 
 static __inline __m512d __attribute__((__always_inline__, __nodebug__))
diff --git a/clang/test/CodeGen/avx512f-builtins.c b/clang/test/CodeGen/avx512f-builtins.c
index ddeb3b6..8a2dd07 100644
--- a/clang/test/CodeGen/avx512f-builtins.c
+++ b/clang/test/CodeGen/avx512f-builtins.c
@@ -102,3 +102,10 @@ __m512d test_mm512_set1_pd(double d)
   // CHECK: insertelement <8 x double> {{.*}}, i32 7
   return _mm512_set1_pd(d);
 }
+
+__m512d test_mm512_castpd256_pd512(__m256d a)
+{
+  // CHECK-LABEL: @test_mm512_castpd256_pd512
+  // CHECK: shufflevector <4 x double> {{.*}} <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
+  return _mm512_castpd256_pd512(a);
+}