Runtime: fixed the region check for three rect region related APIs.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

GBE: improve asin/acos precision

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Tested-by: Zhigang Gong <zhigang.gong@linux.intel.com>

GBE: register width should not exceed execution width

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

GBE: Do not change vertical stride when it is 0

It will change scalar register g3<0,1,0> into g3<16,1,0> which illegally
crosses more than 2 adjacent rows.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

GBE: Fix null register to integer type

GEN 'mach' instruction only support integer type register.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

Fix float to ulong/long fail.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Fix signed to unsinged type sat convert.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@intel.com>

Refine isnan builtin.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@intel.com>

Add FCMP UNO support.

And also correct some UXX compares.
V2: Not use OCL_OPTIMIZE_IMMEDIATE for XOR and ORD compare.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@intel.com>

GBE: filter the unsupported cl compile arguments out.

As the unsupported argument may trigger unexpected compilation
error, we just remove them from the arglist.

If latter clang's cl frontend supports these arguments, we need
to revisit here.

This patch also add a new environment variable
OCL_OUTPUT_BUILD_LOG.
If this variable is set to 1, GBE will print the compile log to
the standard error channel (llvm::errs()). By default, it is false
and GBE will not print any build log.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

When local_work_size is null, try to choose a local_work_size.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>

Complete the feature of clGetEventProfilingInfo API

The profiling feature is now all supported. We use
drm_intel_reg_read to get the current time of GPU
when the event is queued and submitted, and use
PIPI_CONTROL cmd to get the executing time of the
GPU for kernel start and end.
One trivial problem is that:
The GPU timer counter is 36 bits with resolution of
80ns, so 2^36*80 = 5500s, about half an hour.
Some test may last about 2~5 min and if it starts at
about half an hour, this may cause a wrap back problem
and cause the case fail.

Signed-off-by: Junyan He <junyan.he@linux.intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Fix B/UB compare fail.

Because B/UB is treated as W/UW, so can't set src1's type when dismatch.
Set the correct type before getRegisterFromImmediate.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Use -O1 when -cl-opt-disable, for inline function.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Remove test cl_create_kernel.

This test only try to allocate buffer with size large than CL_DEVICE_MAX_MEM_ALLOC_SIZE, and
assert if return status if not CL_INVALID_BUFFER_SIZE. But in openCL spec:
Implementations may return CL_INVALID_BUFFER_SIZE if size is greater than
CL_DEVICE_MAX_MEM_ALLOC_SIZE value specified in table 4.3 for all devices in context.

It don't must return CL_INVALID_BUFFER_SIZE. So remove it.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Runtime: implement the get build log function and fix one build error check issue.

According to spec, we need to support CL_PROGRAM_BUILD_LOG which is
used to get the build log of a cl kernel. And we also need to check
whether a build failure is a generic build fail or a build option
error. This commit also fix the piglit case:
API/clBuildProgram.

Another change in this commit is that it reroute all the output
of the clang excution to internal buffer and don't print to the
console directly. If the user want to get the detail build log,
the CL_PROGRAM_BUILD_LOG could be used.

v2: include both clang error messages and the llvm-to-gen error
messages. Also refine the checking for the error buffer parameter.
If there is no error buffer specified, always flush the build log
to llvm::errs().

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

CL/Runtime: workaround the unused sampler_t kernel argument.

Current implementation is to use a normal integer to represent
a sampler_t, then later when the sampler is used in read_image
or get_sampler_info, the backend will fixup its type to SAMPLER.

But some test case in piglit will define a sampler_t kernel argument
with an empty kernel budy. Then we will not have a chance to fixup
the kernel argument type to sampler, then we will fail at runtime side.

To workaround this issue, we change the sampler_t to short type.
Then when the user call clSetKernelArg to set a sampler, it will pass
in a pointer size with a short value argument type. It will fail
the size checking logic, then we fixup its type to sampler there.

As this workaround will only take effect when error occur, it will
not bring too much side effect to the normal cases. And it can
pass the existing test cases.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

Runtime: fix some piglit failures.

compiler_available should be true. And when a program is retained, we should
not call build on it again.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

Runtime: fixed one missing case for clGetKernelWorkGroupInfo.

CL_KERNEL_PRIVATE_MEM_SIZE is not implemented, this patch fix
this issue and can pass the piglit test case.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

Runtime: fixed parameter error checking in cl create buffer.

This patch can pass piglit test case cl-api-create-buffer.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

Add the drm include and lib path for find when drm is not the system one.

Add the support when the DRM lib is not in the system standard location.
In some cases, we want to debug the libdrm but not want to influence the
whole system.

Signed-off-by: Junyan He <junyan.he@linux.intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Enlarge the global mem size.

When create image, due to alignment, will casue size large than max alloc size.
Enlarge the global memory size and using it to check size when alloc.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Fix some get image info errors.

Get correct grf offset and need clear image set offsets.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Fix a build problem when the llvm version has the fix version digit.

If the llvm version is something like 3.3.1, the previous cmake script
will generate an incorrect cflags as: -DLLVM_33 1 which breaks the build.

This commit also update the stable llvm version from 3.1 to 3.3.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

Add vload_half and vstore_half build in.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Add convert between fp16 and fp32.

Use convert instruction in ir, and ALU1 in gen selection.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Fix a compare immediate optimize error.

When do LOADI/compare -> compare optimize, IMM src1 will using LOADI type,
but LOADI doesn't  care unsigned or signed. Should use the compare type.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

Add FCmpInst ord support.

Because gen do not support isorder direct, use (src0 == src0) && (src1 == src1).
BTW: can't use !unordered.

v2: Refine, don't need AND.
v3: Do not change getGenCompare function.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

fix builtin function "round"

previously using round to even, the result was wrong.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

improve builtin function "rint"

directly use __gen_ocl_rnde

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

fix builtin function "isnormal"

fix a corner case of very small input

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

put a mutex around gbe_program_new_from_llvm

because random crash happens if without the mutex

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

fix ASR operator for 64bit integer

if operand is positive, then pad zero at high 32 bit.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

Remove boolean values cannot cross their definition basic block restrict.

Add mov bool support.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

fix builtin function "ilogb"

add FP_ILOGB0, FP_ILOGBNAN
return FP_ILOGB0 for zero.
return FP_ILOGBNAN for nan.
return INT_MAX for inf.
also improve function code for other cases.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

fix builtin function "ldexp"

fixed corner cases when input parameter has special value

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

fix builtin function "nextafter"

fix for some corner cases

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

fix builtin function "fdim"

check whether input is NaN. fix the code if input is inf

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

GBE: fix the constant data allocation.

Need to keep consistency between the constant data
allocation and the constant register allocation.
So we need to skip the unused constant data at the
constant data allocation stage.

To avoid possible mismatching, add a new assert in
the constant register(address) allocation stage to
make sure the address register match the eaxct constant
data.

Also modify the constant utest slightly to hit this
code path.

Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: Ruiling Song <ruiling.song@intel.com>

GBE: remove all vstore macros for constant memory space.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Ruiling Song <ruiling.song@intel.com>

Add bitcast support between vetor and scalar type.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Add other unsigned interger types mask type of shuffle and shuffle2.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

GBE: Add support for kernel attribute reqd_work_group_size.

When a kernel has __attribute__((reqd_work_group_size(X, Y, Z))) qualifier,
the kernel will only accept that group size.

v2: add binary load/store support.
v3: fix the MDNode parsing according to spir spec. It's using the following
structure rather than a tbaa tree.

!spir.functions = !f !0,!1,...,!N g
; Note: The first element is always an LLVM::Function signature
!0 = metadata !f < function signature >, !01, !02, ..., , !0i g
!1 = metadata !f < function signature >, !11, !12, ..., , !1j g
...
!N = metadata !f < function signature >, !N1, !N2, ..., , !Nk g

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

GBE: handle half type size

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Tested-by: "Sun, Yi" <yi.sun@intel.com>

Runtime: complete the api clGetKernelWorkGroupInfo.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>

ignore a clang unsupported building option

IVB does not support float denorm value.
So the building option "-cl-denorms-are-zero" can be safely ignored.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

gbe_bin_generator: should not use append option when create new binary.

We should use trunc option rather than app when we need to create a new
binrary file.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: Ruiling Song <ruiling.song@intel.com>

Fixup the problem of CL_PROGRAM_BINARIES in clGetProgramInfo API

clGetProgramInfo using CL_PROGRAM_BINARIES to get the binary will
not be right because the binary got is not the serilization one.
Add the serilization there to fix this bug.

Signed-off-by: Junyan He <junyan.he@linux.intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

fix builtin function "fmax"

if an parameter is nan, then returns another parameter.

v2: no need to test nan for integer
Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

GBE: Fix alignment for private variables

Private variables allocated on the stack should be aligned according to OCL spec.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Tested-by: Zhigang Gong <zhigang.gong@linux.intel.com>

GBE: Fix alignment according to OCL spec

The patch simply store a 'align' for each kernel argument.
Then the runtime could align the kernel argument address to 'align'.
This patch works for constant and local address space.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Tested-by: Zhigang Gong <zhigang.gong@linux.intel.com>

GBE: Remove max_limit for struct alignment

a struct may have vector field (like int8/16), max_limit is meaningless.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Tested-by: Zhigang Gong <zhigang.gong@linux.intel.com>

release context in runtime_createcontextfromtype

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Move the gpgpu struct from cl_command_queue to thread specific context

We find some cases will use multi-threads to run on the same queue,
executing the same kernel. This will cause the gpgpu struct which
is very important for GPU context setting be destroyed because we
do not implement any sync protect on it now.
Move the gpgpu struct into thread specific space will fix this problem
because the lib_drm will do the GPU command serialization for us.

Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Zou, Nanhai" <nanhai.zou@intel.com>

Add the clGetMemObjectInfo options for sub-buffer and update the utest case

Signed-off-by: Junyan He <junyan.he@linux.intel.com>
Reviewed-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Add the test case for sub buffer check

Signed-off-by: Junyan He <junyan.he@linux.intel.com>
Reviewed-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Implement the clCreateSubBuffer API

Signed-off-by: Junyan He <junyan.he@linux.intel.com>
Reviewed-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Add the bo's internal offset support when do drm_intel_bo_emit_reloc

Signed-off-by: Junyan He <junyan.he@linux.intel.com>
Reviewed-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

GBE: fix a 64bit scalar register issue.

For scalar register, should use stride 0.
also change the unit test to hit the point.

v2: fix h2()

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

improve multithread calling of llvm

call llvm multithread function instead of using a semaphore.
also exit llvm multithread mode at the end of life.

v2: not call llvm::shutdown() if llvm is older than 3.4
Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>
Tested-by: Zhigang Gong <zhigang.gong@linux.intel.com>

fix builtin function "fract"

v2: return nan for nan, +zero for +inf, -zero for -inf.
Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

fix builtin function "copysign"

using better algorithm

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

fix builtin function 'frexp'

returns correct value for nan or inf.
also returns correct value for very small float value.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

release previous program in cl_kernel_init

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

release previous kernel in cl_kernel_init

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

Runtime: fix some max/alignment values.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

Runtime: fix one bug in clGetProgramInfo.

The CL_PROGRAM_BINARIES forget to return the param value size.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

GBE: Don't modify argument 0 of the get image information instruction.

When the first round compilation fail, GBE will turn to recompile the
sample program by using another profile. If we changed the argument
0 of the get image information instruction, then it will fail the second
round compilation. But the argument 1 is ok to change, as we never change
the first instruction's argument, and all the subsequent instruction's
argument 1 is free to change.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

Runtime: fix the length of properties.

The last zero should also be counted.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

Runtime: implement clGetSamplerInfo.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

Runtime: fix some max values.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

Runtime: fix the incorrect device info string size.

sizeof(str) already includes the '\0', we don't need to add
1 to it.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

GBE: disable MulAdd pattern in instruction selection temporarily.

The story starts from 'FP_CONTRACT'. In c99 spec, it describes constract
expression as:
"A floating expression may be contracted, that is, evaluated as though it
were an atomic operation, thereby omitting rounding errors implied by the
source code and the expression evaluation method."

But user can use 'pragma FP_CONTRACT OFF' to disable float contraction,
in which condition, we should not do contraction like mad optimization.
In SPIR 1.2, named metadata 'opencl.enable.FP_CONTRACT' will be used to do this.
When Clang is ready, we need refine the backend logic.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: Homer Hsing <homer.xing@intel.com>

utests: add test case for structure argument

Signed-off-by: Lu Guanqun <guanqun.lu@intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

fix the error that structure would be pushed twice

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Tested-by: Lu Guanqun <guanqun.lu@intel.com>
Reviewed-by: Lu Guanqun <guanqun.lu@intel.com>

GBE: use ISA mad for mad() builtin function.

directly map mad() to ISA mad. so mad will have better performance and
less precision loss.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

utests: use mad which will get better precision.

Normal mul/add could not met the precision requirement of this case.
Previously it passed because we will do mad optimization in backend.
Use mad directly, so the test case does not depend on backend optimization.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Add a necessary include path for building with mesa.

Reported by Lv Meng.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>

fix operators for 64 bit integer

if operand is signed 64 bit integer, emit -1 for SExt casting

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

fix pointer bugs in linked list

change the header of linked list if the header was deleted

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

fix ill-coded utest_run::main

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

add same type converting

converting a data type to same type ...

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>

runtime: Fix a dangling pointer issue

ctx->events points to the head of 'event list' under the ctx.
When deleting an event from the list, we should also update
the head pointer besides updating its neighbour's next & prev,

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewedy-by: "Xing, Homer" <homer.xing@intel.com>

GBE: fixed one bug for vector relational builtin functions.

For most vector relational builtin functions, we need to
return -1 if the element result is true, return 0 if the element
result is 0. So we can simply put a - in front of the scalar
version of function for each element.

Reported by Yang Rong.

Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>

fix built-in function "normalize"

divide the parameter by its length

ver 2: scalar typed function returns NaN if parameter is NaN.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

fix built-in function "fast_normalize"

if the parameter is zero, then return zero
if the parameter is positive, then return 1.
for other cases, return -1.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: "Lu, Guanqun" <guanqun.lu@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

GBE: Give a zero-initialized register for Undef value.

For instructions that reference an undef value, we simply
allocate a register to the undef operand and set as 0.

v2:
handle float and double type. also fix some typos about double type.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

Refine the build option checking.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>

Fix a event segment fault.

If event type is CL_COMMAND_USER, event->queue is NULL, cause segment fault.
Change the order to fix it.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

GBE: enable bitselect vector builtin functions.

Now we have the scalar version of bitselect, so we
enable the vector version in the def file. Also remove
some comments.

Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>

Runtime: fixed an incorrect error checking for CL_INVALID_GLOBAL_OFFSET.

According to OpenCL spec:

CL_INVALID_GLOBAL_OFFSET if the value specified in global_work_size + the
corresponding values in global_work_offset for any dimensions is greater than the
sizeof(size t) for the device on which the kernel execution will be enqueued.

Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

GBE: fix 3-component vector's astype macros.

According to OpenCL spec,

For 3-component vector data types, the size of the data type is 4 * sizeof(component). This
means that a 3-component vector data type will be aligned to a 4 * sizeof(component)
boundary. The vload3 and vstore3 built-in functions can be used to read and write, respectively,
3-component vector data types from an array of packed scalar data type.

Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

GBE: fix a bug for the cast(FPToUI) instruction.

We need to choose unsigned dst type for this case.

v2:
fix a typo.

Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>

fix built-in function "length"

if vector is zero, then returns zero.
if vector is very large, then do a scaling first.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

delete vec-8 or vec-16 typed geometric built-in

per OpenCL spec (ver 1.1 or 1.2), geometric built-in
should not have vec-8 type or vec-16 type versions.

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

not use "mad" in vector type "dot"

the purpose is just to make code more readable, for float16 case

Signed-off-by: Homer Hsing <homer.xing@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>

Per openCL spec, set p->is_built to 1 when build fail.

Also correct the err code when build fail.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>

Re-build the program when build option changed.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>

Remove CL_FP_DENORM in clGetDeviceInfo.

IVB don't support single float denorm, so compiler option -cl-denorms-are-zero should ingore.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>

Add preprocessor #define that match the extension name string.

Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
Reviewed-by: "Xing, Homer" <homer.xing@intel.com>