mirror of
https://github.com/KhronosGroup/OpenCL-CTS.git
synced 2026-03-19 06:09:01 +00:00
* allocations: Move results array from stack to heap (#1857) * allocations: Fix stack overflow * check format fixes * Fix windows stack overflow. (#1839) * thread_dimensions: Avoid combinations of very small LWS and very large GWS (#1856) Modify the existing condition to include extremely small LWS like 1x1 on large GWS values * c11_atomics: Reduce the loopcounter for sequential consistency tests (#1853) Reduce the loop from 1000000 to 500000 since the former value makes the test run too long and cause system issues on certain platforms * Limit individual allocation size using the global memory size (#1835) Signed-off-by: Ahmed Hesham <ahmed.hesham@arm.com> * geometrics: fix Wsign-compare warnings (#1855) Signed-off-by: Sven van Haastregt <sven.vanhaastregt@arm.com> * integer_ops: fix -Wformat warnings (#1860) The main sources of warnings were: * Printing of a `size_t` which requires the `%zu` specifier. * Printing of `cl_long`/`cl_ulong` which is now done using the `PRI*64` macros to ensure portability across 32 and 64-bit builds. Signed-off-by: Sven van Haastregt <sven.vanhaastregt@arm.com> * Replace OBSOLETE_FORAMT with OBSOLETE_FORMAT (#1776) * Replace OBSOLETE_FORAMT with OBSOLETE_FORMAT In imageHelpers.cpp and few other places in image tests, OBSOLETE_FORMAT is misspelled as OBSOLETE_FORAMT. Fix misspelling by replcaing it with OBSOLETE_FORMAT. Fixes #1769 * Remove code guarded by OBSOLETE_FORMAT Remove code guarded by OBSOLETE_FORMAT as suggested by review comments Fixes #1769 * Fix formating issues for OBSOLETE_FORMAT changes Fix formatting issues observed in files while removing code guarded by OBSOLETE_FORMAT Fixes #1769 * Some more formatting fixes Some more formatting fixes to get CI clean Fixes #1769 * Final Formating fixes Final formatting fixes for #1769 * Enhancement: Thread dimensions user parameters (#1384) * Fix format in the test scope * Add user params to limit testing Add parameters to reduce amount of testing. Helpful for debugging or for machines with lower performance. * Restore default value * Print info only if testing params bigger than 0. * [NFC] conversions: reenable Wunused-but-set-variable (#1845) Remove an assigned-to but unused variable. Reenable the Wunused-but-set-variable warning for the conversions suite, as it now compiles cleanly with this warning enabled. Signed-off-by: Sven van Haastregt <sven.vanhaastregt@arm.com> * Fix bug of conversion from long to double (#1847) * Fix bug of conversion from long to double It the input is long type, it should be load as long type, not ulong. * update long2float * math_brute_force: fix exp/exp2 rlx ULP calculation (#1848) Fix the ULP error calculation for the `exp` and `exp2` builtins in relaxed math mode for the full profile. Previously, the `ulps` value kept being added to while verifying the result buffer in a loop. `ulps` could even become a `NaN` when the input argument being tested was a `NaN`. Signed-off-by: Sven van Haastregt <sven.vanhaastregt@arm.com> * Enable LARGEADDRESSAWARE for 32 bit compilation (#1858) * Enable LARGEADDRESSAWARE for 32 bit compilation 32-bit executables built with MSVC linker have only 2GB virtual memory address space by default, which might not be sufficient for some tests. Enable LARGEADDRESSAWARE linker flag for 32-bit targets to allow tests to handle addresses larger than 2 gigabytes. https://learn.microsoft.com/en-us/cpp/build/reference/largeaddressaware-handle-large-addresses?view=msvc-170 Signed-off-by: Guo, Yilong <yilong.guo@intel.com> * Apply suggestion Co-authored-by: Ben Ashbaugh <ben.ashbaugh@intel.com> --------- Signed-off-by: Guo, Yilong <yilong.guo@intel.com> Co-authored-by: Ben Ashbaugh <ben.ashbaugh@intel.com> * fix return code when readwrite image is not supported (#1873) This function (do_test) starts by testing write and read individually. Both of them can have errors. When readwrite image is not supported, the function returns TEST_SKIPPED_ITSELF potentially masking errors leading to the test returning EXIT_SUCCESS even with errors along the way. * fix macos builds by avoiding double compilation of function_list.cpp for test_spir (#1866) * modernize CMakeLists for test_spir * add the operating system release to the sccache key * include the math brute force function list vs. building it twice * fix the license header on the spirv-new tests (#1865) The source files for the spirv-new tests were using the older Khronos license instead of the proper Apache license. Fixed the license in all source files. * compiler: fix grammar in error message (#1877) Signed-off-by: Sven van Haastregt <sven.vanhaastregt@arm.com> * Updated semaphore tests to use clSemaphoreReImportSyncFdKHR. (#1854) * Updated semaphore tests to use clSemaphoreReImportSyncFdKHR. Additionally updated common semaphore code to handle spec updates that restrict simultaneous importing/exporting of handles. * Fix build issues on CI * gcc build issues * Make clReImportSemaphoreSyncFdKHR a required API call if cl_khr_external_semaphore_sync_fd is present. * Implement signal and wait for all semaphore types. * subgroups: fix for testing too large WG sizes (#1620) It seemed to be a typo; the comment says that it tries to fetch local size for a subgroup count with above max WG size, but it just used the previous subgroup count. The test on purpose sets a SG count to be a larger number than the max work-items in the work group. Given the minimum SG size is 1 WI, it means that there can be a maximum of maximum work-group size of SGs (of 1 WI of size). Thus, if we request a number of SGs that exceeds the local size, the query should fail as expected. * add SPIR-V version testing (#1861) * basic SPIR-V 1.3 testing support * updated script to compile for more SPIR-V versions * switch to general SPIR-V versions test * update copyright text and fix license * improve output while test is running * check for higher SPIR-V versions first * fix formatting * fix the reported platform information for math brute force (#1884) When the math brute force test printed the platform version it always printed information for the first platform in the system, which could be different than the platform for the passed-in device. Fixed by querying the platform from the passed-in device instead. * api tests fix: Use MTdataHolder in test_get_image_info (#1871) * Minor fixes in mutable dispatch tests. (#1829) * Minor fixes in mutable dispatch tests. * Fix size of newWrapper in MutableDispatchSVMArguments. * Fix errnoneus clCommandNDRangeKernelKHR call. Signed-off-by: John Kesapides <john.kesapides@arm.com> * * Set the row_pitch for imageInfo in MutableDispatchImage1DArguments and MutableDispatchImage2DArguments. The row_pitch is used by get_image_size() to calculate the size of the host pointers by generate_random_image_data. Signed-off-by: John Kesapides <john.kesapides@arm.com> --------- Signed-off-by: John Kesapides <john.kesapides@arm.com> * add test for cl_khr_spirv_linkonce_odr (#1226) * initial version of the test with placeholders for linkonce_odr linkage * add OpExtension SPV_KHR_linkonce_odr extension * add check for extension * switch to actual LinkOnceODR linkage * fix formatting * add a test case to ensure a function with linkonce_odr is exported * add back the extension check * fix formatting * undo compiler optimization and actually add the call to function a * [NFC] subgroups: remove unnecessary extern keywords (#1892) In C and C++ all functions have external linkage by default. Also remove the unused `gMTdata` and `test_pipe_functions` declarations. Fixes https://github.com/KhronosGroup/OpenCL-CTS/issues/1137 Signed-off-by: Sven van Haastregt <sven.vanhaastregt@arm.com> * Added cl_khr_fp16 extension support for test_decorate from spirv_new (#1770) * Added cl_khr_fp16 extension support for test_decorate from spirv_new, work in progres * Complemented test_decorate saturation test to support cl_khr_fp16 extension (issue #142) * Fixed clang format * scope of modifications: -changed naming convention of saturation .spvasm files related to test_decorate of spirv_new -restored float to char/uchar saturation tests -few minor corrections * fix ranges for half testing * fix formating * one more formatting fix * remove unused function * use isnan instead of std::isnan isnan is currently implemented as a macro, not as a function, so we can't use std::isnan. * fix Clang warning about inexact conversion --------- Co-authored-by: Ben Ashbaugh <ben.ashbaugh@intel.com> * add support for custom devices (#1891) enable the CTS to run on custom devices --------- Signed-off-by: Ahmed Hesham <ahmed.hesham@arm.com> Signed-off-by: Sven van Haastregt <sven.vanhaastregt@arm.com> Signed-off-by: Guo, Yilong <yilong.guo@intel.com> Signed-off-by: John Kesapides <john.kesapides@arm.com> Co-authored-by: Sreelakshmi Haridas Maruthur <sharidas@quicinc.com> Co-authored-by: Haonan Yang <haonan.yang@intel.com> Co-authored-by: Ahmed Hesham <117350656+ahesham-arm@users.noreply.github.com> Co-authored-by: Sven van Haastregt <sven.vanhaastregt@arm.com> Co-authored-by: niranjanjoshi121 <43807392+niranjanjoshi121@users.noreply.github.com> Co-authored-by: Grzegorz Wawiorko <grzegorz.wawiorko@intel.com> Co-authored-by: Wenwan Xing <wenwan.xing@intel.com> Co-authored-by: Yilong Guo <yilong.guo@intel.com> Co-authored-by: Romaric Jodin <89833130+rjodinchr@users.noreply.github.com> Co-authored-by: joshqti <127994991+joshqti@users.noreply.github.com> Co-authored-by: Pekka Jääskeläinen <pekka.jaaskelainen@tuni.fi> Co-authored-by: imilenkovic00 <155085410+imilenkovic00@users.noreply.github.com> Co-authored-by: John Kesapides <46718829+JohnKesapidesARM@users.noreply.github.com> Co-authored-by: Marcin Hajder <marcin.hajder@gmail.com> Co-authored-by: Aharon Abramson <aharon.abramson@mobileye.com>
223 lines
9.7 KiB
C++
223 lines
9.7 KiB
C++
//
|
|
// Copyright (c) 2017 The Khronos Group Inc.
|
|
//
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
// you may not use this file except in compliance with the License.
|
|
// You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
// See the License for the specific language governing permissions and
|
|
// limitations under the License.
|
|
//
|
|
#include "testBase.h"
|
|
#include "harness/typeWrappers.h"
|
|
#include "harness/conversions.h"
|
|
|
|
const char *subgroup_dispatch_kernel[] = {
|
|
"#pragma OPENCL EXTENSION cl_khr_subgroups : enable\n"
|
|
"__kernel void subgroup_dispatch_kernel(__global int *output)\n"
|
|
"{\n"
|
|
" size_t size = get_num_sub_groups ();\n"
|
|
"\n"
|
|
" output[0] = size;\n"
|
|
"\n"
|
|
"}\n" };
|
|
|
|
size_t flatten_ndrange(size_t* ndrange, size_t dim)
|
|
{
|
|
switch(dim)
|
|
{
|
|
case 1:
|
|
return *ndrange;
|
|
case 2:
|
|
return ndrange[0] * ndrange[1];
|
|
case 3:
|
|
return ndrange[0] * ndrange[1] * ndrange[2];
|
|
default:
|
|
log_error("ERROR: bad ndrange value");
|
|
return 0;
|
|
}
|
|
}
|
|
|
|
cl_int get_sub_group_num(cl_command_queue queue, cl_kernel kernel, clMemWrapper& out, size_t& size, size_t local_size, size_t dim)
|
|
{
|
|
size_t ndrange[3] = {local_size, 1, 1};
|
|
cl_int error = CL_SUCCESS;
|
|
size = 0;
|
|
error = clSetKernelArg(kernel, 0, sizeof(out), &out);
|
|
error += clEnqueueNDRangeKernel(queue, kernel, dim, NULL, ndrange, ndrange, 0, NULL, NULL);
|
|
error += clEnqueueReadBuffer(queue, out, CL_TRUE, 0, 4, &size, 0, NULL, NULL);
|
|
return error;
|
|
}
|
|
|
|
int test_sub_group_dispatch(cl_device_id deviceID, cl_context context, cl_command_queue queue, int num_elements)
|
|
{
|
|
int error;
|
|
size_t realSize;
|
|
size_t kernel_max_subgroup_size, kernel_subgroup_count;
|
|
size_t max_local;
|
|
|
|
cl_platform_id platform;
|
|
clProgramWrapper program;
|
|
clKernelWrapper kernel;
|
|
clMemWrapper out;
|
|
|
|
size_t ret_ndrange1d;
|
|
size_t ret_ndrange2d[2];
|
|
size_t ret_ndrange3d[3];
|
|
|
|
size_t ret_ndrange2d_flattened;
|
|
size_t ret_ndrange3d_flattened;
|
|
|
|
if (get_device_cl_version(deviceID) >= Version(3, 0))
|
|
{
|
|
int error;
|
|
cl_uint max_num_sub_groups;
|
|
|
|
error = clGetDeviceInfo(deviceID, CL_DEVICE_MAX_NUM_SUB_GROUPS,
|
|
sizeof(max_num_sub_groups), &max_num_sub_groups,
|
|
NULL);
|
|
if (error != CL_SUCCESS)
|
|
{
|
|
print_error(error, "Unable to get max num subgroups");
|
|
return error;
|
|
}
|
|
|
|
if (max_num_sub_groups == 0)
|
|
{
|
|
return TEST_SKIPPED_ITSELF;
|
|
}
|
|
}
|
|
|
|
error = create_single_kernel_helper(context, &program, &kernel, 1,
|
|
subgroup_dispatch_kernel,
|
|
"subgroup_dispatch_kernel");
|
|
if (error != 0)
|
|
return error;
|
|
|
|
out = clCreateBuffer(context, CL_MEM_READ_WRITE | CL_MEM_ALLOC_HOST_PTR, sizeof(size_t), NULL, &error);
|
|
test_error(error, "clCreateBuffer failed");
|
|
|
|
error = clGetDeviceInfo(deviceID, CL_DEVICE_MAX_WORK_GROUP_SIZE, sizeof(size_t), &max_local, NULL);
|
|
test_error(error, "clGetDeviceInfo failed");
|
|
|
|
|
|
error = clGetDeviceInfo(deviceID, CL_DEVICE_PLATFORM, sizeof(platform), (void *)&platform, NULL);
|
|
test_error(error, "clDeviceInfo failed for CL_DEVICE_PLATFORM");
|
|
|
|
// Get the max subgroup size
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE,
|
|
sizeof(max_local), &max_local, sizeof(kernel_max_subgroup_size), (void *)&kernel_max_subgroup_size, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE");
|
|
log_info("The CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE for the kernel is %d.\n", (int)kernel_max_subgroup_size);
|
|
|
|
if (realSize != sizeof(kernel_max_subgroup_size)) {
|
|
log_error( "ERROR: Returned size of max sub group size not valid! (Expected %d, got %d)\n", (int)sizeof(kernel_max_subgroup_size), (int)realSize );
|
|
return -1;
|
|
}
|
|
|
|
// Get the number of subgroup for max local size
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE,
|
|
sizeof(max_local), &max_local, sizeof(kernel_subgroup_count), (void *)&kernel_subgroup_count, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE");
|
|
log_info("The CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE for the kernel is %d.\n", (int)kernel_subgroup_count);
|
|
|
|
if (realSize != sizeof(kernel_subgroup_count)) {
|
|
log_error("ERROR: Returned size of sub group count not valid! "
|
|
"(Expected %d, got %d)\n",
|
|
(int)sizeof(kernel_subgroup_count), (int)realSize);
|
|
return -1;
|
|
}
|
|
|
|
// test CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT
|
|
for (size_t i = kernel_subgroup_count; i > 0; --i)
|
|
{
|
|
// test all 3 different dimention of requested local size
|
|
size_t kernel_ret_size = 0;
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT, sizeof(i), &i, sizeof(ret_ndrange1d), &ret_ndrange1d, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT");
|
|
if (realSize != sizeof(ret_ndrange1d)) {
|
|
log_error( "ERROR: Returned size of sub group count not valid! (Expected %d, got %d)\n", (int)sizeof(kernel_subgroup_count), (int)realSize );
|
|
return -1;
|
|
}
|
|
|
|
error = get_sub_group_num(queue, kernel, out, kernel_ret_size, ret_ndrange1d, 1);
|
|
test_error(error, "Failed to query number of subgroups from kernel");
|
|
if (i != kernel_ret_size)
|
|
{
|
|
log_error( "ERROR: Mismatch between requested number of subgroups and what get_num_sub_groups() in kernel returned! (Expected %d, got %d)\n", (int)i, (int)kernel_ret_size );
|
|
return -1;
|
|
}
|
|
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT, sizeof(i), &i, sizeof(ret_ndrange2d), ret_ndrange2d, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT");
|
|
if (realSize != sizeof(ret_ndrange2d)) {
|
|
log_error( "ERROR: Returned size of sub group count not valid! (Expected %d, got %d)\n", (int)sizeof(kernel_subgroup_count), (int)realSize );
|
|
return -1;
|
|
}
|
|
|
|
ret_ndrange2d_flattened = flatten_ndrange(ret_ndrange2d, 2);
|
|
|
|
error = get_sub_group_num(queue, kernel, out, kernel_ret_size, ret_ndrange2d_flattened, 2);
|
|
test_error(error, "Failed to query number of subgroups from kernel");
|
|
if (i != kernel_ret_size)
|
|
{
|
|
log_error( "ERROR: Mismatch between requested number of subgroups and what get_num_sub_groups() in kernel returned! (Expected %d, got %d)\n", (int)i, (int)kernel_ret_size );
|
|
return -1;
|
|
}
|
|
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT, sizeof(i), &i, sizeof(ret_ndrange3d), ret_ndrange3d, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT");
|
|
if (realSize != sizeof(ret_ndrange3d)) {
|
|
log_error( "ERROR: Returned size of sub group count not valid! (Expected %d, got %d)\n", (int)sizeof(kernel_subgroup_count), (int)realSize );
|
|
return -1;
|
|
}
|
|
|
|
ret_ndrange3d_flattened = flatten_ndrange(ret_ndrange3d, 3);
|
|
|
|
error = get_sub_group_num(queue, kernel, out, kernel_ret_size, ret_ndrange3d_flattened, 3);
|
|
test_error(error, "Failed to query number of subgroups from kernel");
|
|
if (i != kernel_ret_size)
|
|
{
|
|
log_error( "ERROR: Mismatch between requested number of subgroups and what get_num_sub_groups() in kernel returned! (Expected %d, got %d)\n", (int)i, (int)kernel_ret_size );
|
|
return -1;
|
|
}
|
|
}
|
|
|
|
// test when input subgroup count exceeds max wg size:
|
|
// there can be at most the local size of (1 WI) subgroups
|
|
size_t large_sg_size = max_local + 1;
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT, sizeof(size_t), &large_sg_size, sizeof(ret_ndrange1d), &ret_ndrange1d, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT");
|
|
if (ret_ndrange1d != 0)
|
|
{
|
|
log_error( "ERROR: Incorrect value returned for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT! (Expected %d, got %d)\n", 0, (int)ret_ndrange1d );
|
|
return -1;
|
|
}
|
|
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT, sizeof(size_t), &large_sg_size, sizeof(ret_ndrange2d), ret_ndrange2d, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT");
|
|
if (ret_ndrange2d[0] != 0 ||
|
|
ret_ndrange2d[1] != 0)
|
|
{
|
|
log_error( "ERROR: Incorrect value returned for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT!" );
|
|
return -1;
|
|
}
|
|
|
|
error = clGetKernelSubGroupInfo(kernel, deviceID, CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT, sizeof(size_t), &large_sg_size, sizeof(ret_ndrange3d), ret_ndrange3d, &realSize);
|
|
test_error(error, "clGetKernelSubGroupInfo failed for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT");
|
|
if (ret_ndrange3d[0] != 0 ||
|
|
ret_ndrange3d[1] != 0 ||
|
|
ret_ndrange3d[2] != 0)
|
|
{
|
|
log_error( "ERROR: Incorrect value returned for CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT!" );
|
|
return -1;
|
|
}
|
|
|
|
return 0;
|
|
}
|