Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

kernel_bundle_ignore_sycl_external.cpp failing on CUDA #16417

Open
lbushi25 opened this issue Dec 18, 2024 · 0 comments
Open

kernel_bundle_ignore_sycl_external.cpp failing on CUDA #16417

lbushi25 opened this issue Dec 18, 2024 · 0 comments
Labels
bug Something isn't working confirmed cuda CUDA back-end

Comments

@lbushi25
Copy link
Contributor

Describe the bug

The test SYCL :: Regression/kernel_bundle_ignore_sycl_external.cpp is failing on CUDA. Upon resolution, please remove XFAIL from test source.

To reproduce

FAIL: SYCL :: Regression/kernel_bundle_ignore_sycl_external.cpp (1843 of 2242)
******************** TEST 'SYCL :: Regression/kernel_bundle_ignore_sycl_external.cpp' FAILED ********************
Exit Code: -6

Command Output (stdout):
--
# RUN: at line 1
/__w/llvm/llvm/toolchain/bin//clang++  -Werror  -fsycl -fsycl-targets=nvptx64-nvidia-cuda  /__w/llvm/llvm/llvm/sycl/test-e2e/Regression/kernel_bundle_ignore_sycl_external.cpp -o /__w/llvm/llvm/build-e2e/Regression/Output/kernel_bundle_ignore_sycl_external.cpp.tmp.out
# executed command: /__w/llvm/llvm/toolchain/bin//clang++ -Werror -fsycl -fsycl-targets=nvptx64-nvidia-cuda /__w/llvm/llvm/llvm/sycl/test-e2e/Regression/kernel_bundle_ignore_sycl_external.cpp -o /__w/llvm/llvm/build-e2e/Regression/Output/kernel_bundle_ignore_sycl_external.cpp.tmp.out
# note: command had no output on stdout or stderr
# RUN: at line 2
env UR_CUDA_ENABLE_IMAGE_SUPPORT=1 ONEAPI_DEVICE_SELECTOR=cuda:gpu  /__w/llvm/llvm/build-e2e/Regression/Output/kernel_bundle_ignore_sycl_external.cpp.tmp.out
# executed command: env UR_CUDA_ENABLE_IMAGE_SUPPORT=1 ONEAPI_DEVICE_SELECTOR=cuda:gpu /__w/llvm/llvm/build-e2e/Regression/Output/kernel_bundle_ignore_sycl_external.cpp.tmp.out
# .---command stderr------------
# | kernel_bundle_ignore_sycl_external.cpp.tmp.out: /__w/llvm/llvm/llvm/sycl/test-e2e/Regression/kernel_bundle_ignore_sycl_external.cpp:22: int main(): Assertion `sycl::get_kernel_ids().size() == 1' failed.
# `-----------------------------
# error: command failed with exit status: -6


Environment

  • OS: Any
  • Target device and vendor: Nvidia GPU
  • DPC++ version: da317d0
  • Dependencies version: 2024-12-17T21:58:43.7814040Z �[36;1msycl-ls --verbose�[0m
2024-12-17T21:58:43.7814444Z shell: sh -e {0}
2024-12-17T21:58:43.7814623Z env:
2024-12-17T21:58:43.7815136Z   PATH: /__w/llvm/llvm/toolchain/bin/:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
2024-12-17T21:58:43.7815893Z   LD_LIBRARY_PATH: /__w/llvm/llvm/toolchain/lib/:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
2024-12-17T21:58:43.7816306Z ##[endgroup]
2024-12-17T21:58:44.7889537Z [cuda:gpu][cuda:0] NVIDIA CUDA BACKEND, NVIDIA GeForce RTX 3090 8.6 [CUDA 12.2]
2024-12-17T21:58:44.7890811Z [native_cpu:cpu][native_cpu:0] SYCL_NATIVE_CPU, SYCL Native CPU 0.1 [0.0.0]
2024-12-17T21:58:44.7891600Z 
2024-12-17T21:58:44.7891859Z Platforms: 2
2024-12-17T21:58:44.7892384Z Platform [#1]:
2024-12-17T21:58:44.7893015Z     Version  : CUDA 12.2
2024-12-17T21:58:44.7893647Z     Name     : NVIDIA CUDA BACKEND
2024-12-17T21:58:44.7894417Z     Vendor   : NVIDIA Corporation
2024-12-17T21:58:44.7894873Z     Devices  : 1
2024-12-17T21:58:44.7895202Z         Device [#0]:
2024-12-17T21:58:44.7895571Z         Type              : gpu
2024-12-17T21:58:44.7895994Z         Version           : 8.6
2024-12-17T21:58:44.7896460Z         Name              : NVIDIA GeForce RTX 3090
2024-12-17T21:58:44.7897034Z         Vendor            : NVIDIA Corporation
2024-12-17T21:58:44.7897582Z         Driver            : CUDA 12.2
2024-12-17T21:58:44.7898120Z         UUID              : 151202218110820252671222472153911713872
2024-12-17T21:58:44.7898682Z         DeviceID          : 0
2024-12-17T21:58:44.7899113Z         Num SubDevices    : 0
2024-12-17T21:58:44.7899560Z         Num SubSubDevices : 0
2024-12-17T21:58:44.7901299Z Images are not fully supported by the CUDA BE, their support is disabled by default. Their partial support can be activated by setting UR_CUDA_ENABLE_IMAGE_SUPPORT environment variable at runtime.
2024-12-17T21:58:44.7914284Z         Aspects           : gpu fp16 fp64 online_compiler online_linker queue_profiling usm_device_allocations usm_host_allocations usm_shared_allocations ext_intel_pci_address usm_atomic_shared_allocations atomic64 ext_intel_device_info_uuid ext_oneapi_native_assert ext_oneapi_cuda_async_barrier ext_intel_free_memory ext_intel_device_id ext_intel_memory_clock_rate ext_intel_memory_bus_width ext_oneapi_bindless_images ext_oneapi_bindless_images_shared_usm ext_oneapi_bindless_images_1d_usm ext_oneapi_bindless_images_2d_usm ext_oneapi_external_memory_import ext_oneapi_external_semaphore_import ext_oneapi_mipmap ext_oneapi_mipmap_anisotropy ext_oneapi_mipmap_level_reference ext_oneapi_ballot_group ext_oneapi_fixed_size_group ext_oneapi_opportunistic_group ext_oneapi_graph ext_oneapi_limited_graph ext_oneapi_cubemap ext_oneapi_cubemap_seamless_filtering ext_oneapi_bindless_sampled_image_fetch_1d_usm ext_oneapi_bindless_sampled_image_fetch_2d_usm ext_oneapi_bindless_sampled_image_fetch_2d ext_oneapi_bindless_sampled_image_fetch_3d ext_oneapi_queue_profiling_tag ext_oneapi_virtual_mem ext_oneapi_image_array ext_oneapi_unique_addressing_per_dim ext_oneapi_bindless_images_sample_1d_usm ext_oneapi_bindless_images_sample_2d_usm
2024-12-17T21:58:44.7931906Z         info::device::sub_group_sizes: 32
2024-12-17T21:58:44.7932863Z         Architecture: nvidia_gpu_sm_86
2024-12-17T21:58:44.7933758Z Platform [#2]:
2024-12-17T21:58:44.7934438Z     Version  : 0.1
2024-12-17T21:58:44.7935152Z     Name     : SYCL_NATIVE_CPU
2024-12-17T21:58:44.7935953Z     Vendor   : tbd
2024-12-17T21:58:44.7936651Z     Devices  : 1
2024-12-17T21:58:44.7937353Z         Device [#0]:
2024-12-17T21:58:44.7938085Z         Type              : cpu
2024-12-17T21:58:44.7938885Z         Version           : 0.1
2024-12-17T21:58:44.7939713Z         Name              : SYCL Native CPU
2024-12-17T21:58:44.7940641Z         Vendor            : Intel(R) Corporation
2024-12-17T21:58:44.7941541Z         Driver            : 0.0.0
2024-12-17T21:58:44.7942355Z         DeviceID          : UNKNOWN
2024-12-17T21:58:44.7943485Z         Num SubDevices    : 0
2024-12-17T21:58:44.7944275Z         Num SubSubDevices : 0
2024-12-17T21:58:44.7947019Z         Aspects           : cpu fp16 fp64 online_compiler online_linker queue_profiling usm_device_allocations usm_host_allocations usm_shared_allocations usm_system_allocations usm_atomic_host_allocations usm_atomic_shared_allocations atomic64
2024-12-17T21:58:44.7949755Z         info::device::sub_group_sizes: 1
2024-12-17T21:58:44.7950612Z         Architecture: unknown
2024-12-17T21:58:44.7951795Z default_selector()      : gpu, NVIDIA CUDA BACKEND, NVIDIA GeForce RTX 3090 8.6 [CUDA 12.2]
2024-12-17T21:58:44.7953336Z accelerator_selector()  : No device of requested type available.
2024-12-17T21:58:44.7954685Z cpu_selector()          : cpu, SYCL_NATIVE_CPU, SYCL Native CPU 0.1 [0.0.0]
2024-12-17T21:58:44.7956134Z gpu_selector()          : gpu, NVIDIA CUDA BACKEND, NVIDIA GeForce RTX 3090 8.6 [CUDA 12.2]
2024-12-17T21:58:44.7957796Z custom_selector(gpu)    : gpu, NVIDIA CUDA BACKEND, NVIDIA GeForce RTX 3090 8.6 [CUDA 12.2]
2024-12-17T21:58:44.7959264Z custom_selector(cpu)    : cpu, SYCL_NATIVE_CPU, SYCL Native CPU 0.1 [0.0.0]
2024-12-17T21:58:44.7960574Z custom_selector(acc)    : No device of requested type available.


Additional context

No response

@lbushi25 lbushi25 added bug Something isn't working cuda CUDA back-end labels Dec 18, 2024
@lbushi25 lbushi25 changed the title kernel_bundle.cpp failing on CUDA kernel_bundle_ignore_sycl_external.cpp failing on CUDA Dec 18, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working confirmed cuda CUDA back-end
Projects
None yet
Development

No branches or pull requests

1 participant