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

BFloat16 tests XPASSing in nightly on OCL CPU #14562

Closed
sarnex opened this issue Jul 12, 2024 · 3 comments · Fixed by #14571
Closed

BFloat16 tests XPASSing in nightly on OCL CPU #14562

sarnex opened this issue Jul 12, 2024 · 3 comments · Fixed by #14571
Assignees
Labels
bug Something isn't working confirmed OCL CPU Experimental RT Issues in Experimental Intel(R) CPU Runtime for OpenCL(TM) Applications with SYCL support

Comments

@sarnex
Copy link
Contributor

sarnex commented Jul 12, 2024

Describe the bug

https://github.com/intel/llvm/actions/runs/9901741273/job/27354870210

 Unexpectedly Passed Tests (3):
  SYCL :: BFloat16/bfloat16_builtins.cpp
  SYCL :: BFloat16/bfloat16_builtins_cuda_generic.cpp
  SYCL :: BFloat16/bfloat16_type.cpp

example:

********************
XPASS: SYCL :: BFloat16/bfloat16_type.cpp (136 of 2099)
******************** TEST 'SYCL :: BFloat16/bfloat16_type.cpp' FAILED ********************
Exit Code: 0

Command Output (stdout):
--
# RUN: at line 1 has no command after substitutions
# RUN: at line 2 has no command after substitutions
# RUN: at line 3
/__w/llvm/llvm/toolchain/bin//clang++   -fsycl -fsycl-targets=spir64  /__w/llvm/llvm/llvm/sycl/test-e2e/BFloat16/bfloat16_type.cpp -o /__w/llvm/llvm/build-e2e/BFloat16/Output/bfloat16_type.cpp.tmp.out
# executed command: /__w/llvm/llvm/toolchain/bin//clang++ -fsycl -fsycl-targets=spir64 /__w/llvm/llvm/llvm/sycl/test-e2e/BFloat16/bfloat16_type.cpp -o /__w/llvm/llvm/build-e2e/BFloat16/Output/bfloat16_type.cpp.tmp.out
# note: command had no output on stdout or stderr
# RUN: at line 4
env ONEAPI_DEVICE_SELECTOR=opencl:cpu  /__w/llvm/llvm/build-e2e/BFloat16/Output/bfloat16_type.cpp.tmp.out
# executed command: env ONEAPI_DEVICE_SELECTOR=opencl:cpu /__w/llvm/llvm/build-e2e/BFloat16/Output/bfloat16_type.cpp.tmp.out
# .---command stdout------------
# | PASS verify_conv_implicit
# | PASS verify_conv_explicit
# | PASS verify_add
# | PASS verify_sub
# | PASS verify_mul
# | PASS verify_div
# | PASS verify_logic
# | PASS verify_add
# | PASS verify_sub
# | PASS verify_minus
# | PASS verify_mul
# | PASS verify_div
# | PASS verify_logic
# `-----------------------------

To reproduce

No response

Environment

No response

Additional context

No response

@sarnex sarnex added the bug Something isn't working label Jul 12, 2024
@sarnex
Copy link
Contributor Author

sarnex commented Jul 12, 2024

@frasercrmck The XFAIL was added in 0f0b699, can you please take a look at why the nightly passes but it failed in whatever case cause you to add XFAIL? Thanks

@sarnex sarnex changed the title BFloat16 tests XPASSing in nightly BFloat16 tests XPASSing in nightly on OCL CPU Jul 12, 2024
@sarnex sarnex added the OCL CPU Experimental RT Issues in Experimental Intel(R) CPU Runtime for OpenCL(TM) Applications with SYCL support label Jul 12, 2024
@frasercrmck
Copy link
Contributor

Ah sorry, it was failing for me locally in a multi-target build with CUDA but perhaps I applied the wrong kind fix. I was applying the suggestion made here perhaps a little too zealously, and took the FIXME comment in the test too literally.

@sarnex
Copy link
Contributor Author

sarnex commented Jul 15, 2024

@frasercrmck Thanks! FYI there are 2 other tests failing, here is the full failing list:

 Unexpectedly Passed Tests (3):
  SYCL :: BFloat16/bfloat16_builtins.cpp
  SYCL :: BFloat16/bfloat16_builtins_cuda_generic.cpp
  SYCL :: BFloat16/bfloat16_type.cpp

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 OCL CPU Experimental RT Issues in Experimental Intel(R) CPU Runtime for OpenCL(TM) Applications with SYCL support
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants