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

[Driver] Save PTX files for SYCL kernels in the user input directory. #12422

Merged
merged 7 commits into from
Feb 21, 2024

Conversation

srividya-sundaram
Copy link
Contributor

@srividya-sundaram srividya-sundaram commented Jan 17, 2024

Save PTX files generated (for SYCL kernels) during PTX target processing for CUDA backend using the -fsycl-dump-device-code option.

Example usage:
clang++ -fsycl -fsycl-targets=nvptx64-nvidia-cuda -fsycl-dmp-device-code=/path/to/ptx syclfile.cpp

The PTX files (.s files for each kernel) will be saved under /path/to/ptx

@jinz2014
Copy link
Contributor

Thank you for the pull request.

@srividya-sundaram
Copy link
Contributor Author

@intel/llvm-gatekeepers Hi, this is ready for merge. Thanks.

@sommerlukas sommerlukas merged commit 16e06ff into intel:sycl Feb 21, 2024
11 checks passed
@sommerlukas
Copy link
Contributor

Unrelated failures in post-commit:

Failed Tests (2):
  SYCL :: ESIMD/unified_memory_api/scatter_lacc_dg2_pvc.cpp
  SYCL :: Reduction/reduction_range_3d_rw_reducer_skip.cpp

Windows LevelZero failed with fatal: Cannot list device aspects for level_zero:gpu.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants