[SYCL][NativeCPU] Limit generic ABI to builtins. #19564
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
In intel#17408, NativeCPU became a target in order to be able to pick the ABI for its own libclc functions consistently, without having targets affect this. This was, and is, required to be able to use libclc independent of target and target options. However, it breaks some calls into libc. Therefore, this PR allows the calling convention to be explicitly specified, ensures it is specified for any libclc functions, and ensures it is not specified for any libc functions. Fixes the SYCL-E2E acos, cmath, and exp-std-complex tests.
|
This needs compiler tests, |
|
Just a heads up that after today I will be on holiday. I am not expecting reviews to be done today, that is fine, but it will be a little while before I'll be able to respond to any issues or concerns. |
aelovikov-intel
approved these changes
Jul 31, 2025
Comment on lines
+178
to
+182
| // Note: These signatures for parallel_for_work_item are intentionally | ||
| // non-conforming. The spec says this should take const WorkItemFunctionT &, | ||
| // but we take it by value, and rely on passing by value being done as passing | ||
| // a copy by reference (ptr byval) to ensure that the special handling in | ||
| // SYCLLowerWGScopePass to mutate the passed functor object works. |
There was a problem hiding this comment.
Hierarchical parallelism is deprecated in SYCL 2020, so nobody cares, I guess.
npmiller
approved these changes
Aug 4, 2025
MrSidims
approved these changes
Aug 14, 2025
maksimsab
reviewed
Aug 20, 2025
coldav
approved these changes
Aug 25, 2025
|
It seems weird, but this PR hung twice on Windows BMG where other PRs passed, I don't expect it would effect GPU since this is NativeCPU. Are there any changes here that could cause this? |
No, nothing; I would be surprised if it's caused by anything in this PR. I'll retry a few more times. |
|
@intel/llvm-gatekeepers This is ready to be merged, thanks. |
|
We suspect this or something related is responsible for the CI struggles. Looks like and then Github Actions struggle processing all that output. @hvdijk please make take a look ASAP. |
|
Can you point me to where you are seeing these warnings, please? This PR adds both the definition of the |
|
We're seeing that in pre-commit build jobs, e.g. https://github.com/intel/llvm/actions/runs/17445027793/job/49537268401 And a bit more copy-paste context: |
|
Ah, thanks. The warning follows "warning: 'libclc_call' calling convention is not supported for this target" in files such as The libclc CMake is not setting |
|
Untested fix: #19971. I'll look at the CI output when it's ready to see if this works as intended. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
9 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In #17408, NativeCPU became a target in order to be able to pick the ABI for its own libclc functions consistently, without having targets affect this. This was, and is, required to be able to use libclc independent of target and target options. However, it breaks some calls into libc. Therefore, this PR allows the calling convention to be explicitly specified, ensures it is specified for any libclc functions, and ensures it is not specified for any libc functions.
Fixes the SYCL-E2E acos, cmath, and exp-std-complex tests.