[libcudacxx] Work around NVCC issue with __restrict__ in launch#8289
Open
pciolkosz wants to merge 2 commits intoNVIDIA:mainfrom
Open
[libcudacxx] Work around NVCC issue with __restrict__ in launch#8289pciolkosz wants to merge 2 commits intoNVIDIA:mainfrom
pciolkosz wants to merge 2 commits intoNVIDIA:mainfrom
Conversation
Update cudax launch call sites to pass reinterpret_cast<const void*> to __get_cufunction_of for raw function pointers, matching the new signature. Apply clang-format fixes.
This comment has been minimized.
This comment has been minimized.
Contributor
Contributor
Author
|
/ok to test 9a80f37 |
Contributor
🥳 CI Workflow Results🟩 Finished in 1h 20m: Pass: 100%/149 | Total: 1d 13h | Max: 56m 25s | Hits: 98%/307392See results here. |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
__get_kernel_launcherreturned a function pointer whose type could be corrupted by__restrict__qualifiers leaking through nvcc's template machinery. Changed it to returnconst void*instead, avoiding the type mismatch.__get_cufunction_ofto takeconst void*directly, removing the unnecessary template parameter.__restrict__-qualified pointer parameter.Test plan
"Launch functor with __restrict__ pointer arg"test validates the fix