Repository navigation
Conversation
Contributor
|
No actionable comments were generated in the recent review.
 |
| Layer / File(s) | Summary |
|---|---|
Hint conversion and single-attempt behavior libcudacxx/include/cuda/__barrier/shared_memory_barrier.h |
Adds hint range checking and conversion. Nonpositive hints perform one test-wait check; positive hints use the SM 90 hinted wait path or report unsupported storage on unsupported targets. |
Hinted wait overloads and tests libcudacxx/include/cuda/__barrier/shared_memory_barrier.h, libcudacxx/test/libcudacxx/cuda/barrier/shared_memory_barrier.cu |
Adds hinted overloads for arrival tokens, primary phases, and conditional phases. Tests check status and completion results for several hint values before and after barrier completion. |
Possibly related PRs
- NVIDIA/cccl#12131: Changes
operation_statusdecoding used by the new status-returning hinted waits.
Suggested reviewers: fbusato, wmaxey, jacobfaib
Priority:
Change: Feature
Merge Risk: 
8682d
-
The hinted waits and their tests show no identified issue that should block merging after normal checks.
Comment @coderabbitai help to get the list of available commands.
Contributor
🥳 CI Workflow Results🟩 Finished in 2h 59m: Pass: 100%/228 | Total: 4d 04h | Max: 1h 32m | Hits: 92%/687189See results here. |
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
We currently have
try_waitthat will do a single wait for an unspecified amount of time andtry_wait_forthat takes a time limit and the API will keep callingtry_waitin a loop until that time limit passes. We also havewaitthat will wait however long it takes in a loop.try_waitPTX instruction actually does take a time limit hint, its not promised by the implementation, that's whytry_wait_forhas a loop. But we should expose another wait variant that can supply that hint without the extra cost of a loop.This PR adds a new overload of
try_wait. It is still documented to do a one-shot wait for unspecified amount of time like the currenttry_waitvariant, but it takes an extra argument with the wait time hint.