Fix AllReduce3 auto-tune launch parameters#827
Open
sugarlxn wants to merge 1 commit into
Open
Conversation
Author
|
Hi maintainers, could someone please review this PR and approve the pending workflows when convenient? This PR fixes the AllReduce3 auto-tune path by replacing the hardcoded kernel launch parameters with the configured Thanks! |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
This PR fixes the launch parameters used by the AllReduce3 auto-tune path.
Previously, the kernel launch used hardcoded values:
This prevented the auto-tuned
nblocksandblock_sizevalues from actually being applied during execution. The launch now uses the configured parameters:Changes
self.nblocksandself.block_size.Motivation
The auto-tune logic should be able to control the kernel launch configuration. Using fixed values makes tuning ineffective and may lead to suboptimal performance across different message sizes, GPU counts, or hardware environments.
Testing
nblocksandblock_sizevalues.