[XLA:GPU] Adjust GetNumWarps heuristic in Tiled Cost Model. #19540
Conversation
This file contains 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
63da492 to
3765dc1
Compare
We need to adjust the heuristic because before our emitter had an issue that prevented Triton from doing proper layout optimizations. It was fixed in 7280b9a. We needed to use higher number of warps (up to 32) before to cover the lack of layout optimization, but now it can cause performance regressions, because Triton likes to insert shmem usage and barrier syncs. PiperOrigin-RevId: 698416298
3765dc1 to
6fc61c5
Compare
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Outdated suggestions cannot be applied.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.
[XLA:GPU] Adjust GetNumWarps heuristic in Tiled Cost Model.
We need to adjust the heuristic because before our emitter had an issue that prevented Triton from doing proper layout optimizations. It was fixed in 7280b9a.
We needed to use higher number of warps (up to 32) before to cover the lack of layout optimization, but now it can cause performance regressions, because Triton likes to insert shmem usage and barrier syncs.