Fix Half2 UVM performance regression with vectorized store#5499
Open
q10 wants to merge 1 commit intopytorch:mainfrom
Open
Fix Half2 UVM performance regression with vectorized store#5499q10 wants to merge 1 commit intopytorch:mainfrom
q10 wants to merge 1 commit intopytorch:mainfrom
Conversation
Summary: Apply vectorized store optimization pattern for fbgemm_gpu::rocm::Half2 class. This ensures Half2 store operations use efficient 32-bit memory operations instead of scalar element-by-element access. With UVM (managed memory), each separate store can trigger a page fault, causing significant slowdown. Using vectorized operations reduces this overhead. Reviewed By: henrylhtsang Differential Revision: D96381300
Contributor
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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:
Apply vectorized store optimization pattern for fbgemm_gpu::rocm::Half2 class. This ensures Half2 store operations use efficient 32-bit memory operations instead of scalar element-by-element access.
With UVM (managed memory), each separate store can trigger a page fault, causing significant slowdown. Using vectorized operations reduces this overhead.
Reviewed By: henrylhtsang
Differential Revision: D96381300