MLAS/POWER10: Optimize Sgemm PackA kernel using VSX intrinsics and assembly.#27575
Open
BODAPATIMAHESH wants to merge 1 commit intomicrosoft:mainfrom
Open
MLAS/POWER10: Optimize Sgemm PackA kernel using VSX intrinsics and assembly.#27575BODAPATIMAHESH wants to merge 1 commit intomicrosoft:mainfrom
BODAPATIMAHESH wants to merge 1 commit intomicrosoft:mainfrom
Conversation
…sembly Introduce an optimized POWER10 PackA implementation leveraging VSX builtins and assembly to pre-pack 8 rows of matrix A, packing 64 bytes per row per iteration. Performance improvements observed in prompt processing: - 14% speedup (batch size 1) - 6% speedup (batch size 4) - 4% speedup (batch size 8) Tested with granite-3.1-8b Signed-off-by: Mahesh Bodapati <bmahi496@linux.ibm.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Introduce an optimized POWER10 PackA implementation leveraging VSX builtins and assembly to pre-pack 8 rows of matrix A, packing 64 bytes per row per iteration.
Motivation and Context
Performance improvements observed in prompt processing:
Tested with granite-3.1-8b