[VectorDistribute] Enable intrinsic usage for unaligned attention#24510
Draft
sommerlukas wants to merge 3 commits into
Draft
[VectorDistribute] Enable intrinsic usage for unaligned attention#24510sommerlukas wants to merge 3 commits into
sommerlukas wants to merge 3 commits into
Conversation
Signed-off-by: Lukas Sommer <lukas.sommer@amd.com>
Signed-off-by: Lukas Sommer <lukas.sommer@amd.com>
Signed-off-by: Lukas Sommer <lukas.sommer@amd.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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Enable attention to use intrinsics for matrix multiplication, even if the attention shape is not evenly divided by the intrinsic size.
Previous work on to allow padding/masking in such cases allows to use intrinsics in these cases.
Additionally, also adds configuration and pipeline tests for unaligned shapes, mainly based on shapes reported in #24221 and #24319. The pipeline tests not only check cases using intrinsics, but also smaller shapes that do not use intrinsics.
Assisted-by: Codex