Enable VECT_MUL for blackwell attention #536
Open
Conversation
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
Enabling `VECT_MUL` previously caused a regression, but this seems to come from the fma vectorisation in particular. Enable it just for multiplication for now, which seems to be a performance win.
neildhar
temporarily deployed
to
docker-s3-upload
GitHub Actions
Inactive
neildhar
temporarily deployed
to
docker-s3-upload
GitHub Actions
Inactive
neildhar
temporarily deployed
to
docker-s3-upload
GitHub Actions
Inactive
manman-ren
reviewed
Oct 9, 2025
| if VECT_MUL: | ||
| # TODO: Figure out why vector FMA slows things down. | ||
| if VECT_MUL and False: | ||
| qk = _fma_f32x2(qk, qk_scale, -m_ij[:, None]) |
There was a problem hiding this comment.
Wondering if we should make VECT_MUL an integer with each bit representing one vectorization so we can autotune.
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Outdated suggestions cannot be applied.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.
Enabling
VECT_MULpreviously caused a regression, but this seems to come from the fma vectorisation in particular. Enable it just for multiplication for now, which seems to be a performance win.