[cute] Support explicit rank-1 column broadcasts - #3159
Open
ethche wants to merge 1 commit into
Open
Conversation
meta-cla
Bot
added
the
CLA Signed
Contributor
|
Failing CI? |
ethche
force-pushed
the
helion-cute-leading-k-tail
branch
from
July 25, 2026 01:50
c81f7ca to
11b4e53
Compare
ethche
force-pushed
the
helion-cute-rank1-leading-broadcast
branch
from
July 25, 2026 01:50
7b3d281 to
a159d68
Compare
ethche
force-pushed
the
helion-cute-rank1-leading-broadcast
branch
from
July 28, 2026 18:03
a159d68 to
523c0be
Compare
Contributor
Author
|
@jansel should be okay now? |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Recognize explicit column-vector epilogues such as:
acc + bias[tile_m].unsqueeze(-1)The auxiliary-operand analyzer now normalizes supported casts, scalar scaling, and rank-1 unsqueezes while retaining the explicit data axis. Classification validates index provenance, tile extent, global
extent, and storage layout.
Codegen constructs a logical 2D CuTe view with stride
(1, 0)over the original rank-1 storage and reuses the existing column-vector epilogue. No expanded tensor or materialized transpose is required.Bare rank-1 operands retain normal PyTorch trailing-axis broadcast semantics.