Repository navigation
[CUDA] Fix gather_mm - #3503
Merged
Merged
[CUDA] Fix gather_mm#3503
Conversation
zcbenz
force-pushed
the
cuda-gather-gemm
branch
2 times, most recently
from
May 10, 2026 11:12
30ef7e1 to
b104e41
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The sm70/sm80 kernels of CUTLASS v3 are in a weird state: no one uses them, no one maintains them, they have bugs and I had a really hard time trying to find a maintainer who actually understand the kernels to look into the bugs.
Anyway it is much easier to just write the kernel in plain CuTe and it should be better for long term maintenance since the code no longer requires domain-specific knowledge of CUTLASS C++ APIs.
This PR removes the last entries in
cuda_skip.pyand I'm deleting the file, for future TODO items in CUDA backend we can just use@unittest.skipIf.