[Bugfix] [B200] cutlass_mla - ensure kv_split == 1 for batch size > 1 #25509
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
Signed-off-by: Alexander Matveev <[email protected]>
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request addresses a critical bug that causes hangs on SM100 architectures when the batch size is greater than one. The fix involves limiting max_splits to 1 in such cases, which correctly disables the problematic reduction kernel as described. The code change is accurate, and the corresponding comment has been updated to reflect the new logic. This is a solid workaround for a significant stability issue.
mgoin
approved these changes
Sep 23, 2025
mgoin
added
bug
pavanimajety
approved these changes
Sep 23, 2025
FeiDaLI
pushed a commit
to FeiDaLI/vllm
that referenced
this pull request
Sep 25, 2025
…vllm-project#25509) Signed-off-by: Alexander Matveev <[email protected]>
yewentao256
pushed a commit
that referenced
this pull request
Oct 3, 2025
…#25509) Signed-off-by: Alexander Matveev <[email protected]> Signed-off-by: yewentao256 <[email protected]>
gjc0824
pushed a commit
to gjc0824/vllm
that referenced
this pull request
Oct 10, 2025
…vllm-project#25509) Signed-off-by: Alexander Matveev <[email protected]> Signed-off-by: gaojc <[email protected]>
xuebwang-amd
pushed a commit
to xuebwang-amd/vllm
that referenced
this pull request
Oct 10, 2025
…vllm-project#25509) Signed-off-by: Alexander Matveev <[email protected]> Signed-off-by: xuebwang-amd <[email protected]>
choprahetarth
pushed a commit
to Tandemn-Labs/vllm
that referenced
this pull request
Oct 11, 2025
…vllm-project#25509) Signed-off-by: Alexander Matveev <[email protected]>
lywa1998
pushed a commit
to lywa1998/vllm
that referenced
this pull request
Oct 20, 2025
…vllm-project#25509) Signed-off-by: Alexander Matveev <[email protected]>
xuebwang-amd
pushed a commit
to xuebwang-amd/vllm
that referenced
this pull request
Oct 24, 2025
…vllm-project#25509) Signed-off-by: Alexander Matveev <[email protected]> Signed-off-by: xuebwang-amd <[email protected]>
rtourgeman
pushed a commit
to rtourgeman/vllm
that referenced
this pull request
Nov 10, 2025
…vllm-project#25509) Signed-off-by: Alexander Matveev <[email protected]>
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.
Tests showed that limiting kv_split to 2 is not enough and it needs to be limited to 1 to ensure no-hangs (1 disables the sm100 cutlass reduction kernel)