Add VarLen support to new flash attention api #616
+233
−54
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: Chen, Xi2 <[email protected]>
Signed-off-by: Chen, Xi2 <[email protected]>
Signed-off-by: Chen, Xi2 <[email protected]>
ClarkChin08
changed the title
Signed-off-by: Chen, Xi2 <[email protected]>
tdeng5
requested review from
Antonyvance,
jiyang1011,
petercad,
rolandschulz and
tdeng5
petercad
approved these changes
Nov 7, 2025
petercad
left a comment
petercad
left a comment
There was a problem hiding this comment.
varlen-related changes look OK (reviewed causal mask changes in a separate PR). Just to check, we have some accuracy issues with the old FlashAttention + varlen + causal code when the driver stopped zero-initializing memory. Should double-check that it's not happening here.
tdeng5
approved these changes
Nov 9, 2025
taozha2
approved these changes
Nov 10, 2025
jiyang1011
approved these changes
Nov 10, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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Outdated suggestions cannot be applied.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.
No description provided.