[Release v0.1.21] Cherry-pick _fold_seqlen_indptr cudagraph-safe fix (#5202) - #5222
Merged
Conversation
…copy) (#5202) * make _fold_seqlen_indptr cudagraph safe Signed-off-by: Micah Williamson <micah.williamson@amd.com> * [Claude] add unit tests Signed-off-by: Micah Williamson <micah.williamson@amd.com> * [Claude] fix formatting and tests Signed-off-by: Micah Williamson <micah.williamson@amd.com> --------- Signed-off-by: Micah Williamson <micah.williamson@amd.com> (cherry picked from commit 5d23231)
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Cherry-pick #5202 onto release/v0.1.21.
Original PR:
_fold_seqlen_indptrcudagraph-safe (avoid scalar H2D copy) #5202: [Bugfix] Make_fold_seqlen_indptrcudagraph-safe (avoid scalar H2D copy) #5202Original commit:
Backport commit:
This makes
_fold_seqlen_indptrcudagraph-safe by avoiding the scalar H2D copy (out[0] = 0), replacing the empty+indexed-write pattern withtorch.nn.functional.pad(cumsum, (1, 0), value=0), and adds unit tests for the folded indptr path in the persistent MLA decode tests.To be published as
v0.1.21.post1once merged (tag on the merge commit, release automation builds the wheel set).