CORE: Fix deadlock in asymmetric collectives check - #1300
Open
QiaoK wants to merge 1 commit into
Open
Conversation
QiaoK
force-pushed
the
ucc_dt_check_fix
branch
3 times, most recently
from
April 18, 2026 01:19
d98507a to
e2b8317
Compare
Scatterv/gatherv in TL/UCP and scatter/scatterv/gatherv in TL/NCCL can hang when ranks use non-predefined (derived) datatypes. The TL rejects unsupported types locally without collective coordination, so some ranks fall back to the MPI layer while others proceed into UCC, causing a deadlock. Disable these collectives by default via score=0 tuning until collective fallback is properly supported.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Scatterv/gatherv in TL/UCP and scatter/scatterv/gatherv in TL/NCCL can hang when ranks use non-predefined (derived) datatypes. The TL rejects unsupported types locally without collective coordination, so some ranks fall back to the MPI layer while others proceed into UCC, causing a deadlock. Disable these collectives by default via score=0 tuning until collective fallback is properly supported.