fix allgather for empty buffers (facebookincubator#179) #356
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
malfet
approved these changes
Apr 20, 2023
|
@malfet has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
I used PyTorch-Ignite metrics that were all-gathered on both NCCL and Gloo backends.
In one of the cases the tensor (buffer) was empty and while NCCL tolerated this (hopefully as a no-op), Gloo gave me SIGFPE(8). I think this is because of the modulo by
dataSizepresent today in Gloo all-gather code paths.I believe this can be a no-op for Gloo, especially since
contextSize=1(rank) is already a no-op.