Safe GPU-aware MPI communication#1313
Merged
Merged
Conversation
Technici4n
approved these changes
May 5, 2026
Collaborator
Technici4n
left a comment
Technici4n
left a comment
There was a problem hiding this comment.
Ah yes I remember this
mfherbst
reviewed
May 5, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Before MPI communication of GPU buffers, the device must be synchronized. If not, there is a real risk of race condition because CPU and GPU run asynchronously. This has no effect on CPU performance. On the GPU, this cannot be avoided.
For reference, here is a PR updating the relevant documentation in MPI.jl. The problem has been known for a while, but never properly documented. The PR also links to a slack thread where the problem is discussed.