Uh oh!
There was an error while loading. Please reload this page.
Fix incorrect return value from broker queue enqueue - #1
Open
gosLp wants to merge 1 commit into
Open
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
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Return the result of
ensureEnqueue()fromBrokerWorkDistributorT::enqueue()instead of returningfalseunconditionally.Problem
When queue capacity is available,
enqueue()reserves a slot successfully, callsputData(), but then returnsfalse. It therefore reports failure even though the item was inserted.The current state of Mustard ignores this return value so this wouldn't change existing LU or Cholesky results. However, any caller that checks the Boolean result may incorrectly report failure or retry on a previously successful insertion.
Tests
Manually tested with small single-process, single-GPU NVSHMEM reproducer on one NVIDIA A100 (node with 4xA100) using CUDA 12.6 and NVSHMEM from NVHPC 24.11.
The reproducer:
0xC0FFEE.Before the change, the item was successfully inserted and dequeued, but
enqueue()returnedfalse.After the change, the same operation returned
true, and the inserted value was successfully dequeued.