Add MPI barrier before all tests#2148
Closed
brownbaerchen wants to merge 1 commit into
Closed
Conversation
Contributor
|
Thank you for the PR! |
Member
|
Should we remove some other barriers at the start of tests within this PR? |
Collaborator
Author
Actually, I didn't find any places where there is a barrier at the start of the test. Only at the end. This is very similar, but since we don't understand the problem, I don't want to do everything at once. Can you approve this PR, such that the CI runs and we see if this fixes the problem at all? Then we can try removing the barriers at the end of IO tests. |
JuanPedroGHM
approved these changes
Feb 24, 2026
Collaborator
Author
|
So, this is not the solution, sadly. Let's go with the barrier in a specific place suggested in #2147 then.. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The CI currently runs into deadlocks with some IO tests. We have not understood why, but seemingly adding some MPI barriers helps. In this PR, I simply add a global MPI barrier before every test that is derived from the basic
TestCaseusing thesetUpfunction.Since I cannot reproduce the problem locally and we don't understand the problem, I have no clue if this fixes the issue. Nevertheless, I don't think this hurts and I'd rather do this in a centralized place for all tests than randomly spamming barriers in the code.
Schauen wir mal was wird.
Looking forward to being able to just do this on my fork without requiring a review from somebody else once we have merged #2139 ;)