ch4/{ofi,ucx}: Improved error checking for business card exchange

raffenet commented 2 weeks ago

Pull Request Description

Business card exchange in roots only mode involves communication over a special "init" communicator. Take care to check for errors that may arise during the exchange and skip the cleanup of the init comm in that case to avoid masking the source of the error. Fixes pmodels/mpich#7200.

Author Checklist

[x] Provide Description Particularly focus on why, not what. Reference background, issues, test failures, xfail entries, etc.
[x] Commits Follow Good Practice Commits are self-contained and do not do two things at once. Commit message is of the form: module: short description Commit message explains what's in the commit.
[ ] Passes All Tests Whitespace checker. Warnings test. Additional tests via comments.
[x] Contribution Agreement For non-Argonne authors, check contribution agreement. If necessary, request an explicit comment from your companies PR approval manager.

raffenet commented 2 weeks ago

test:mpich/ch4/most

raffenet commented 2 weeks ago

test:mpich/ch4/most

raffenet commented 2 weeks ago

test:mpich/ch4/most

raffenet commented 2 weeks ago

test:mpich/ch4/most

raffenet commented 2 weeks ago

test:mpich/ch4/most

raffenet commented 2 weeks ago

test:mpich/ch4/most

raffenet commented 2 weeks ago

test:mpich/ch4/most

colleeneb commented 2 weeks ago

This is ok to try testing on Aurora now, correct?

raffenet commented 2 weeks ago

This is ok to try testing on Aurora now, correct?

Yes, this branch is OK to test now.

pmodels / mpich

ch4/{ofi,ucx}: Improved error checking for business card exchange #7201

Pull Request Description

Author Checklist