Skip to content

Conversation

jjhursey
Copy link
Member

For user defined MPI_Op types the o_name field is never set. Some components use this field for verbose output, and leaving it undefined makes the output difficult to review. So this commit makes sure that it gets initialized in all paths.

@jjhursey jjhursey added this to the v2.0.1 milestone Jun 30, 2016
@jjhursey jjhursey self-assigned this Jun 30, 2016
@ibm-ompi
Copy link

Build Failed with XL compiler! Please review the log, and get in touch if you have questions.

@ibm-ompi
Copy link

Build Failed with GNU compiler! Please review the log, and get in touch if you have questions.

@jsquyres
Copy link
Member

👍

@jjhursey
Copy link
Member Author

(IBM Jenkins) Something strange is happening with the IBM cluster this morning - I'm trying to diagnose.

@jjhursey
Copy link
Member Author

bot:retest

@ibm-ompi
Copy link

Build Failed with XL compiler! Please review the log, and get in touch if you have questions.

Gist: https://gist.github.com/5bc41e28bae363aaacfe61645dd6425f

@hjelmn
Copy link
Member

hjelmn commented Jul 1, 2016

:bot:retest

@lanl-ompi
Copy link
Contributor

Test FAILed.

@jjhursey
Copy link
Member Author

jjhursey commented Jul 1, 2016

Humm. The LANL Jenkins is showing something the the IBM Jenkins was showing on some other PRs:

[ompi:32674] listen_thread: accept() failed: Invalid argument (22).

I wonder if this is a regression in the master branch that only gets triggered on occasion. It is not related to this PR.

bot:retest

@jjhursey jjhursey force-pushed the topic/op-init-fix branch from c7adddd to 0671e45 Compare July 1, 2016 18:46
@ibm-ompi
Copy link

ibm-ompi commented Jul 1, 2016

Build Failed with XL compiler! Please review the log, and get in touch if you have questions.

Gist: https://gist.github.com/812577ac894303c1148f2b700a7e715a

@hjelmn
Copy link
Member

hjelmn commented Jul 1, 2016

@jjhursey Yeah. Seeing that periodically on all Jenkins. Maybe something in pmix?

@jjhursey
Copy link
Member Author

jjhursey commented Jul 1, 2016

The IBM failure is due to an unexpected system outage. Once it's back i'll get it to retest.

@jjhursey jjhursey merged commit 9b4ed96 into open-mpi:master Jul 1, 2016
@jjhursey jjhursey deleted the topic/op-init-fix branch July 1, 2016 20:22
@Di0gen
Copy link

Di0gen commented Jul 3, 2016

bot:mellanox:retest

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants