-
Notifications
You must be signed in to change notification settings - Fork 910
Fix some minor compatibility issues #3104
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
…tored against wildcard rank in the cray, s1, and s2 components, and that the ext1 component translates all wildcard rank requests into the peer's rank since v1.x of PMIx doesn't understand wildcard ranks Closes #3101 Signed-off-by: Ralph Castain <[email protected]>
This is not correct. In 2.1 we use this proc: |
@karasevb, please verify |
@artpol84 Kewl - I didn't know you had used the different name. However, just to be correct, that name should be changed to preserve the abstraction - it should be |
Agreed |
@rhc54, we did runtime evaluation. so it looks like Also it is funny, but pmix v1.2 doesn't care about wildcard to be passed on his input, the internal logic is written in the way that it will ignore that fact and work correctly. The change is good tough because otherwise we were breaking API and it was just a luck that pmix v1.2 is working. |
@jsquyres I don't want to merge this till I understand what happened on 2/21/17. After that point many tests failed with direct launch using aprun. Even ring_c failed. This was on master. |
@hppritcha #3011 was merged on 02/28/2017 so it should be unrelated. |
given that cray/pmix specific code was changed in this PR, and I haven't tried it yet. that's why. |
sorry I thought this was for v2.x since that's what the target label was. don't care if it goes in to master since its already broken for aprun direct launch. |
Ensure job-level data gets stored against wildcard rank in the cray, s1, and s2 components, and that the ext1 component translates all wildcard rank requests into the peer's rank since v1.x of PMIx doesn't understand wildcard ranks
Closes #3101
Needs to come over to 2.1 since the change that exposed the issue was already taken there.
Signed-off-by: Ralph Castain [email protected]