Skip to content

Conversation

@rolfv
Copy link

@rolfv rolfv commented Aug 14, 2015

This is to fix things so we set the macro to 0 when no CUDA-aware. Otherwise, the macro is always set to 1 which is obviously wrong.

@jsquyres can you look at this one?

I still may try and change it so things are undefined, but that is still under discussion and I would like it to at least work as it was originally intended.

@rolfv rolfv added the bug label Aug 14, 2015
@rolfv rolfv added this to the Open MPI v2.0.0 milestone Aug 14, 2015
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You should put $CUDA_SUPPORT in brackets, i.e., [$CUDA_SUPPORT].

@jsquyres
Copy link
Member

Other than those 2 minor comments, looks good.

@rolfv rolfv force-pushed the pr/fix-cuda-ext-again branch from c37b1a8 to 0e87478 Compare August 14, 2015 17:57
@rolfv
Copy link
Author

rolfv commented Aug 14, 2015

Fixed those two issues. Will wait for more testing before merging.

@rolfv
Copy link
Author

rolfv commented Aug 14, 2015

bot:retest

rolfv pushed a commit that referenced this pull request Aug 14, 2015
Fix macro return value when not CUDA-aware
@rolfv rolfv merged commit 652a685 into open-mpi:master Aug 14, 2015
@rolfv rolfv deleted the pr/fix-cuda-ext-again branch August 14, 2015 18:40
jsquyres pushed a commit to jsquyres/ompi that referenced this pull request Sep 19, 2016
pml/ob1: add missing ompi_request_wait_completion for buffered sends
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants