openib: Disable CUDA async by default #5748
Merged
Conversation
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
Disable async receive for CUDA under OpenIB. While a performance optimization, it also causes incorrect results for transfers larger than the GPUDirect RDMA limit. This change has been validated and approved by Akshay. References open-mpi#3972 Signed-off-by: Brian Barrett <[email protected]> (cherry picked from commit 9344afd) Signed-off-by: Brian Barrett <[email protected]>
Akshay-Venkatesh
approved these changes
Sep 20, 2018
|
@bwbarret would you mind changing the commit message to 'disabling cuda async recv'? Cuda async is broader. |
|
@Akshay-Venkatesh kind of ugly to do that for just the branch commits :/. |
|
Should be ok then. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Disable async receive for CUDA under OpenIB. While a performance
optimization, it also causes incorrect results for transfers
larger than the GPUDirect RDMA limit. This change has been validated
and approved by Akshay.
References #3972
Signed-off-by: Brian Barrett [email protected]
(cherry picked from commit 9344afd)
Signed-off-by: Brian Barrett [email protected]