Skip to content

OSC rdma: make sure accumulating in shared memory is safe #6747

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

Merged
merged 1 commit into from
Jun 14, 2019

Conversation

devreal
Copy link
Contributor

@devreal devreal commented Jun 7, 2019

Even if the peer is reachable through shared memory there may still be remote peers updating directly through the network, which leads to conflicting updates, i.e., if osc_rdma_acc_single_intrinsic is set. This PR adds more thorough checks to make sure it's safe to accumulate in shared memory.

This is #6649 for v4.0.x

Fixes #6536

Signed-off-by: Joseph Schuchart [email protected]

Signed-off-by: Joseph Schuchart <[email protected]>
(cherry picked from commit c67e229)
@ompiteam-bot
Copy link

Can one of the admins verify this patch?

@gpaulsen
Copy link
Member

gpaulsen commented Jun 7, 2019

ok to test.

@gpaulsen gpaulsen added the NEWS label Jun 7, 2019
@devreal devreal changed the title OSC rdma: make sure accumulating in shared memory is safe Rdma fetchop local v4.0.x OSC rdma: make sure accumulating in shared memory is safe Jun 7, 2019
@gpaulsen gpaulsen added this to the v4.0.2 milestone Jun 7, 2019
@gpaulsen gpaulsen requested a review from hppritcha June 7, 2019 18:53
@gpaulsen gpaulsen added NEWS and removed NEWS labels Jun 7, 2019
@hppritcha hppritcha merged commit cb8dd56 into open-mpi:v4.0.x Jun 14, 2019
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.

4 participants