Skip to content

Conversation

@hjelmn
Copy link
Member

@hjelmn hjelmn commented Oct 20, 2015

This commit fixes several bugs in the osc/rdma component:

  • Complete aggregated requests immediately. Completion of RMA
    requests indicates local completion anyway. This fixes a hang in
    the c_reqops test.
  • Correctly mark Rget_accumulate requests.
  • Set the local base flag correctly on the local peer.
  • Clear or set the no locks flag on the window if the value is
    changed by MPI_Win_set_info.
  • Actually update the target when using MPI_OP_REPLACE.

Fixes #1010

Signed-off-by: Nathan Hjelm hjelmn@lanl.gov

This commit fixes several bugs in the osc/rdma component:

 - Complete aggregated requests immediately. Completion of RMA
   requests indicates local completion anyway. This fixes a hang in
   the c_reqops test.

 - Correctly mark Rget_accumulate requests.

 - Set the local base flag correctly on the local peer.

 - Clear or set the no locks flag on the window if the value is
   changed by MPI_Win_set_info.

 - Actually update the target when using MPI_OP_REPLACE.

Fixes open-mpi#1010

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
@hjelmn
Copy link
Member Author

hjelmn commented Oct 20, 2015

@rolfv This fixes the MTT problems you identified.

There is still an outstanding issue with the openib btl that MTT does not test. The connection manager is not completely asynchronous. This can cause issues when using passive-target to a peer that has not been communicated with before. I am working on a fix for that issue.

hjelmn added a commit that referenced this pull request Oct 20, 2015
@hjelmn hjelmn merged commit 763744a into open-mpi:master Oct 20, 2015
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

ibm/c_accumulate (and friends) all fail with master

1 participant