1. 26 Jun, 2015 1 commit
  2. 23 Jun, 2015 1 commit
  3. 15 Jun, 2015 1 commit
  4. 12 Jun, 2015 1 commit
  5. 11 Jun, 2015 1 commit
  6. 02 Jun, 2015 1 commit
  7. 30 May, 2015 3 commits
  8. 24 Apr, 2015 2 commits
  9. 16 Apr, 2015 1 commit
  10. 20 Mar, 2015 1 commit
    • Wesley Bland's avatar
      Extend timeouts for putfence1/getfence1 · 5c060dfd
      Wesley Bland authored
      These tests take a longer time on OSX due to the poor performance of
      malloc/calloc on OSX platforms. The extended timeout gives them a chance
      to complete.
      
      This could be better fixed in the future by only allocating the memory
      for the window once instead of over and over (~5000 times).
      
      No reviewer
      5c060dfd
  11. 09 Mar, 2015 1 commit
  12. 04 Mar, 2015 2 commits
  13. 01 Mar, 2015 1 commit
  14. 08 Feb, 2015 2 commits
  15. 20 Nov, 2014 1 commit
    • Min Si's avatar
      Add tests for SHM detection in win_create. · 0a69a9d0
      Min Si authored
      
      
      This program creates window with shm window buffer and checks the
      correctness of RMA operations issued through that window. It generates
      two tests with and without alloc_shm info, in which operations are
      issued out as SHM OP and as AM respectively.
      Signed-off-by: default avatarXin Zhao <xinzhao3@illinois.edu>
      0a69a9d0
  16. 12 Nov, 2014 1 commit
  17. 08 Nov, 2014 1 commit
  18. 05 Nov, 2014 1 commit
  19. 04 Nov, 2014 1 commit
    • Min Si's avatar
      Add req RMA op tests checking local completion. · db61d8d8
      Min Si authored
      
      
      Rput/accumulate + wait guarantees local completion, which means we can
      modify local buffer after wait is finished. These two tests check the local
      completion of Rput and Raccumulate by modifying local buffer after wait
      and then checking remote data. We expect the remote data should be equal
      to the local data before modifying.
      Signed-off-by: default avatarXin Zhao <xinzhao3@illinois.edu>
      db61d8d8
  20. 28 Sep, 2014 1 commit
    • Xin Zhao's avatar
      Add a RMA test for target side completion in Active Target. · 0e945ae7
      Xin Zhao authored
      
      
      In this test, during Active Target epoch, the origin process
      issues a large GET operation followed by a small PUT operation
      to the target process. Win_wait / Win_fence should guarantee
      all operations are completed at target when they returns. So
      after they return, we make the target process to update one
      window location accessed by that GET before. The GET operation
      should not get the updated value.
      Signed-off-by: Pavan Balaji's avatarPavan Balaji <balaji@anl.gov>
      0e945ae7
  21. 19 Sep, 2014 1 commit
  22. 03 Sep, 2014 1 commit
    • Min Si's avatar
      Added a test for flush on shared window. · f4518bf6
      Min Si authored
      
      
      Flush should guarantee operations are finished on both origin and target
      side. However, flush may return before the completion on target side in
      MPI implementation. It makes an error in this case: P0 and P1 allocate a
      shared window, and P2 locks both of them; P2 first put to P0 and flush,
      then get the updated data from P1. The put may complete on P0 after the
      completion of get on P1.
      Signed-off-by: default avatarXin Zhao <xinzhao3@illinois.edu>
      Signed-off-by: default avatarAntonio J. Pena <apenya@mcs.anl.gov>
      f4518bf6
  23. 27 Aug, 2014 1 commit
    • Xin Zhao's avatar
      Add a failed test for RMA on SHM. · cd0d4177
      Xin Zhao authored
      
      
      The added test mutex_bench_shm_ordered causes deadlock when
      MPIR_PARAM_CH3_ODD_EVEN_CLIQUES is set to 1. It is modified
      from mutex_bench_shm by changing the work distribution pattern
      from odd/even ranks to continuous ranks.
      
      See #2127.
      Signed-off-by: Pavan Balaji's avatarPavan Balaji <balaji@anl.gov>
      cd0d4177
  24. 07 Aug, 2014 2 commits
  25. 08 Jul, 2014 1 commit
  26. 07 Apr, 2014 1 commit
  27. 02 Feb, 2014 1 commit
    • Pavan Balaji's avatar
      Reduce the number of processes needed by rma-contig. · d1ac7712
      Pavan Balaji authored
      
      
      This test does not do anything special for 4 processes.  rank 0
      communicates with each other process sequentially, and waits for all
      communication to be over before communicating with a different target.
      Reducing the number of processes allows us to reduce the time taken by
      the test without affecting any tested parameter.
      Signed-off-by: default avatarWesley Bland <wbland@mcs.anl.gov>
      d1ac7712
  28. 16 Jan, 2014 1 commit
  29. 04 Dec, 2013 1 commit
  30. 31 Oct, 2013 2 commits
  31. 19 Oct, 2013 1 commit
  32. 01 Aug, 2013 1 commit
  33. 28 Jul, 2013 1 commit