[mpich-commits] [mpich] MPICH primary repository annotated tag, v3.1.3, created. v3.1.3

Service Account noreply at mpich.org
Wed Oct 8 09:35:44 CDT 2014


This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "MPICH primary repository".

The annotated tag, v3.1.3 has been created
        at  6845c55245868068a2794c9ee1205deefb46c41c (tag)
   tagging  835ab3bc2d051a44dc75aadd6ac1f9ace9b43c61 (commit)
  replaces  v3.2a1
 tagged by  Sangmin Seo
        on  Wed Oct 8 09:19:50 2014 -0500

- Log -----------------------------------------------------------------
tagging 'v3.1.3'

Igor Ivanov (4):
      netmod/mxm: Cleanup code from warnings
      netmod/mxm: Change supported MXM version to 3.1+
      netmode/mxm: Fix issue in anysource match
      mpi/comm: Fix MPI_Intercomm_merge

Junchao Zhang (7):
      Fix switch(datatype) kind of errors
      Fix F77/90 binding errors in MPI_Pcontrol
      Fix an "unused variable" warning in F77/90 MPI_PCONTROL
      Fix mixed code warnings reported by --enable-strict=c89
      Change // comments to /* */ style
      Remove unnessary array init in MPIU_CHKLMEM_DECL
      Revised the description of two MPI pvars

Ken Raffenetti (4):
      portals4: support for MPI_ANY_TAG recvs
      add libraries needed to build MPICH to EXTERNAL_LIBS
      portals4: cleanup usage of datatype info
      portals4: invalidate ME handle before request dequeue

Michael Blocksome (2):
      pamid: Add bgq libraries to wrapper scripts
      pamid: Do not compile/link C++ source if it is not used.

Min Si (1):
      Added mb in flush handler. Resolves #2177.

Norio Yamaguchi (2):
      Divide memory area to segments when registering
      Modify the method of IB-dereg_mr

Nysal Jan K.A (1):
      pamid: Fix a hang in scatter

Pavan Balaji (4):
      Fix incorrect portals library name.
      Cleanup release.pl to remove unnecessary code.
      Bug-fix: Expose sufficient memory as public.
      Bug-fix: look for man pages in the build directory, not source.

Rob Latham (4):
      no reason for a huge stack in these tests
      fix logic in syshint processing check
      Remove comm_split in deferred open case
      romio: remove allreduce in flush

Sameh Sharkawi (3):
      pamid: Adding empty implementation new MPID_Comm functions
      pamid: Fix for configure error in src/mpid/pamid/subconfigure.m4
      pamid: Collectives using process topology segv if FCA on

Sangmin Seo (3):
      Remove 3.2a1 changes in the CHANGES file.
      Update CHANGES for 3.1.3 release.
      Update version number for 3.1.3 release.

Su Huang (1):
      pamid: LE RDMA hit "mlx5: got completion with error" and even coredump

Wesley Bland (2):
      Turn on some of the FT tests that are now passing
      Revert "Turn on some of the FT tests that are now passing"

William Gropp (1):
      Create doctext index and suppress location

Xin Zhao (13):
      Bug-fix: avoid wrong synchronization in MPI_Win_complete.
      Bug-fix: skip synchronization in Win_start for MODE_NOCHECK.
      Enable SHM in PSCW tests.
      Bug-fix: waiting for ACKs for Active Target Synchronization.
      Pass tail pointer to MPL single direction linked-list API.
      Revert "Bug-fix: waiting for ACKs for Active Target Synchronization."
      Fix completion on target side in Active Target synchronization.
      Add a RMA test for target side completion in Active Target.
      Bug-fix: call RMA finish function when RMA ops finishes.
      Bug-fix: correct decrementing counter in GET-like operations.
      Modify variable name my_counter to at_completion_counter.
      Bug-fix: correct SHM barrier in MPI_Win_fence.
      Decrease iteration number in manyrma2 test.

halim amer (2):
      Added an example for checking on which CPU a process is running on.
      Added missing incrementation of the posted_recvq_match_attempts pvar

-----------------------------------------------------------------------


hooks/post-receive
-- 
MPICH primary repository


More information about the commits mailing list