- 06 Oct, 2010 6 commits
-
-
William Gropp authored
[svn-r7312] Correct use of MPI_ERR_NO_MEM (should be MPI_ERR_OTHER) for out-of-memory that does not pertain to MPI_Alloc_mem and add detail to error messages about out-of-memory
-
William Gropp authored
-
William Gropp authored
-
Darius Buntinas authored
-
Darius Buntinas authored
-
Guillaume Mercier authored
-
- 05 Oct, 2010 3 commits
-
-
Jayesh Krishna authored
[svn-r7305] Adding windows dll stubs and exports for the new wrapper func for fortran interface to grab CS. The new wrapper func was added in r7290
-
Jayesh Krishna authored
[svn-r7303] Let the dll be built inside the TraceInput dir instead of rlog dir - The installer expects the dll to be inside TraceInput dir. Also see r7301
-
Jayesh Krishna authored
-
- 04 Oct, 2010 2 commits
-
-
Darius Buntinas authored
-
Darius Buntinas authored
-
- 03 Oct, 2010 4 commits
-
-
Pavan Balaji authored
-
Pavan Balaji authored
is always generated. We might not use it when it's packaged within MPICH2, but it makes the release tarball creation easier.
-
Pavan Balaji authored
checkpoint-restart.
-
Pavan Balaji authored
-
- 01 Oct, 2010 2 commits
-
-
Darius Buntinas authored
[svn-r7292] replacing progress_test() in shared memory send functions with calls to shared memory progress function. This addresses several issues related to recursive calls to progress.
-
Darius Buntinas authored
-
- 30 Sep, 2010 15 commits
-
-
Darius Buntinas authored
-
Jayesh Krishna authored
[svn-r7289] MPI Cluster debugger settings are different for different versions of VS - Add remark that the current settings are for VS 2010
-
David Goodell authored
No reviewer.
-
Jayesh Krishna authored
[svn-r7287] Fixing windows makefile due to changes in r7262 - src files for TraceInput were moved to a subdir
-
David Goodell authored
No reviewer.
-
David Goodell authored
These are caused by padding in C's long double _Complex type. No reviewer.
-
Pavan Balaji authored
-
Pavan Balaji authored
-
Pavan Balaji authored
-
Pavan Balaji authored
deprecated function MPI_Attr_put was. Reviewed by goodell.
-
Pavan Balaji authored
other MPI functions. So before they are called, we need to release the internally held lock. Reviewed by goodell.
-
Pavan Balaji authored
MPIR_ functions can be called by multiple different MPI_ functions causing deadlocks if they try to acquire locks again. Reviewed by goodell.
-
David Goodell authored
No reviewer.
-
William Gropp authored
-
William Gropp authored
-
- 29 Sep, 2010 8 commits
-
-
David Goodell authored
Fixes the specific warning from tt#1038. There are still a few other valgrind warnings from ch3:sock with the MPICH2 test suite. Reviewed by buntinas@.
-
Rajeev Thakur authored
[svn-r7274] reverting r7261 so that we can fix the --enable-threads=multiple issue. Maybe we should also change the name to enable-threads=force_multiple or something
-
William Gropp authored
[svn-r7273] Added a option, similar to one available in MPICH2, to force MTest_Init to select a thread-level other than THREAD_SINGLE, permitting more extensive testing of thread-enabled MPI implementations
-
William Gropp authored
-
Pavan Balaji authored
multi-threaded mode. Fixes ticket #1105. Reviewed by goodell.
-
Darius Buntinas authored
-
Jayesh Krishna authored
[svn-r7269] Handling VS warnings - Removing unreferenced labels/vars, type casting appropriately, fix func defns to use the *right* types
-
Darius Buntinas authored
-