- 05 Nov, 2012 10 commits
-
-
James Dinan authored
Reviewer: buntinas
-
James Dinan authored
Updated RMA implementation to track the passive target status individually, for each target. Includes new implementation for lock/unlock_all. Lock_all is currently unoptimized, see #1734 for future plans. Reviewer: buntinas
-
James Dinan authored
Remove ugly rank 0 hack and updated the active target code to maintain and process a proper RMA ops list. Reviewer: goodell
-
James Dinan authored
The use of a dense array is a temporary measure to support the reference implementation. This will be much improved by ticket #1735. Reviewer: goodell
-
James Dinan authored
A couple of the RMA synchronization error checks were returning the wrong error class. reviewer: buntinas
-
James Dinan authored
The old "lockRank" error checking is no longer sufficient in MPI 3.0 and must be removed to add support for locking multiple targets. Reviewer: balaji
-
James Dinan authored
Removed the use of lockRank in the passive target RMA code. This was a hack to start with, since lockRank belonged to the the front-end error checking and should not have been used below the ADI. Reviewer: buntinas
-
James Dinan authored
This is a temporary measure for the reference implementation that should be improved by ticket #1733. Reviewer: buntinas
-
James Dinan authored
In this patch, I have refactored the RMA ops list again to use the MPL UTList doubly-linked list and to treat the list as a proper object. This should set us up to work with multiple lists, as we will soon have one list per target. Doubly-linking the list is a big help in terms of maintainability (no more prevNext pointers) and flexibility (better implementation of request-based ops and other optimizations). Reviewer: goodell
-
Darius Buntinas authored
-
- 03 Nov, 2012 1 commit
-
-
William Gropp authored
[svn-r10506] More corrections for new mpi.h.in plus added missing special cases for old form where int* used instead of int[], in collective routines
-
- 02 Nov, 2012 4 commits
-
-
James Dinan authored
-
James Dinan authored
Request refs at the target were incorrectly set to two. Only one ref is held, by the progress engine. Reviewer: none
-
William Gropp authored
-
William Gropp authored
-
- 01 Nov, 2012 4 commits
-
-
Pavan Balaji authored
-
Pavan Balaji authored
Reviewed by buntinas.
-
William Gropp authored
-
William Gropp authored
-
- 31 Oct, 2012 2 commits
-
-
Pavan Balaji authored
tickets 125 and 126. Reviewed by buntinas.
-
Pavan Balaji authored
Reviewed by buntinas.
-
- 30 Oct, 2012 4 commits
-
-
William Gropp authored
[svn-r10495] More changes to adapt to limitations of early Fortran 90/95 compilers, for the new required (why?) MPI_Status derived type
-
William Gropp authored
-
William Gropp authored
[svn-r10493] Update C++ binding generation to correct handling of const and updates for MPI-3 prototype changes, including const and array notation
-
Darius Buntinas authored
-
- 29 Oct, 2012 2 commits
-
-
James Dinan authored
Reviewer: buntinas
-
James Dinan authored
Factored MPI_Set_info into separate external and internal interfaces, so we can use it from within MPI_Win/Comm_set_info. Reviewer: buntinas
-
- 28 Oct, 2012 1 commit
-
-
Pavan Balaji authored
No reviewer.
-
- 27 Oct, 2012 3 commits
-
-
William Gropp authored
-
William Gropp authored
-
William Gropp authored
-
- 26 Oct, 2012 8 commits
-
-
Darius Buntinas authored
-
Darius Buntinas authored
-
Pavan Balaji authored
-
Pavan Balaji authored
-
Pavan Balaji authored
and should not reside in the MPICH trunk. No reviewer.
-
Darius Buntinas authored
-
Darius Buntinas authored
-
Darius Buntinas authored
-
- 25 Oct, 2012 1 commit
-
-
James Dinan authored
Test for MPI-3 window info query routines. Reviewer: balaji
-