- 16 Oct, 2012 2 commits
-
-
James Dinan authored
This fixes FOP segv failures when configured with --enable-g=mem, and partially addresses ticket #1697.
-
William Gropp authored
-
- 15 Oct, 2012 3 commits
-
-
Anthony Chan authored
[svn-r10384] added check of receive buffer values for the inplace blocking collectives added in r10186. Added the equivalent tests of the inplace nonblocking collectives. This resolves req#1623.
-
William Gropp authored
-
William Gropp authored
[svn-r10377] More fixes for inter-language attribute use. Tested on x86 and 64-bit SPARC; there are still errors because of the incorrect use of int where MPI_Fint is needed, but this will work for most users, and the MPI_Fint case has been removed from MPI-3
-
- 12 Oct, 2012 2 commits
-
-
Darius Buntinas authored
-
Darius Buntinas authored
-
- 11 Oct, 2012 9 commits
-
-
James Dinan authored
Removed dead code from passive target RMA processing. Reviewer: none
-
James Dinan authored
True_lb is always zero for predefined types; there's no need to include this in address calculations when processing predefined types. Reviewer: goodell
-
Pavan Balaji authored
No reviewer.
-
James Dinan authored
Removed unused local index variables in strided tests with indexed datatypes. Reviewer: none
-
James Dinan authored
Reviewer: goodell
-
James Dinan authored
Reviewer: goodell
-
James Dinan authored
-
James Dinan authored
-
James Dinan authored
-
- 10 Oct, 2012 2 commits
-
-
David Goodell authored
By setting "indent-tabs-mode:nil" we should hopefully begin to slowly squeeze out hard tabs from the source without a disruptive (to downstream projects) whitespace-fixing change. No reviewer.
-
William Gropp authored
-
- 09 Oct, 2012 1 commit
-
-
Pavan Balaji authored
Reviewed by goodell.
-
- 08 Oct, 2012 3 commits
-
-
Pavan Balaji authored
Kaz!
-
David Goodell authored
The ABI string (c:r:a) was bumped from 8:0:0 to 8:1:0 since the interface has not changed AFAICT. No reviewer.
-
William Gropp authored
[svn-r10342] Added test of the MPI-1 version of the Fortran attributes. This is needed to ensure that these work correctly without needing to run the MPICH-1 or Intel tests. Added because the current MPICH2 build appears to fail to handle the MPI-1 Fortran attribute put/get correctly
-
- 05 Oct, 2012 2 commits
-
-
Pavan Balaji authored
distributed independently. Thanks to IBM for reporting the issue with "make dist" in the test suite. No reviewer.
-
James Dinan authored
An error class conversion was missing from test/mpi/error/errinstatwa.c Reviewer: Balaji
-
- 04 Oct, 2012 12 commits
-
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
build example executables. No reviewer.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
library as well. Reviewed by goodell.
-
Pavan Balaji authored
Added a configure flag --enable-versioning, which is on by default. This causes libtool to version shared libraries, using libtool conventions. Some platforms may use other versioning schemes, for example PE MPI uses directory structure layout to version libraries. In addition, PE must generate consistent, non-versioned "libmpi.so" output from ldd on the final link of the application to allow for 2 way binary compatibility with other MPI implementations. Reviewed by goodell.
-
Pavan Balaji authored
success, instead of relying on an assignment statement.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
libraries. This is required on BG/Q. Reviewed by goodell.
-
Pavan Balaji authored
pamid. No reviewer.
-
Pavan Balaji authored
No reviewer.
-
- 03 Oct, 2012 4 commits
-
-
Darius Buntinas authored
-
Darius Buntinas authored
-
Darius Buntinas authored
-
Darius Buntinas authored
-