- 07 Jun, 2013 4 commits
-
-
Pavan Balaji authored
-
Pavan Balaji authored
Fixes tt#1871.
-
Pavan Balaji authored
-
Pavan Balaji authored
-
- 05 Jun, 2013 1 commit
-
-
William Gropp authored
This must be re-enabled once the errors in the error calls are fixed.
-
- 02 Jun, 2013 1 commit
-
-
William Gropp authored
Discovered this error while fixing an unrelated problem. Note that autogen will detect that errors were encountered while building the error message tables and will fail, so the problems discovered by this fix need to be addressed.
-
- 01 Jun, 2013 1 commit
-
-
William Gropp authored
This bug was found by the Tianhe-2 software group, and applies only to the device-override path for the MPI_Bcast operation. Because of a missing trailing underscore, the wrong bcast override function could be selected. As our tests don't check the override path, this went undetected.
-
- 31 May, 2013 1 commit
-
-
Rajeev Thakur authored
-
- 29 May, 2013 2 commits
-
-
Replace all instances of bzero(x,n) with memset(x,0,n) per the recommendation of http://pubs.opengroup.org/onlinepubs/009696899/functions/bzero.html Signed-off-by:
Michael Blocksome <blocksom@us.ibm.com> Signed-off-by:
Pavan Balaji <balaji@mcs.anl.gov>
-
Signed-off-by:
Pavan Balaji <balaji@mcs.anl.gov>
-
- 22 May, 2013 4 commits
-
-
Pavan Balaji authored
-
William Gropp authored
-
William Gropp authored
-
William Gropp authored
Updated buildiface for C++ (relevant for MPI 2.2) to create the Distgraph C++ interfaces. Also fixed errors in the C++ tests where the tests failed to include the configure-generated configuration file mpitestconf.h before included files that depended upon that file. In the process of creating the tests for the C++ Distgraph routines, discovered errors in the error checking in the MPI dist graph routine to extract the graph data. This commit does not fix that error.
-
- 21 May, 2013 3 commits
-
-
Michael Blocksome authored
-
Michael Blocksome authored
This reverts commit e4fa6c06.
-
Michael Blocksome authored
-
- 20 May, 2013 5 commits
-
-
-
-
Attempting to use 'MPI_IN_PLACE' with a pamid collective and a pami library that does not support 'PAMI_IN_PLACE' will abort. Versions of pami that support `PAMI_IN_PLACE` will be available in the near future.
-
William Gropp authored
Observed that while there was a "NEEDS_DECL" for getsid in the pm/util programs, there was no test for this case (getsid is not defined by default and requires the correct level of XOPEN_SOURCE). Also removed now extraneous tests from remshell subconfigure.
-
Michael Blocksome authored
configure now checks for: - PAMI_CLIENT_NONCONTIG - PAMI_CLIENT_MEMORY_OPTIMIZE - PAMI_GEOMETRY_NONCONTIG - PAMI_GEOMETRY_MEMORY_OPTIMIZE
-
- 19 May, 2013 7 commits
-
-
Pavan Balaji authored
-
Pavan Balaji authored
-
Pavan Balaji authored
-
Pavan Balaji authored
Fixed a typo where we used HAVE_SYSCTL_H instead of HAVE_SYS_SYSCTL_H. This was causing the check to always fail. No reviewer.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
No reviewer.
-
- 18 May, 2013 11 commits
-
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
For the VA_ARGS test, clang was accepting it in C89 mode, but throwing warnings. This causes configure to enable it, but the build to throw many warnings. This patch will disable VA_ARGS support unless the compiler can accept it without warnings. No reviewer.
-
Pavan Balaji authored
Added a check for inline in the MPL configure. Also made C89 headers available by default. We should not be checking for them. No reviewer.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
Thanks to Jeff Hammond for reporting it. No reviewer.
-
Pavan Balaji authored
No reviewer.
-
Pavan Balaji authored
We have been allocating a "large enough" pool of temporary strings and hoping that we don't overflow it for our temporary allocations. This commit provides macros that better check for overflow problems without making the code too clumsy. No reviewer.
-
With MS-MPI 64-bit from HPC Pack 2008 and 2012, MPI_Attr_get returns error code 773 when called from delete_fn on a communicator obtained from MPI_Comm_split. The standard is not explicit that the 'comm' argument of delete_fn must be valid, so this test is only in effect when !USE_STRICT_MPI. Signed-off-by:
Jed Brown <jedbrown@mcs.anl.gov> Signed-off-by:
Satish Balay <balay@mcs.anl.gov>
-
The old method attempts to create a failed compile when the function *is* already declared, by creating an incompatible declaration. That is insufficient with clang-3.2, which only warns in case of incompatible redeclaration of a library function: conftest.c:44:5: warning: incompatible redeclaration of library function 'strncmp' int strncmp(double, int, double, const char *); ^ /usr/include/string.h:143:12: note: 'strncmp' is a builtin with type 'int (const char *, const char *, size_t)' extern int strncmp (const char *__s1, const char *__s2, size_t __n) ^ 1 warning generated. Instead, we take the opposite approach, generating a failed compile when the library function has *not* been declared in the header. void (*fptr)(void) = (void(*)(void))strncmp; Signed-off-by:
Jed Brown <jedbrown@mcs.anl.gov>
-
(ibm) D189371 (ibm) b02a38e0680dbd1471884439214ee6afe13baeb4 (ibm) c8ebe0da80732297de985c3640caa4c5623c55d4 (ibm) 48d1993041c3820a468d2ee6c838ff0a5bc1154a
-