jsquyres [Tue, 08 Dec 2009 20:32:37 +0000] rev 17048
We should never make another tarball off the 1.3 branch, but just in
case we do, ensure to require LT 2.2.6b.
jsquyres [Fri, 04 Feb 2011 20:51:10 +0000] rev 17047
Nope -- not ready for rc1 yet. RM wants to wait for the MPI_GATHER
fix first.
jsquyres [Fri, 04 Feb 2011 13:38:03 +0000] rev 17046
Make rc1
jsquyres [Wed, 02 Feb 2011 17:40:01 +0000] rev 17045
As discussed here:
http://www.open-mpi.org/community/lists/devel/2011/01/8894.php
http://blogs.cisco.com/performance/building-3rd-party-open-mpi-components/
Contribute a sample of how to build MCA components outside of the Open
MPI source tree.
This version is specific to the v1.4 tree.
jsquyres [Wed, 02 Feb 2011 15:00:46 +0000] rev 17044
Fixes #2412: Prevent the SM BTL from creating the mmaped file on remote mounted directories
Custom patch from Jeff Squyres; looks like it was somehow missed when
initially merging this stuff from trunk -> v1.4.
jsquyres [Tue, 01 Feb 2011 15:21:20 +0000] rev 17043
Fixes #2703: Move debugger multi-token fix to v1.4
Patch based on r24322:
As noted on the mailing list by Gabriele Fatigati
(http://www.open-mpi.org/community/lists/users/2011/01/15427.php), the
--tv (and friends) switches to mpirun would effectively munge the
orterun command line together and then split it apart again before
exec'ing the underlying debugger. We would therefore lose multi-token
argv[x] value and split them into multiple tokens. For example:
mpirun --tv -np 2 a.out "foo bar"
would get launched with "foo" and "bar" as separate arguments; not one
argument. This was due to the underlying code joining the argv into a
single string and then re-splitting it. This commit removed the argv
join; it now does the parsing and re-jigering of the argv by only
looking at each individual argv item; multi-word tokens like "foo bar"
will never be split into separate tokens.
jsquyres [Thu, 27 Jan 2011 20:29:24 +0000] rev 17042
Fixes #2697: Move uninitialized variable fix to v1.4
r24296
Also need to check the "flag" value, because if flag!=true, then the
value of "local_spawn" (and "non_mpi") is not set by ompi_info_get*().
jsquyres [Thu, 27 Jan 2011 20:27:09 +0000] rev 17041
Fixes #2657: Move MPI_INIT_THREAD man page clarification to v1.4
r24173
Clarification that calling MPI_INIT_THREAD with MPI_THREAD_SINGLE is
the same as calling MPI_INIT.
jsquyres [Thu, 27 Jan 2011 20:23:32 +0000] rev 17040
Fixes #2695: CMR: Update LANL Platform Files in 1.4 Branch
jsquyres [Tue, 18 Jan 2011 16:35:09 +0000] rev 17039
Fixes #2676: Fix incorrect text in MPI_FILE_SET_VIEW man page
r24201
Correct an error in MPI_FILE_SET_VIEW man page: the fh does ''not''
have to have an identical ''value'' (it must be the same file handle,
but that means nothing about its actual value). But the datarep must
have an identical value. Additionaly, the etype does not need to be
identical, but the extent of all the etypes supplied must be
identical.
See MPI-2.2 p401-402 for further details.