openmpi

Eugene Loh e91d3a6f13 Fix test that checked whether C data type matched Fortran REAL*16.

The test was comparing 1.1L to Fortran's 1.1, which was sometimes
a single-precision/truncated value, even if stored in a REAL*16 word.

This commit was SVN r26543.

2012-06-01 23:33:00 +00:00

attribute

== Highlights ==

2012-04-18 15:57:29 +00:00

class

Cleanup some compiler warnings when doing an optimized/non-debug build.

2012-04-04 20:40:16 +00:00

communicator

Per RFC, bring in the following changes:

2012-05-07 14:52:54 +00:00

config

Fix test that checked whether C data type matched Fortran REAL*16.

2012-06-01 23:33:00 +00:00

contrib

Changes to VT:

2012-05-23 09:06:22 +00:00

datatype

Fixes trac:3109: patch from Patrick LeDot/Bull -- fix MPI_COMPLEX8, 16,

2012-05-30 14:08:30 +00:00

debuggers

My apologies for doing this outside of the usual time restrictions, but we need to get this in so we can make progress.

2012-01-11 15:53:09 +00:00

errhandler

Fixes trac:3093: when cleaning up, only reset items in the

2012-05-16 15:26:32 +00:00

etc

Many thanks to Ralf W. for finding a subtle bug in these Makefile.am's

2008-06-04 01:28:03 +00:00

file

== Highlights ==

2012-04-18 15:57:29 +00:00

group

== Highlights ==

2012-04-18 15:57:29 +00:00

include

A bunch of fixes to the f08 bindings:

2012-05-29 17:54:33 +00:00

info

== Highlights ==

2012-04-18 15:57:29 +00:00

mca

ugni: update to latest btl code. bug fixes and cleanup

2012-05-31 20:02:41 +00:00

message

* Implement matched probe for the CM PML. Required adding a peer field to

2012-03-22 22:55:59 +00:00

mpi

A bunch of fixes to the f08 bindings:

2012-05-29 17:54:33 +00:00

mpiext

Move hwloc_cpuset_t prettyprint routines down into the hwloc base:

2012-06-01 16:02:18 +00:00

== Highlights ==

2012-04-18 15:57:29 +00:00

peruse

- Sanity check initialization and finalization of PERUSE.

2010-01-12 16:36:24 +00:00

proc

Per RFC, bring in the following changes:

2012-05-07 14:52:54 +00:00

request

== Highlights ==

2012-04-18 15:57:29 +00:00

runtime

Per RFC, bring in the following changes:

2012-05-07 14:52:54 +00:00

tools

add exit code to mpijavac dependent on the result of javac command

2012-05-29 18:44:08 +00:00

win

== Highlights ==

2012-04-18 15:57:29 +00:00

CMakeLists.txt

* s/f77/fortran/gi in a lot of places

2012-04-27 01:09:37 +00:00

Makefile.am

Oops - if we don't want the Java bindings, then we really shouldn't be building them :-/

2012-05-23 19:51:27 +00:00