1
1
openmpi/opal
Brian Barrett 2a16c094f7 A slightly more correct fix for issues with reordering unlock calls. We want
to flush all writes pending (ie, the data being protected) out of the memory
manager before we write the spinlock unlock. Only need a wmb instead of
full mb, which is at least slightly less intrusive.  Also, after much
thought, no need for a memory barrier in init.

This commit was SVN r13649.
2007-02-14 16:37:31 +00:00
..
asm remove unneeded footer 2006-09-26 20:30:49 +00:00
class set object information in the internal macro, as that's what is used by 2007-01-31 21:51:06 +00:00
etc * fix a typo 2007-02-06 21:59:55 +00:00
event Force only one thread to come out from the 2007-01-25 07:36:42 +00:00
include A slightly more correct fix for issues with reordering unlock calls. We want 2007-02-14 16:37:31 +00:00
mca Don't complain on 64 bits architectures. 2007-02-09 03:17:02 +00:00
memoryhooks Rename liborte to libopen-rte and libopal to libopen-pal per telecon today 2006-12-05 18:27:24 +00:00
runtime Complete backout of prior sched_yield and paffinity changes 2007-02-07 14:22:37 +00:00
threads - Fix further compiler warnings regarding initialization 2007-01-30 06:34:38 +00:00
tools The extra includes should prefix the includedir, not the compile-time INCLUDEDIR when creating the -I arguments. 2007-02-12 22:43:50 +00:00
util Protect unistd.h usage. 2007-02-07 22:36:50 +00:00
win32 Be even more Windows friendly. 2007-02-07 22:39:03 +00:00
Makefile.am Rename liborte to libopen-rte and libopal to libopen-pal per telecon today 2006-12-05 18:27:24 +00:00
win_makefile Move all win32 related files in opal, and modify all the Makefiles.am to 2005-12-08 06:17:15 +00:00