1
1
openmpi/ompi
Jeff Squyres c8bb7537e7 Remove include/opal/sys/cache.h -- its only purpose in life was to
#define CACHE_LINE_SIZE to 128.  This name has a conflict on NetBSD,
and it seems kinda odd to have a header file that ''only'' defines a
single value.  Also, we'll soon be raising hwloc to be a first-class
item, so having this file around seemed kinda weird.

Therefore, I replaced CACHE_LINE_SIZE with opal_cache_line_size, an
int (in opal/runtime/opal_init.c and opal/runtime/opal.h) on the
rationale that we can fill this in at runtime with hwloc info (trunk
and v1.5/beyond, only).  The only place we ''needed'' a compile-time
CACHE_LINE_SIZE was in the BTL SM (for struct padding), so I made a
new BTL_SM_ preprocessor macro with the old CACHE_LINE_SIZE value
(128).  That use isn't suitable for run-time hwloc information,
anyway.

This commit was SVN r23349.
2010-07-06 14:33:36 +00:00
..
attribute * Wrap all the direct error-code checks of the form (OMPI_ERR_* == ret) with 2010-05-17 23:08:56 +00:00
class Remove include/opal/sys/cache.h -- its only purpose in life was to 2010-07-06 14:33:36 +00:00
communicator * Wrap all the direct error-code checks of the form (OMPI_ERR_* == ret) with 2010-05-17 23:08:56 +00:00
config Merge in the modified thread configure option branch per today's telecon. 2010-03-16 23:10:50 +00:00
contrib Added workaround for PGI compiler bug (see http://www.pgroup.com/support/release_tprs_90.htm TPR 4337): 2010-06-15 07:16:13 +00:00
datatype Refs trac:2273 2010-03-16 00:47:10 +00:00
debuggers Also add <stdlib.h> for exit(). 2010-06-28 15:17:42 +00:00
errhandler Fix bugs where (OMPI_ERROR == *) checks cannot be converted to (OMPI_SUCCESS != *) since the return codes are overloaded to return an "index" on success. 2010-05-18 20:54:11 +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 Clean up request handling in the I/O framework to be more consistent with 2009-11-26 05:13:43 +00:00
group Fix bugs where (OMPI_ERROR == *) checks cannot be converted to (OMPI_SUCCESS != *) since the return codes are overloaded to return an "index" on success. 2010-05-18 20:54:11 +00:00
include Convert the line endings for the added header files. They were changed automatically by Windows when adding new files. 2010-02-16 17:24:44 +00:00
info - As discussed revert r21330, Fortran-configure info should 2009-06-01 19:02:34 +00:00
mca Remove include/opal/sys/cache.h -- its only purpose in life was to 2010-07-06 14:33:36 +00:00
mpi Updates about thread support levels. 2010-07-02 13:14:09 +00:00
mpiext Add an "affinity" Open MPI extension (also describe the 2010-04-21 17:28:08 +00:00
op Fix bugs where (OMPI_ERROR == *) checks cannot be converted to (OMPI_SUCCESS != *) since the return codes are overloaded to return an "index" on success. 2010-05-18 20:54:11 +00:00
peruse - Sanity check initialization and finalization of PERUSE. 2010-01-12 16:36:24 +00:00
proc * Wrap all the direct error-code checks of the form (OMPI_ERR_* == ret) with 2010-05-17 23:08:56 +00:00
request Fix the configure logic for --with-ft so that it properly takes a comma separated list. 2010-03-12 23:57:50 +00:00
runtime Some paffinity functions had their return status overloaded: 2010-05-21 16:55:28 +00:00
tools Fix a typo that breaks ompi_info if --enable-sensors 2010-07-02 02:38:55 +00:00
win - Replace combinations of 2009-08-20 11:42:18 +00:00
CMakeLists.txt Use variables instead of hard-coded compiler flags, in order to support various C/C++ compilers on Windows. 2010-04-21 12:45:00 +00:00
Makefile.am This commit converts us to the "one big libmpi" scheme that has been 2010-02-23 22:20:01 +00:00