George Bosilca 20f0ec584a A tricky optimization. On my test machine it improve the bandwidth by about 3Mb/s out of 580Mb/s. But ... the real interest is for small to middle size unexpected messages. The unexpected messages are copied by the PML in it's own unexpected buffers. Therefore, there is no reason to make a first copy in the TCP BTL. The BTL can handle to the PML it's own buffer, and can be sure that once the callback completed it can reuse the buffer, no matter what happened with the fragment. This commit was SVN r14320.		2007-04-12 04:52:29 +00:00
..
allocator	Change the way we handle unexpected messages,	2007-03-22 01:00:34 +00:00
bml	Small cleanup.	2007-04-12 04:34:53 +00:00
btl	A tricky optimization. On my test machine it improve the bandwidth by about 3Mb/s out of 580Mb/s. But	2007-04-12 04:52:29 +00:00
coll	Remove some warnings.	2007-04-02 19:11:06 +00:00
common	This fixes the initialization of the usable size of the shared memory.	2007-03-07 13:28:06 +00:00
crcp	Some cleanup.	2007-04-05 13:52:05 +00:00
io	fix for aggregated writes	2007-04-11 22:07:19 +00:00
mpool	bring disable_sbrk back online, there was a change to properly support AIX	2007-03-29 16:11:00 +00:00
mtl	Merging in the jjhursey-ft-cr-stable branch (r13912 : HEAD).	2007-03-16 23:11:45 +00:00
osc	Added new module for intercomm collectives. This will require an	2007-03-27 02:06:42 +00:00
pml	two fixes, one mine, the other gleb's, I'm committing for gleb due to	2007-04-11 22:03:06 +00:00
rcache	Be consistent about the definitions of mca_mpool_base_page_size and	2007-03-28 14:14:05 +00:00
topo	Merging in the jjhursey-ft-cr-stable branch (r13912 : HEAD).	2007-03-16 23:11:45 +00:00