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
..
base	ORTE_NAME_ARGS casts to long, not unsigned long.	2007-03-08 21:42:29 +00:00
gm	Finally the convertor export a convenience function to allow a consistent	2007-03-30 22:02:45 +00:00
mvapi	Finally the convertor export a convenience function to allow a consistent	2007-03-30 22:02:45 +00:00
mx	make the mx btl compile again...	2007-04-01 02:49:23 +00:00
openib	Finally the convertor export a convenience function to allow a consistent	2007-03-30 22:02:45 +00:00
portals	Correct an error introduced by commit r14180.	2007-04-02 02:59:23 +00:00
self	Finally the convertor export a convenience function to allow a consistent	2007-03-30 22:02:45 +00:00
sm	fixed typo	2007-04-03 17:21:54 +00:00
tcp	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
template	Finally the convertor export a convenience function to allow a consistent	2007-03-30 22:02:45 +00:00
udapl	Remove some warnings.	2007-04-02 19:11:06 +00:00
btl.h	Merging in the jjhursey-ft-cr-stable branch (r13912 : HEAD).	2007-03-16 23:11:45 +00:00
Makefile.am	fixes suggested by Ralf for supporting both Libtool 1 and 2 in Open MPI...	2005-12-19 03:10:23 +00:00