George Bosilca a9c6ae8f15 Minimize the number of branches, and orce the correct prediction for the ...

most usual one. Most of the time we expect the functions which allocate
requests to succeed.

This commit was SVN r12344.

2006-10-27 23:16:13 +00:00

..

configure.params

Update the copyright notices for IU and UTK.

2005-11-05 19:57:48 +00:00

Makefile.am

A long time waiting patch. Get rid of the comm->c_pml_procs. It was (and that was

2006-09-20 22:14:46 +00:00

pml_ob1_comm.c

A long time waiting patch. Get rid of the comm->c_pml_procs. It was (and that was

2006-09-20 22:14:46 +00:00

pml_ob1_comm.h

A long time waiting patch. Get rid of the comm->c_pml_procs. It was (and that was

2006-09-20 22:14:46 +00:00

pml_ob1_component.c

A long time waiting patch. Get rid of the comm->c_pml_procs. It was (and that was

2006-09-20 22:14:46 +00:00

pml_ob1_component.h

The last patch for Windows support. Mostly casting and conversion to C++ friendly headers.

2006-08-24 16:38:08 +00:00

pml_ob1_endpoint.c

Next step in the project split, mainly source code re-arranging

2006-02-12 01:33:29 +00:00

pml_ob1_endpoint.h

Minor cleanups. On the OB1 PML the endpoint is not used => remove it from the build.

2006-07-13 00:07:13 +00:00

pml_ob1_hdr.h

More endian-related cleanups:

2006-02-26 00:45:54 +00:00

pml_ob1_iprobe.c

Next step in the project split, mainly source code re-arranging

2006-02-12 01:33:29 +00:00

pml_ob1_irecv.c

PERUSE support for OB1. There we go, now the trunk has a partial peruse implementation.

2006-03-31 17:09:09 +00:00

pml_ob1_isend.c

A long time waiting patch. Get rid of the comm->c_pml_procs. It was (and that was

2006-09-20 22:14:46 +00:00

pml_ob1_progress.c

Next step in the project split, mainly source code re-arranging

2006-02-12 01:33:29 +00:00

pml_ob1_rdma.c

The last patch for Windows support. Mostly casting and conversion to C++ friendly headers.

2006-08-24 16:38:08 +00:00

pml_ob1_rdma.h

allow pml pipeline to cache memory registrations

2006-02-09 15:49:51 +00:00

pml_ob1_rdmafrag.c

make ompi_free_list_item_t a class..

2006-06-12 16:44:00 +00:00

pml_ob1_rdmafrag.h

The last patch for Windows support. Mostly casting and conversion to C++ friendly headers.

2006-08-24 16:38:08 +00:00

pml_ob1_recvfrag.c

Fix for a common race condition when running the Sandia mt_send_recv.cc test.

2006-10-26 19:09:39 +00:00

pml_ob1_recvfrag.h

The last patch for Windows support. Mostly casting and conversion to C++ friendly headers.

2006-08-24 16:38:08 +00:00

pml_ob1_recvreq.c

Look like a big commit but in fact it address only one issue. The way we're working with

2006-10-17 20:20:58 +00:00

pml_ob1_recvreq.h

Big datatype commit. Remove all unused features of the datatype engine. As the memory

2006-10-26 23:11:26 +00:00

pml_ob1_sendreq.c

Minimize the number of branches, and orce the correct prediction for the

2006-10-27 23:16:13 +00:00

pml_ob1_sendreq.h

Last set of explicit conversions. We are now close to the zero warnings on

2006-10-20 03:57:44 +00:00

pml_ob1_start.c

Update the comment to reflect what's inside the code.

2006-10-27 23:09:37 +00:00

pml_ob1.c

Some process_pending() functions get bml_btl on which resource was freed as a

2006-10-26 13:21:47 +00:00

pml_ob1.h

Some process_pending() functions get bml_btl on which resource was freed as a

2006-10-26 13:21:47 +00:00

post_configure.sh

* include the correct file if we are doing the component bypass thing with ob1

2006-02-22 16:16:38 +00:00