Mike Dubman d6ead2a3a5 Add support for routable ROCE where different subnet_id is a valid to proceed with MPI routing. ...

(can happen in the same LAN)
developed by vasily, reviewed by miked
cmr=v1.7.4:reviewer=ompi-gk1.7

This commit was SVN r29479.

2013-10-23 06:08:54 +00:00

..

base

As per the email discussion, revise the sparse handling of hostnames so that we avoid potential infinite loops while allowing large-scale users to improve their startup time:

2013-08-20 18:59:36 +00:00

mx

rename ompi_free_list operations to _mt, as per discussions at last face to face meeting

2013-07-08 22:07:52 +00:00

openib

Add support for routable ROCE where different subnet_id is a valid to proceed with MPI routing.

2013-10-23 06:08:54 +00:00

portals4

Remove the proc_pml and proc_bml fields from ompi_proc_t and replace with a

2013-08-30 16:54:55 +00:00

sctp

Silence few warnings.

2013-08-06 09:38:30 +00:00

self

rename ompi_free_list operations to _mt, as per discussions at last face to face meeting

2013-07-08 22:07:52 +00:00

sm

Remove two warnings from the SM BTL. The return code can be safely ignored

2013-08-28 06:36:01 +00:00

smcuda

Add runtime support to turn off CUDA IPC support.

2013-10-16 16:48:18 +00:00

tcp

Enable some CUDA-aware support on tcp btl. Only when configured in.

2013-10-04 12:50:16 +00:00

template

rename ompi_free_list operations to _mt, as per discussions at last face to face meeting

2013-07-08 22:07:52 +00:00

ugni

Fix typo in uGNI prepare source that could cause incorrect results with

2013-09-30 16:00:58 +00:00

usnic

Remove all of our "to-do" items from the README.txt.

2013-10-11 16:43:56 +00:00

vader

Make btl/vader valgrind safe.

2013-10-22 15:33:32 +00:00

btl.h

Fix support in smcuda btl so it does not blow up when there is no CUDA IPC support between two GPUs. Also make it so CUDA IPC support is added dynamically.

2013-08-21 21:00:09 +00:00

Makefile.am

Fix mistake that came in via the ompi-agen tree in r23764. The mistake wasn't part of the core autogen upgrade; it was an additional 'bonus' cleanup. Oops. The mistake will always create a set of directories under installdir, even if you do not --with-devel-headers. The set of directories will be empty, but still -- they should not be there at all. This commit fixes that -- the directories are not created at all if you do not --with-devel-headers

2010-09-24 22:53:28 +00:00