Mikhail Brinskii 404c480068 COLL/TUNED: Update alltoall selection rule for mlx ...

Use linear with sync alltoall algorithm for certain message/comm size
ranges. Does not affect default fixed decision, unless HPCX (with its
custom parameters) is used or corresponding mca is set.

Signed-off-by: Mikhail Brinskii <mikhailb@mellanox.com>

2019-07-13 23:27:40 +03:00

..

clang

Save the clang settings for detecting alignment issues - I don't want to have to remember the cmd line jango

2014-09-10 18:56:19 +00:00

embedded

orte/ras: remove loadleveler support

2017-03-21 10:32:28 -06:00

hadoop

Purge whitespace from the repo

2015-06-23 20:59:57 -07:00

ibm

platform: Remove "with_verbs" from all the platform files.

2019-02-07 05:36:06 -08:00

intel/bend

platform: Remove "with_verbs" from all the platform files.

2019-02-07 05:36:06 -08:00

lanl

platform: Remove "with_verbs" from all the platform files.

2019-02-07 05:36:06 -08:00

mellanox

COLL/TUNED: Update alltoall selection rule for mlx

2019-07-13 23:27:40 +03:00

ornl

Purge whitespace from the repo

2015-06-23 20:59:57 -07:00

snl

platform: Remove "with_verbs" from all the platform files.

2019-02-07 05:36:06 -08:00

utk

platform: Remove "with_verbs" from all the platform files.

2019-02-07 05:36:06 -08:00

cray_xt_cnl_romio

- Change the property of a few files, that obviously

2009-08-11 01:40:00 +00:00

optimized

tarball doesn't disable threads, neither should we

2007-02-19 19:23:15 +00:00

ps3

Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately.

2008-02-28 01:57:57 +00:00