ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
840e2cb4a2
openmpi/orte/mca
История
Mike Dubman 840e2cb4a2 mindist: cosmetic, use fallback to byslot if unable to read NUMA info, small fix. 
fixed by Elena, reviewed by Ralph/Mike
cmr=v1.7.4:reviewer=ompi-gk1.7

This commit was SVN r29679.
2013-11-13 09:26:40 +00:00
..
dfs As per the RFC, bring in the ORTE async progress code and the rewrite of OOB: 2013-08-22 16:37:40 +00:00
errmgr ***** THIS INCLUDES A SMALL CHANGE IN THE MPI-RTE INTERFACE ***** 2013-10-08 18:37:59 +00:00
ess Initialize orte_coprocessors hash table to NULL. Delay coprocessor detection on HNP until after node topology final definition in case rmaps changes it. Minor spacing change. 2013-10-24 00:08:47 +00:00
filem As per the RFC, bring in the ORTE async progress code and the rewrite of OOB: 2013-08-22 16:37:40 +00:00
grpcomm **************************************************************** 2013-10-14 16:52:58 +00:00
iof The function orte_iof_base_setup_prefork attempts to create a pty for 2013-09-15 15:33:51 +00:00
odls Singletons currently start out by spawning an HNP - this is required solely in the cases where the singleton subsequently calls MPI_Comm_spawn or publishes port info without support from an external orte-server. In all other cases, the HNP is of no value and can actually be a detriment by creating additional overhead on the node. This is particularly concerning for async operations where processes may begin as singletons and then dynamically wireup to perform pt2pt communications. 2013-10-04 02:58:26 +00:00
oob Ouch - remove an extraneous line. 2013-11-13 04:02:05 +00:00
plm Revise the launch timeout detection so we don't mistakenly declare "failed to start". Recognize that timeout is at the per-job level, and define the timeout param as a total value instead of seconds/daemon as it otherwise can get to be an enormous (and useless) number. 2013-11-11 23:50:40 +00:00
ras As per the RFC, bring in the ORTE async progress code and the rewrite of OOB: 2013-08-22 16:37:40 +00:00
rmaps mindist: cosmetic, use fallback to byslot if unable to read NUMA info, small fix. 2013-11-13 09:26:40 +00:00
rml Revert r29166 in favor of a better solution from George 2013-09-18 01:41:26 +00:00
routed Singletons currently start out by spawning an HNP - this is required solely in the cases where the singleton subsequently calls MPI_Comm_spawn or publishes port info without support from an external orte-server. In all other cases, the HNP is of no value and can actually be a detriment by creating additional overhead on the node. This is particularly concerning for async operations where processes may begin as singletons and then dynamically wireup to perform pt2pt communications. 2013-10-04 02:58:26 +00:00
sensor As per the RFC, bring in the ORTE async progress code and the rewrite of OOB: 2013-08-22 16:37:40 +00:00
snapc As per the RFC, bring in the ORTE async progress code and the rewrite of OOB: 2013-08-22 16:37:40 +00:00
sstore Correctly register the component MCA parameters. 2013-06-15 16:05:09 +00:00
state As per the RFC, bring in the ORTE async progress code and the rewrite of OOB: 2013-08-22 16:37:40 +00:00