openmpi

comm

We stripped the process info packing routine to minimize message size when sending the launch message, but tools still require all the info. So modify the tool-hnp handshake to explicitly add the missing info

2013-12-19 20:42:20 +00:00

dash_host

Initialize variable so dash-host option gets correctly parsed

2014-01-08 15:17:16 +00:00

hostfile

Prevent a segfault if we encounter an error while parsing a hostfile. Don't issue and error_log output as the hostfile code already prints an error message

2014-04-12 21:32:10 +00:00

context_fns.c

MCA/base: Add new MCA variable system

2013-03-27 21:09:41 +00:00

context_fns.h

... Delayed due to notifier commits earlier this day ...

2009-04-29 01:32:14 +00:00

error_strings.c

When we comm_spawn, we really want to respect the original -host directives and not expand the daemon virtual machine unless directed to do so in the comm_spawn command. Otherwise, we will automatically launch daemons on every node in the allocation.

2014-04-30 22:26:18 +00:00

error_strings.h

When we comm_spawn, we really want to respect the original -host directives and not expand the daemon virtual machine unless directed to do so in the comm_spawn command. Otherwise, we will automatically launch daemons on every node in the allocation.

2014-04-30 22:26:18 +00:00

help-regex.txt

Fix regular expression analyzer for slurmd - use a slurm-specific version

2011-07-13 22:49:56 +00:00

hnp_contact.c

As per the RFC, bring in the ORTE async progress code and the rewrite of OOB:

2013-08-22 16:37:40 +00:00

hnp_contact.h

As requested by Aurelien at the July design meeting - long time coming, but finally got around to it.

2008-12-10 17:10:39 +00:00

Makefile.am

Use the correct abstraction layer name for the data dirs

2014-05-08 14:32:24 +00:00

name_fns.c

As per the RFC, bring in the ORTE async progress code and the rewrite of OOB:

2013-08-22 16:37:40 +00:00

name_fns.h

Roll in the rest of the modex change. Eliminate all non-modex API access of RTE info from the MPI layer - in some cases, the info was already present (either in the ompi_proc_t or in the orte_process_info struct) and no call was necessary. This removes all calls to orte_ess from the MPI layer. Calls to orte_grpcomm remain required.

2012-06-27 14:53:55 +00:00

nidmap.c

More leak fixes.

2014-05-15 16:38:13 +00:00

nidmap.h

Introduce staged execution. If you don't have adequate resources to run everything without oversubscribing, don't want to oversubscribe, and aren't using MPI, then staged execution lets you (a) run as many procs as there are available resources, and (b) start additional procs as others complete and free up resources. Adds a new mapper as well as a new state machine.

2012-08-28 21:20:17 +00:00

parse_options.c

Use ports as multicast channels instead of networks so we avoid stepping into reserved spaces.

2011-04-29 18:46:40 +00:00

parse_options.h

Use ports as multicast channels instead of networks so we avoid stepping into reserved spaces.

2011-04-29 18:46:40 +00:00

pre_condition_transports.c

Addressing Jeff's comments:

2014-02-24 23:18:35 +00:00

pre_condition_transports.h

In the case of direct-launched processes running under slurm, psm requires that the pre_condition_transports MCA param be set. This is normally computed by mpirun and inserted into each proc's environ, but that doesn't work here.

2011-04-28 13:54:33 +00:00

proc_info.c

proc_info.c: uncomment C99 struct member initialization usage

2014-04-18 17:26:07 +00:00

proc_info.h

Update a couple of fields, add a scheduler field to proc_info

2014-02-13 23:30:04 +00:00

regex.c

Extend the regular expression parsing support

2014-03-17 21:25:05 +00:00

regex.h

Extend the regular expression parsing support

2014-03-17 21:25:05 +00:00

session_dir.c

Per Jeff's suggestion, remove function that has duplicate functionality and just use one to check if session_dir directory should be removed.

2014-05-08 17:22:43 +00:00

session_dir.h

- Revert r20739

2009-03-05 21:56:03 +00:00

show_help.c

Revert r31596 and r31595 as they recreate the "abort" problem - all they did was move the blocking send to another point in the code. An alternative solution to the "show_help and abort" problem. will come in another commit

2014-05-02 10:38:30 +00:00

show_help.h

Remove the old configure option for disabling full rte support - we now use the OMPI rte framework for such purposes

2013-02-28 01:35:55 +00:00