ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
29 649 коммитов 3 Ветки 0 Теги
Граф коммитов

635 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
044c561cba Roll to latest PMIx master 2016-06-16 17:30:30 -07:00
Ralph Castain
5d330d5220 Enable the PMIx event notification capability and use that for all error notifications, including debugger release. This capability requires use of PMIx 2.0 or above as the features are not available with earlier PMIx releases. When OMPI master is built against an earlier external version, it will fallback to the prior behavior - i.e., debugger will be released via RML and all notifications will go strictly to the default error handler. 
Add PMIx 2.0

Remove PMIx 1.1.4

Cleanup copying of component

Add missing file

Touchup a typo in the Makefile.am

Update the pmix ext114 component

Minor cleanups and resync to master

Update to latest PMIx 2.x

Update to the PMIx event notification branch latest changes
 2016-06-14 13:08:41 -07:00
Ralph Castain
d58da99dbc Shift to memcpy to avoid Solaris issues 2016-06-09 12:07:17 -07:00
Ralph Castain
8fa935534b Abstract the strnlen function for environments that do not have it (e.g., Solaris 10) 2016-06-08 10:12:43 -07:00
Gilles Gouaillardet
b707d138fe pmix114/pmix1_client: fix misc memory leaks 
Fixes CID 1325146-1325149
 2016-06-06 09:33:35 +09:00
Ralph Castain
ecea1e3bb5 Update to 1.1.4rc3 2016-06-01 20:56:07 -07:00
Ralph Castain
12ecf972af Split the pmix external component into one for the 1.1.4 release, and another for the upcoming 2.0 release. Clean up the configury so the components look for a series-specific function instead of running a program. 
NOTE: the changes for the 2.0 series are not yet in the PMIx master.
 2016-06-01 14:15:24 -07:00
Ralph Castain
55923eacd3 Stealing some pieces of Josh Hursey's PR #1583 and modifying a bit, allow the opal/pmix external component to handle both PMIx 1.1.4 and PMIx 2.0 versions. Automatically detect the version of the target external library and adjust the only two APIs that changed (PMIx_Init and PMIx_Finalize) 
Rename temp vars in .m4 to avoid conflict with Travis
 2016-05-27 08:06:31 -07:00
Artem Polyakov
725eea2819 Fix base64 implementation in pmix framework. 
In the commit 80f07b65f16e9538aca7fc5e124d2074e7e0b69e setting of '-' marker used
as the string termination sign was moved from base64 code:
from: 80f07b65f1 (diff-1b10896c267d2591dc2c08fd0542ab67L491)
to: 80f07b65f1 (diff-1b10896c267d2591dc2c08fd0542ab67R189)

However the decoding function wasn't fixed and still expects on extra
byte at the end of the encoded string which leads to data truncation
during extraction (was noticed on standalone code that was using base64
from OMPI).
 2016-05-23 23:30:31 +06:00
Gilles Gouaillardet
8466a3daf3 pmix: update .gitignore 
git ignore opal/mca/pmix/pmix114/pmix/include/pmix/autogen/config.h.in
git rm opal/mca/pmix/pmix114/pmix/include/pmix/autogen/config.h.in
git ignore opal/mca/pmix/pmix*/...
 2016-05-23 11:58:07 +09:00
Gilles Gouaillardet
cbbdce05b1 pmix/pmix114: silence a warning 2016-05-20 09:35:26 +09:00
Ralph Castain
6f743f81b6 Update PMIx 114 to current release candidate 2016-05-19 12:55:05 -07:00
rhc54
8b534e9897 Merge pull request #1668 from rhc54/topic/slurm 
When direct launching applications, we must allow the MPI layer to pr…
 2016-05-16 12:23:19 -07:00
Howard Pritchard
1a676e5b35 pmix/cray: fix some breakage 
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
 2016-05-16 12:45:05 -05:00
Ralph Castain
01ba861f2a When direct launching applications, we must allow the MPI layer to progress during RTE-level barriers. Neither SLURM nor Cray provide non-blocking fence functions, so push those calls into a separate event thread (use the OPAL async thread for this purpose so we don't create another one) and let the MPI thread sping in wait_for_completion. This also restores the "lazy" completion during MPI_Finalize to minimize cpu utilization. 
Update external as well

Revise the change: we still need the MPI_Barrier in MPI_Finalize when we use a blocking fence, but do use the "lazy" wait for completion. Replace the direct logic in MPI_Init with a cleaner macro
 2016-05-14 16:37:00 -07:00
Ralph Castain
7767882346 Per user request, add some missing data and definitions: 
OPAL_PMIX_UNIV_RANK - synonym for OPAL_PMIX_GLOBAL_RANK
OPAL_PMIX_APP_SIZE - #ranks in the application of this proc
 2016-05-09 08:39:01 -07:00
Ralph Castain
8ec1891d11 Silence warning 2016-05-05 20:04:10 -07:00
Ralph Castain
08022d7af1 Some minor cleanups of warnings from gcc 6.0.0. Update s1/s2 pmix to get max_procs as required. 2016-05-05 15:28:13 -07:00
rhc54
648043597a Merge pull request #1612 from ggouaillardet/poc/pmix_external_configury 
pmix/external: revamp external pmix package detection
 2016-05-02 09:46:05 -07:00
Jeff Squyres
265e5b9795 Merge pull request #1552 from kmroz/wip-hostname-len-cleanup-1 
ompi/opal/orte/oshmem/test: max hostname length cleanup
 2016-05-02 09:44:18 -04:00
Gilles Gouaillardet
45f9a47d77 pmix/external: fix typo and silence a warning 2016-05-02 17:15:52 +09:00
Gilles Gouaillardet
08d91b9a03 pmix/external: revamp external pmix package detection 2016-05-02 16:23:31 +09:00
Ralph Castain
42d9d861fc Fix minor typo in PMIx packing of pmix_app_t - thanks to Gilles for pointing it out 2016-04-29 08:55:46 -07:00
Howard Pritchard
f52dd511d4 Merge pull request #1600 from hppritcha/topic/pmix_fix_for_finalize 
pmix/cray: set fence_nb to NULL
 2016-04-28 13:50:15 -06:00
hppritcha
aa1d7b9c50 pmix/cray: set fence_nb to NULL 
Rather than have a stub function for the pmix fence_nb
operation, just set to NULL.  Causes fewer problems.

Fixes #1597
Fixes #1527

Signed-off-by: hppritcha <howardp@lanl.gov>
 2016-04-28 13:48:54 -05:00
Ralph Castain
02876564d4 Silence warning of zero-byte malloc 2016-04-26 11:55:59 -07:00
Karol Mroz
e1c64e6e59 opal: standardize on max hostname length 
Define OPAL_MAXHOSTNAMELEN to be either:
  (MAXHOSTNAMELEN + 1) or
  (limits.h:HOST_NAME_MAX + 1) or
  (255 + 1)

For pmix code, define above using PMIX_MAXHOSTNAMELEN.

Fixup opal layer to use the new max.

Signed-off-by: Karol Mroz <mroz.karol@gmail.com>
 2016-04-24 08:19:47 +02:00
Gilles Gouaillardet
d96919638f pmix: remote autogenerated file and update .gitignore 
removed: opal/mca/pmix/pmix114/pmix/src/include/private/autogen/config.h.in
 2016-04-18 12:57:41 +09:00
Ralph Castain
b009e58d25 Roll to PMIx 1.1.4rc2 - replaces some code that was incorrectly removed in prior update 2016-04-16 18:24:24 -07:00
Ralph Castain
8ff114e668 Update to official PMIx 1.1.4rc1 2016-04-15 21:47:46 -07:00
Ralph Castain
449ec41532 Roll to PMIx 1.1.4rc1 and remove the PMIx 1.2.0 directory as the community has decided to not do that release version. This incorporates a number of bug fixes that have been identified and repaired in the PMIx and OMPI code bases. Also includes several minor corrections to the PMIx code so it now supports run-thru without hanging on collectives involving a process that exits 2016-04-15 10:11:11 -07:00
Ralph Castain
2432daf065 Some minor cleanups of a memory leak and error output 2016-04-08 07:46:18 -07:00
Rainer Keller
52080a5736 As per the pull request to pmix/master: 
https://github.com/pmix/master/pull/71

Have OMPI's current version of pmix120 nicely fail in case of
too long sun_path (longer than 108 or in case of OSX 103 chars).
And have OMPI return proper error messages with hints how to
amend.
 2016-04-07 22:12:53 +02:00
Gilles Gouaillardet
6f450630d8 pmix/external: fix misc missing conversion and type issues 2016-04-04 10:12:34 +09:00
Gilles Gouaillardet
2ede47c462 pmix: fix misc missing conversion and type issues 2016-04-04 10:12:34 +09:00
Nysal Jan K.A
75233573d1 pmix: Increment the reference count in PMIx_Init 
The reference counting was broken which led PMIx_Finalize
to release resources early. This fixes the "use after free" scenarios
that I encountered.

(based on commit pmix/master@abfaa4c)
 2016-03-27 04:11:25 -04:00
Josh Hursey
099170bb31 Merge pull request #1496 from jjhursey/topic/pmix120-obj-patch 
pmix/pmix120: Fix OBJ_ to PMIX_ symbol name
 2016-03-25 19:35:43 -05:00
Ralph Castain
0b4310b186 Remove an unnecessary header that forced exposure of the PMIx internal headers 2016-03-25 16:57:41 -07:00
Ralph Castain
e8246e079b Minor cleanup to match the changes in the PMIx master 2016-03-25 15:12:41 -07:00
Joshua Hursey
8ebeaa5861 pmix/pmix120: Fix OBJ_ to PMIX_ symbol name 2016-03-25 16:17:08 -05:00
Ralph Castain
af1444b6e1 Cleanup a debug statement. Plug a memory leak 2016-03-08 18:27:55 -08:00
Ralph Castain
bac6290b22 Ensure the process name is positive when using direct launch 
Fixes #1425
 2016-03-08 08:31:05 -08:00
Ralph Castain
b57a191ccc Update the external client to the new PMIx init/finalize signatures 2016-03-03 20:50:20 -08:00
Ralph Castain
4a55fba414 Fix registration of error handlers thru the pmix120 component. A thread-shift operation was hanging on the sync_event_base, which made it dependent on someone calling opal_progress. Unfortunately, a process in "sleep" or spinning outside the MPI library won't do that, and so we never complete errhandler registration. 2016-03-02 15:01:01 -08:00
Ralph Castain
011403c04a Fix a number of issues, some of which have lingered for a long time: 
* provide a more reliable way of determining that a process is a singleton by leveraging the schizo framework. Add new components for slurm, alps, and orte to detect when we are in a managed environment, and if we have been launched by mpirun or a native launcher. Set the correct envars to control ess and pmix selection in each case.

* change the relative priority of the pmix120 and pmix112 components to make pmix120 the default

* fix singleton comm-spawn by correctly setting the num_apps field of the orte_job_t created by the daemon - this fixes a segfault in register_nspace on newly created daemons

* ensure orterun doesn't propagate any ess or pmix directives in its environment

* Cleanup a few valgrind issues and memory leaks

* Fix a race condition that prevented the client from completing notification registrations (missing thread shift)

* Ensure the shizo/alps component detects launch by mpirun
 2016-03-01 06:53:00 -08:00
Ralph Castain
d28d3ee901 Make the error message on external pmix library a little clearer by separating out the libevent from the libhwloc checks 2016-02-24 11:20:25 -06:00
Ralph Castain
d653cf2847 Convert the orte_job_data pointer array to a hash table so it doesn't grow forever as we run lots and lots of jobs in the persistent DVM. 2016-02-21 11:55:49 -08:00
Ralph Castain
8c92a179c0 Minor memory leak 2016-02-19 15:05:39 -08:00
Ralph Castain
6e68d758b9 Cleanup some valgrind complaints about jumps with uninitialized values. Fix a few IOF issues reported by Mark Santcroos when submitting jobs from tools. Add the ability to pass directives to the --output-filename option that tell ORTE to (a) not include the jobid in the path to the output files, and (b) not to copy the output to the tool (i.e., just store it in the files). 
ck

Remove stale debug

Fix a segfault if no subscribers are present
 2016-02-18 16:30:37 -08:00
Ralph Castain
60a7bc2e50 Enable the PMIx notification callback system. This currently is only supported by the pmix120 component, which is not selected by default. All other components will ignore error registration requests, and thus do not support debugger attach when launched via mpirun. Note that direct launched applications will support such attachment, but may not do so in a scalable fashion. 
Fixes ##1225
 2016-02-18 09:29:12 -08:00