1
1
Граф коммитов

24161 Коммитов

Автор SHA1 Сообщение Дата
Igor Ivanov
36d3a7aa6c contrib: Add bash script to measure performance
This script is useful to measure times from launching ompi
application to different internal points. A user can easy add
it`s test basing on existing tests.
See readme information inside the script.
2015-12-14 17:42:19 +02:00
rhc54
513092bc25 Merge pull request #1201 from rhc54/topic/essdflt
Don't be so prescriptive about the ess component to be used
2015-12-09 23:12:42 -08:00
Nathan Hjelm
219a7cde40 Merge pull request #1200 from hjelmn/var_group_fix
mca/base: remove erroneous check in var group register function
2015-12-09 22:18:56 -08:00
Ralph Castain
1db3db022a Don't be so prescriptive about the ess component to be used - we just need to protect against the proc incorrectly taking the singleton component, so rule that one out. Ensure that the other components understand that they are only for use by daemons. 2015-12-09 19:54:44 -08:00
Nathan Hjelm
772172a99b mca/base: remove erroneous check in var group register function
This commit removes a check that causes mca_base_group_register to
improperly create a new group instead of using an existing group
when the project and framework names are the same. This check was
originally intended to prevent forming groups with names like
ompi_ompi, opal_opal, etc but there is no reason why we shouldn't
allow that.

Fixes open-mpi/ompi#1155

Signed-off-by: Nathan Hjelm <hjelmn@me.com>
2015-12-09 19:48:39 -07:00
Nathan Hjelm
bf5b2bb74f Merge pull request #1199 from hjelmn/mlx5_attr_check
btl/openib: add check for IBV_EXP_QP_INIT_ATTR_ATOMICS_ARG
2015-12-09 18:35:42 -08:00
Nathan Hjelm
f692576f1e btl/openib: add check for IBV_EXP_QP_INIT_ATTR_ATOMICS_ARG
Mofed 2.2 does not have the IBV_EXP_QP_INIT_ATTR_ATOMICS_ARG attribute
flag. Add a check to fix compilation for mofed 2.2. This commit only
fixes complilation with the older mofed. It will not allow an Open MPI
compiled with mofed 2.3 or newer to work on a machine with mofed 2.2.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-12-09 17:02:36 -07:00
Ryan Grant
e5ea2e3248 Merge pull request #1193 from tkordenbrock/topic/fix.btl.logical.endpoint.rank
btl-portals4: set endpoint rank even if endpoint already exists

--Needs to be pulled over to 2.0.0 still @tkordenbrock
2015-12-09 13:49:44 -08:00
Nathan Hjelm
645d4099d4 Merge pull request #1197 from artpol84/comm_create_fix
Fix ompi_comm_create when source communicator is inter-communicator.
2015-12-09 09:18:24 -08:00
Artem Polyakov
ee71e35a90 Fix ompi_comm_create when source communicator is inter-communicator.
This bug was triggered by probe-intercom and icm tests from MPICH suite.
2015-12-09 15:44:26 +02:00
Gilles Gouaillardet
3a62341b30 Merge pull request #1189 from ggouaillardet/topic/empty_ddt_fix
ddt: duplicate MPI_DATATYPE_NULL when ompi_datatype_create_indexed of…
2015-12-09 15:29:03 +09:00
Jeff Squyres
00c5dc9449 rml oob: C99-ification of structure member assignment 2015-12-08 17:05:16 -08:00
Howard Pritchard
c2ea018ce5 Merge pull request #1194 from hppritcha/topic/fix_cray_pmix_locality
pmix/cray: fix locality bug
2015-12-08 16:25:30 -07:00
Howard Pritchard
fecb326256 pmix/cray: fix locality bug
There was a bug with the way the cray pmix component
was setting the locality property for ranks on the
same node, etc.

Improve location/syntax of a comment block.

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2015-12-08 11:13:48 -08:00
Todd Kordenbrock
2b7e983989 btl-portals4: set endpoint rank even if endpoint already exists
If btl-portals4 is configured to use logical mapping of ranks to
physical nodes, then the endpoint must have the rank field set.
This commit fixes a bug that caused the endpoint to have the
nid/pid instead of the rank if the endpoint already exists.
2015-12-08 12:29:00 -06:00
Nathan Hjelm
2ff16c6ba5 Merge pull request #1192 from hjelmn/mlx5_atomics_fix
mlx5: need to set comp_mask to get experimental verbs attributes
2015-12-08 11:20:37 -07:00
Nathan Hjelm
f317ba5262 Merge pull request #1163 from hjelmn/ompi_proc_threads
ompi/proc: make proc system always thread safe
2015-12-08 10:36:55 -07:00
Nathan Hjelm
c9382f23e9 mlx5: need to set comp_mask to get experimental verbs attributes
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-12-08 10:34:16 -07:00
Nathan Hjelm
b47a64f27d Merge pull request #1188 from artpol84/intercomm_split_fix
Yet one more fix to intercommunicator splitting logic.
2015-12-08 07:09:46 -07:00
Nathan Hjelm
dae3746d2f Merge pull request #1190 from kawashima-fj/pr/sm-win-test-fix
osc/sm: Fix a bug that `MPI_WIN_TEST` does not update `flag` to 0
2015-12-08 06:39:16 -07:00
KAWASHIMA Takahiro
9c7b6a4352 osc/sm: Fix a bug that MPI_WIN_TEST does not update flag to 0.
`MPI_WIN_TEST` must update the `flag` parameter to 0 when not all
origin processes called `MPI_WIN_COMPLETE`. But sm OSC doesn't.
If the caller initialize the `flag` argument to a non-0 value,
the caller will receive the non-0 `flag` value.
2015-12-08 19:23:21 +09:00
Gilles Gouaillardet
59a361b781 ompio: correctly handle zero f_cc_size in mca_io_ompio_simple_grouping 2015-12-08 17:00:11 +09:00
Gilles Gouaillardet
d43ad3fada ddt: duplicate MPI_DATATYPE_NULL when ompi_datatype_create_indexed of ompi_datatype_create_indexed_block is invoked with a zero count 2015-12-08 16:25:36 +09:00
Artem Polyakov
7690f4027a Yet one more fix to intercommunicator splitting logic.
Previous commit f2794740 reverts Nathans changes. However it turns out
that I was unable to trace his logic until I started investigation of
icsplit hang. Bug was triggered when splitting Intercom was giving a group
where on side of the communicator was empty (icsplit, intercom create #2).
in this case remote_size == 0 and there is no way to distinguish between
inter- and intra-communicator.
Conclusion: We do need to distinguish between intra- and inter-communicators.
So we should use ompi_mpi_group_null.group.
2015-12-08 08:43:08 +02:00
Nathan Hjelm
63d8feb31c Merge pull request #1187 from hjelmn/bsend_fix
pml/ob1: add missing ompi_request_wait_completion for buffered sends
2015-12-07 23:09:04 -07:00
Nathan Hjelm
f68c315188 pml/ob1: add missing ompi_request_wait_completion for buffered sends
This commit adds a call to ompi_request_wait_completion for buffered
sends. Without this line it is possible to get into a state where the
data is never sent.

Fixes open-mpi/ompi#1185

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-12-07 22:28:07 -07:00
Artem Polyakov
90b4148f9b Merge pull request #1184 from artpol84/intercomm_split_fix
Fix intercommunicator split (was triggered by MPICH/icsend test)
2015-12-08 08:46:38 +05:00
Artem Polyakov
f2794740b3 Fix intercommunicator split (was triggered by MPICH/icsend test) 2015-12-07 15:41:29 +02:00
rhc54
04fc4060db Merge pull request #1143 from rhc54/topic/noompi
Fix "autogen.pl --no-ompi"
2015-12-07 00:40:16 -08:00
Gilles Gouaillardet
4d2c7f7de1 cuda: fix missing #include opal/util/argv.h 2015-12-07 14:10:32 +09:00
Ralph Castain
e33515c87c Fix "autogen.pl --no-ompi", which was broken due to inclusion of a conditional in the test/monitoring Makefile.am that is only defined if OMPI is built.
Per suggestion from @bosilca, comment out generation of the shared library

Use the patch from Gilles instead
2015-12-06 21:10:17 -08:00
Gilles Gouaillardet
bfe8e03d9d fcoll/two_phase: use ompi_mpi_abort instead of PMPI_Abort
Thanks Jeff for the review
2015-12-07 11:34:36 +09:00
Gilles Gouaillardet
ef03bc726c ompi: fix comment in ompi/mpi/c/Makefile.am
Thanks Jeff for the review
2015-12-07 11:34:01 +09:00
Gilles Gouaillardet
37c978f5e9 coll/libnbc: correctly handle changed types.
this fixes open-mpi/ompi@d816d1c194
thanks Jeff for the review
2015-12-07 10:13:43 +09:00
Howard Pritchard
d7b437ecaa Merge pull request #1157 from gpichot/adds-darwin-headers-for-java
Adds darwin headers directory for Darwin JDK
2015-12-06 14:49:53 -07:00
George Bosilca
3a9664ac9d Fix Coverity CIDs 1341584-1341589. 2015-12-06 14:06:36 -05:00
Gabriel Pichot
ff5af73676 Adds usage of /usr/libexec/java_home for OS X platforms 2015-12-05 11:51:51 +01:00
Ralph Castain
10db7ebfab Update the symbol-hiding script to capture a broader range of symbols 2015-12-04 21:05:57 -08:00
rhc54
1568d6b33c Merge pull request #1181 from rhc54/topic/tools
Provide a mechanism by which a tool can request async progress thread support for ORTE
2015-12-04 12:24:05 -08:00
Ralph Castain
8823069fe9 Provide a mechanism by which a tool can request async progress thread support for ORTE 2015-12-04 08:26:57 -08:00
Jeff Squyres
ad35a363fa Merge pull request #1179 from jsquyres/pr/mpi-testsome-man-page-update
Pr/mpi testsome man page update
2015-12-04 05:55:33 -05:00
bosilca
8fee96c086 Merge pull request #1091 from bosilca/topic/datatype_span
Cleanup the temporary memory allocation in collectives
2015-12-03 19:25:04 -05:00
Jeff Squyres
0adcb5b5cd MPI_Testsome.3in: wrap some long lines
Wrap some long lines; no other text or semantics changes.
2015-12-03 17:06:43 -05:00
Jeff Squyres
11c571b568 MPI_Testsome.3in: add explicit verbiage about return values
Instead of solely relying on the out value definitions in
MPI_Waitsome.3, explicitly copy this text here.

Note that the original text in this man page was copied verbatim from
the MPI spec; we've now added a bit more text (copied from
MPI_Waitsome.3in) that explains the out values so that users don't
have to cross-reference to another man page.

Thanks to Eric Schnetter for the suggestion.

Fixes open-mpi/ompi#1153
2015-12-03 17:06:22 -05:00
Mike Dubman
dc3b138b7f Merge pull request #1178 from igor-ivanov/pr/oshmem-fix-preconnect_all
oshmem: Add skew to the oshmem_preconnect_all
2015-12-03 21:00:52 +02:00
Jeff Squyres
a25cf8065d Merge pull request #1176 from igor-ivanov/pr/oshmem-rm-java
oshmem: Remove JAVA support
2015-12-03 13:26:09 -05:00
Igor Ivanov
c4fc1ec081 oshmem: Add skew to the oshmem_preconnect_all 2015-12-03 18:32:26 +02:00
Igor Ivanov
63bc655b53 oshmem: Remove JAVA support
There is a conclusion to remove java support
(see https://github.com/open-mpi/ompi-release/pull/787)
2015-12-03 11:59:07 +02:00
Gilles Gouaillardet
a5440ade5f topo/treematch: do not invoke hwloc_topology_{init,load}
* this is not necessary
 * this overwrites existing topology, that could be different if hwloc_base_topo_file is used
2015-12-03 11:24:32 +09:00
George Bosilca
688108cf7f Patch submitted by @ggouaillardet on ticket #1091. 2015-12-02 20:42:18 -05:00