ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
26 045 коммитов 3 Ветки 0 Теги
Граф коммитов

26045 Коммитов

This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
Josh Hursey
b18598f6c7 Merge pull request #2329 from jjhursey/topic/short-hostname-lsf-fix 
ras/*: Fix !orte_keep_fqdn_hostnames for RAS components
 2016-11-02 10:49:08 -05:00
Jeff Squyres
a4ffa590c8 Merge pull request #2308 from hjelmn/vader_mem 
btl/vader: reduce memory footprint when using xpmem
 2016-11-02 10:28:26 -04:00
Jeff Squyres
27fcd2d6ba Merge pull request #2336 from larrystevenwise/master 
openib btl: remove BTL_OPENIB_FAILOVER_ENABLED code
 2016-11-02 10:27:26 -04:00
Jeff Squyres
479fe59d9e Merge pull request #2343 from karasevb/fix_spml_yoda 
oshmem/spml/yoda: fixed the btl operations
 2016-11-02 10:25:47 -04:00
Boris Karasev
68b5acd9f4 oshmem/spml/yoda: fixed the btl operations 
Fixed the shmem OOM error which is referenced on #2028

Signed-off-by: Boris Karasev <karasev.b@gmail.com>
 2016-11-02 13:38:35 +02:00
rhc54
4c9f7af4cb Merge pull request #2340 from rhc54/topic/dvm 
Fix the radix routed component to correctly handle connected tools - …
 2016-11-01 20:28:02 -07:00
Ralph Castain
435d771e76 Fix the radix routed component to correctly handle connected tools - in such cases, the route must be direct to the tool. 
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
 2016-11-01 19:03:26 -07:00
Jeff Squyres
5e7d8636f6 Merge pull request #2338 from jsquyres/pr/rename-nightly-tarball 
create_tarball.sh: update snapshot filename
 2016-11-01 20:11:32 -04:00
Jeff Squyres
78d1e4ebff create_tarball.sh: update snapshot filename 
Nightly snapshots will now be named:

openmpi-${BRANCHNAME}-${YYYYMMDDHHMM}-${SHORTHASH}.tar.${COMPRESSION}.

Fixes #2337

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
 2016-11-01 17:09:17 -07:00
Steve Wise
7050969d47 openib btl: remove BTL_OPENIB_FAILOVER_ENABLED code 
Remove BTL_OPENIB_FAILOVER_ENABLED code in the openib btl source.

Remove the failover-specific files from the openib btl.

Update the openib/Makefile.am accordingly.

Remove the -enable-openib-failover config logic.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
 2016-11-01 14:45:36 -07:00
rhc54
0cd200da80 Merge pull request #2332 from rhc54/topic/routed2 
Remove the max_connections parameter from the radix component as it i…
 2016-11-01 13:49:51 -07:00
Ralph Castain
64873487b4 Remove the max_connections parameter from the radix component as it is confusing. Modify PMIx client init so that it simply returns the nspace/rank if called by a server - this allows the server to retrieve its assigned ID. Register the server's nspace so client-side operations can succeed 
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
 2016-11-01 12:17:11 -07:00
Joshua Hursey
ed5268a96a ras/slurm: Fix !orte_keep_fqdn_hostnames for Slurm 
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
 2016-11-01 13:21:30 -05:00
Joshua Hursey
5a4c52d9cb ras/loadleveler: Fix !orte_keep_fqdn_hostnames for Loadleveler 
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
 2016-11-01 13:21:30 -05:00
Joshua Hursey
8230201ad1 ras/gridengine: Fix !orte_keep_fqdn_hostnames for GridEngine 
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
 2016-11-01 13:21:30 -05:00
Joshua Hursey
9643175e40 ras/tm: Fix !orte_keep_fqdn_hostnames for TORQUE 
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
 2016-11-01 13:21:24 -05:00
Mike Dubman
0b14928b45 Merge pull request #2327 from yosefe/topic/pml-ucx-status-cancelled 
pml_ucx: fix uninitialized field req_status->_cancelled.
 2016-11-01 20:06:25 +02:00
Joshua Hursey
8d02a33639 ras/lsf: Fix !orte_keep_fqdn_hostnames for LSF 
* By default, make sure that we are using the short hostnames and not
   the fully qualified hostnames when running under LSF.
 * Related to commit open-mpi/ompi@d26dd2c20e

Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
 2016-11-01 13:04:52 -05:00
Yossi Itigin
17c8f76411 pml_ucx: fix uninitialized field req_status->_cancelled. 
Signed-off-by: Yossi Itigin <yosefe@mellanox.com>
 2016-11-01 17:02:22 +02:00
rhc54
6074c2a2a9 Merge pull request #2322 from rhc54/topic/routed 
Update the routed components as we no longer need to init_routes.
 2016-10-31 13:37:07 -07:00
Ralph Castain
b8c5d1ad88 Update the routed components as we no longer need to init_routes. Fixes case of direct launch via srun 
Signed-off-by: Ralph Castain <rhc@open-mpi.org>

Signed-off-by: Ralph Castain <rhc@open-mpi.org>
 2016-10-31 12:38:13 -07:00
Jeff Squyres
773d6039e7 Merge pull request #2306 from hjelmn/alps_cores 
ras/alps: use cpuCnt if using hwthreads as cores
 2016-10-31 15:22:13 -04:00
Gilles Gouaillardet
30298cc83c oob/tcp: remove debug that should have never been commited 
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-31 16:41:14 +09:00
Gilles Gouaillardet
75e96004a4 oob/tcp: fix a typo in mca_oob_tcp_component_no_route() 
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-31 16:30:24 +09:00
Jeff Squyres
25c3edb6d8 Merge pull request #2312 from jsquyres/pr/usnic-warning-fix 
btl/usnic: fix compiler warning
 2016-10-28 12:08:13 -04:00
Joshua Ladd
d27b680de2 Merge pull request #2305 from vspetrov/hcoll_fortran_pair_types 
coll/hcoll fortran pair types
 2016-10-28 12:05:00 -04:00
Jeff Squyres
149b660666 btl/usnic: fix compiler warning 
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
 2016-10-28 07:36:20 -07:00
Gilles Gouaillardet
52a1f96726 fortran/use-mpi-tkr: update .gitignore 
ignore automatically generated mpi-tkr-sizeof.*

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-28 10:41:59 +09:00
Gilles Gouaillardet
efac15e9a1 ompi: use opal_setenv instead of putenv 
this fixes a memory leak at finalize

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-28 09:32:30 +09:00
Gilles Gouaillardet
981dccab8d ompi: cleanup environment at finalize 
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-28 09:32:30 +09:00
Gilles Gouaillardet
fb5bcc47ce ess/singleton: use opal_setenv instead of putenv 
so it fixes a memory leak on finalize

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-28 09:32:30 +09:00
Gilles Gouaillardet
af67183e2f pml/v: fix a memory leak 
close the framework if no more component should be used

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-28 09:32:30 +09:00
Gilles Gouaillardet
ef2b3ac8d2 rml/oob: fix misc memory leaks in open_conduit() 
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-28 09:28:42 +09:00
Gilles Gouaillardet
831f7d9c9d rml/base: plug misc memory leaks 
plug leaks in orte_rml_API_get_contact_info() and orte_rml_base_close()

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-28 09:28:05 +09:00
rhc54
698dac108b Merge pull request #2309 from rhc54/topic/pmix 
Update to latest PMIx master - mostly updates example codes, but incl…
 2016-10-27 14:21:46 -07:00
Ralph Castain
f4a55118e6 Update to latest PMIx master - mostly updates example codes, but includes one critical cleanup during finalize. 
NOTE: set dstore shared memory storage option to "on" by default

Signed-off-by: Ralph Castain <rhc@open-mpi.org>
 2016-10-27 11:16:03 -07:00
Jeff Squyres
e9aab634af Merge pull request #2294 from ggouaillardet/topic/fortran_use_mpi_tkr 
fortran/use-mpi-tkr misc fixes
 2016-10-27 13:09:06 -04:00
Nathan Hjelm
9d92075e60 btl/self: rewrite to decrease memory usage (#2307) 
This commit rewrites much of the btl/self component to fix a long
standing memory usage bug. Before this commit the prepare_src path
would always allocate a max send fragment (256kB). This caused the
rank to allocate 32 * 256k useless buffers from one send. This commit
makes the following changes:

 - Add the MCA_BTL_FLAGS_GET flag by default. No reason not to set it.

 - Reduce the eager limit, max send size, buffers per allocation, and
   maximum buffer count per fragment size. These changes should have
   no noticible affect on performance but should greatly reduce the
   memory usage of the component.

 - Implement the sendi function. This should reduce self send latency
   somewhat.

 - Rewrite prepare_src to never allocate a eager or max send fragment
   for contiguous data.

 - add_procs needs to return something in the peer array for the proc
   self not just set the reachability bit. Now stores (void *) 1.

 - Various cleanups. Removed and unused file.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2016-10-27 12:34:54 -04:00
Nathan Hjelm
a652a193ea btl/vader: reduce memory footprint when using xpmem 
The vader btl kept a per-peer registration cache to keep track of
attachments. This is not really a problem with small numbers of local
ranks but can be a problem with large SMP machines. To reduce the
footprint there is now one registration cache for all xpmem
attachments. This will probably increase the lookup time for large
transfers but is a worthwhile trade-off.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2016-10-27 10:09:43 -06:00
Nathan Hjelm
c3614d30fa ras/alps: use cpuCnt if using hwthreads as cores 
This commit updates the alps ras component to allow the use of
hyperthreads on compute nodes. In this case we need to use the cpuCnt
value from the node structure instead of numPEs.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2016-10-27 09:51:17 -06:00
Valentin Petrov
2b7e362e56 coll/hcoll fortran pair types 
Adds mapping of the MPI Fortran pair types (2INTEGER, 2REAL, 2DBLPREC)
    to the corresponding hcoll dtypes.

Signed-off-by: Valentin Petrov <valentinp@mellanox.com>
 2016-10-27 18:24:07 +03:00
Mike Dubman
83e3323646 Merge pull request #2299 from alex-mikheev/topic/oshmem_init_fix 
OSHMEM: fixes crash during initialization
 2016-10-27 14:48:33 +03:00
Gilles Gouaillardet
3d4285b04d oob/tcp: silence valgrind warning 
fully initialize allocated memory to keep valgrind happy

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-27 17:12:46 +09:00
Alex Mikheev
511dd43736
oshmem: fixes typo in the error message 
Signed-off-by: Alex Mikheev <alexm@mellanox.com>
 2016-10-27 09:27:45 +03:00
Gilles Gouaillardet
ad7f3f93b0 mpi: support MPI_Dims_create(..., ndims=0, ...) 
this is a bozo case, but it should not fail with MPI_ERR_DIMS

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-27 14:13:51 +09:00
Gilles Gouaillardet
bf789762a7 fortran/use-mpi-tkr: fix MPI_Sizeof handling 
MPI_Sizeof related stuff has been moved to their own files.
Remove MPI_Sizeof from Fortran interfaces when it cannot be built
(e.g. stock gcc 4.8 on CentOS 7)

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-27 08:54:28 +09:00
Gilles Gouaillardet
1a16e68c26 fortran/use-mpi-trk: add PMPI_* interfaces in mpi module 
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-27 08:54:07 +09:00
Gilles Gouaillardet
8e26e78728 fortran/use-mpi-tkr: only build MPI_File support if requested 
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-27 08:54:07 +09:00
Gilles Gouaillardet
5543b19e9a fortran/use-mpi-tkr: rename mpi-f90-cptr-interfaces.F90 into mpi-f90-cptr-interfaces.h 
this file is meant to be included and not compiled, so use a consistent naming

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
 2016-10-27 08:54:07 +09:00
rhc54
2b18044051 Merge pull request #2301 from rhc54/topic/update 
Update PMIx to latest master tarball. Ensure we set the HNP name for …
 2016-10-26 16:42:15 -07:00