1
1
Граф коммитов

26834 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
d645557fa0 Update to include the PMIx 2.0 APIs for monitoring and job control. Include required integration, but leave the monitors off for now. Move the sensor framework out of ORTE as it is being absorbed into PMIx
Fix typo and silence warnings

Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-03-21 17:47:08 -07:00
Jeff Squyres
20bf0dd7c6 Merge pull request #3214 from hppritcha/travis_remove_osx
travis: remove os-x from OS test matrix
2017-03-21 13:48:28 -04:00
Howard Pritchard
f06a5d93ea travis: remove os-x from OS test matrix
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-03-21 10:36:30 -06:00
Joshua Ladd
f8d7324b4f Merge pull request #3168 from xinzhao3/topic/ucx-mt-support
Add multithreading support in PML UCX framework.
2017-03-20 15:50:50 -04:00
Xin Zhao
6a99c60fbd Add multithreading support in PML UCX framework.
Signed-off-by: Xin Zhao <xinz@mellanox.com>
2017-03-20 19:55:00 +02:00
Ralph Castain
40a0144e76 Merge pull request #3208 from rhc54/topic/s2
You cannot include both pmi.h and pmi2.h as they have conflicting defines in them.
2017-03-19 14:21:04 -07:00
Ralph Castain
4b6d220a83 You cannot include both pmi.h and pmi2.h as they have conflicting defines in them.
Thanks to Kilian Cavalotti for pointing it out

Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-03-19 11:53:54 -07:00
Jeff Squyres
ce0e1cd32c Merge pull request #3201 from hppritcha/jjhursey-topic/timer-gettimeofday
Jjhursey topic/timer gettimeofday
2017-03-18 20:12:36 -04:00
Howard Pritchard
b9331527f5 timer: hack use of clock_gettime
better solution needed later
 workaround for #3003

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-03-18 15:08:59 -05:00
Jeff Squyres
19b2454d9d Merge pull request #3198 from jsquyres/pr/master/fix-hwloc-dist-autogen
hwloc: re-enable use of autogen.pl in a tarball
2017-03-17 14:44:54 -04:00
Jeff Squyres
b8dfd49e97 hwloc: re-enable use of autogen.pl in a tarball
Commit fec519a793 broke the ability to
run autogen.pl in a distribution tarball.  This commit restores that
ability by also distributing opal/mca/hwloc/autogen.options in the
tarball.

Skipping CI because CI does not test this functionality:

[skip ci]
bot:notest

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-03-17 11:41:17 -07:00
Pavel Shamis / Pasha
78cd5de5d6 Merge pull request #3191 from shamisp/topic/oshmem_wait_cleanup
OSHMEM: shmem_wait code cleanup
2017-03-17 13:35:10 -05:00
Ralph Castain
afcc33862e Merge pull request #3197 from rhc54/topic/errors
Provide a little more help on the error messages when an executable i…
2017-03-17 11:29:39 -07:00
Ralph Castain
dc85e7fde7 Provide a little more help on the error messages when an executable isn't found so we have some better idea where we were looking for it. Don't double-report such errors. Ensure the ORTE_ERROR_NAME doesn't get a NULL back for the string name of an error code as that might cause some systems to segfault
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-03-17 09:54:37 -07:00
Ralph Castain
45b46dc446 Merge pull request #3181 from artpol84/add_proc_fix_2/master
ompi: Avoid unnecessary PMIx lookups when adding procs (step 2).
2017-03-16 15:06:08 -07:00
Jeff Squyres
5219054d29 Merge pull request #3185 from jsquyres/pr/master/compiler-warning-squashes
Compiler warning squashes
2017-03-16 10:12:08 -04:00
Howard Pritchard
65d0372e84 Merge pull request #3179 from hppritcha/topic/remove_osx_builtin_atomics
OSx: remove built-in atomics support
2017-03-16 07:58:41 -06:00
Jeff Squyres
760db0d5ce osc/pt2pt: fix compiler warning
Remove unused variable.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-03-16 05:46:11 -07:00
Jeff Squyres
1947280865 topo/treematch: squash some compiler warnings
Only define MIN/MAX if they are not already defined.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-03-16 05:44:26 -07:00
Jeff Squyres
b51c4e2797 memory/patcher: fix a compiler warning
Don't define the madvise intercept functions since we're not currently
intercepting madvise.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-03-16 05:43:51 -07:00
Joshua Hursey
48d13aa8ef mpi/c: Force wtick/wtime to use gettimeofday
* See https://github.com/open-mpi/ompi/issues/3003 for a discussion about
   this patch. Once we get a better version in place we can revert this
   change.

Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
2017-03-15 21:24:37 -05:00
Jeff Squyres
616f20c52c timer/linux: rename component-specific functions
Several component-specific functions were named with a prefix of
"opal_timer_base", which was quite confusing.  Rename them to have a
prefix "opal_timer_linux" to make it clear that they are here in this
component (and different than *actual* opal_timer_base symbols).

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-03-15 21:03:13 -05:00
Jeff Squyres
290d4598df timer/linux: remove global variable
This variable is only used in one file, so make it static.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-03-15 21:03:06 -05:00
Artem Polyakov
1f7a3a2d54 ompi: Avoid unnecessary PMIx lookups when adding procs (step 2).
Follow-up for 717f3fef62.

Signed-off-by: Artem Polyakov <artpol84@gmail.com>
2017-03-16 07:47:27 +07:00
Pavel Shamis (Pasha)
95c440683b OSHMEM: shmem_wait code cleanup
* updating naming convention for the arguments in order to ensure
that the name aligns with an actual meaning of the argument
* remove local variable references in the macro
* adding volatile for the poll variables

Signed-off-by: Pavel Shamis (Pasha) <pasharesearch@gmail.com>
2017-03-15 21:53:44 +00:00
Howard Pritchard
db2e1298fb OSx: remove built-in atomics support
It was decided to remove support for os-x builtin atomics

Fixes #2668

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-03-15 12:45:33 -06:00
Jeff Squyres
60ca372d60 NEWS: Sync with v2.0.x and v1.10 releases
Pull in content from v1.10 and v2.0.x branches.

[skip ci]
bot:notest

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-03-15 09:56:31 -07:00
Howard Pritchard
1709febdea Merge pull request #3166 from hppritcha/topic/swat_state_orted_comp_warning
ORTED: swat another compiler warning
2017-03-15 08:40:59 -06:00
Howard Pritchard
1f7378d7e4 Merge pull request #3151 from hppritcha/topic/update_license_file
LICENSE: update according to copyrights in source files
2017-03-15 07:58:27 -06:00
Howard Pritchard
8e4689c2b8 v3.x:updates for branching v3.x
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-03-14 14:03:47 -06:00
Howard Pritchard
c2da14d514 AUTHORS: update for 3.0.0 branching
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-03-14 13:54:56 -06:00
Ralph Castain
96d7d10c1d Merge pull request #3170 from rhc54/topic/reg
Ensure the backend daemons know if we are in a managed allocation and if the HNP was included in the allocation
2017-03-14 12:48:09 -07:00
Nathan Hjelm
db9232f8d6 Merge pull request #3169 from hjelmn/btl_ugni_2_0
More btl/ugni updates
2017-03-14 13:23:13 -06:00
Nathan Hjelm
37214eda09 Merge pull request #3164 from hjelmn/ob1_pinned
pml/ob1: do not cache leave_pinned
2017-03-14 13:22:18 -06:00
Mike Dubman
ccac7e5363 Merge pull request #3157 from vspetrov/c_coll_allgather_usage_bugfix
Fixes the coll_allgather usage bug
2017-03-14 18:42:04 +01:00
Ralph Castain
24e8639826 Platform file update
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-03-14 11:11:48 -06:00
Ralph Castain
955fa0456d Merge pull request #3161 from rhc54/topic/cov2
Silence Coverity warnings
2017-03-14 10:10:11 -07:00
Ralph Castain
61a71e25ef Ensure the backend daemons know if we are in a managed allocation and if
the HNP was included in the allocation

Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-03-14 10:06:43 -07:00
Nathan Hjelm
6b210fa2c4 btl/ugni: do not return a frag from sendi if an endpoint is waitlisted
This fixes a hang that can occur when running bandwidth tests.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-03-14 10:14:13 -06:00
Nathan Hjelm
2e42b0afbd btl/ugni: move connection check into sync event
This commit makes datagram checks time based and reduces their
frequency when only the wildcard datagram is posted. This change
improves latency on knl systems.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-03-14 10:10:05 -06:00
Nathan Hjelm
3e7ef48c13 pml/ob1: do not cache leave_pinned
This commit fixes a bug that disabled both the RDMA pipeline and RDMA
protocols in ob1. ob1 was internally caching the values of
opal_leave_pinned and opal_leave_pinned_pipeline at init time. This is
no longer valid as opal_leave_pinned may be set by any call to a btl's
add_procs.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-03-14 09:00:40 -06:00
Howard Pritchard
5daaf7f3fd ORTED: swat another compiler warning
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-03-14 08:41:51 -06:00
Ralph Castain
52c9e631de Silence Coverity warnings
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-03-14 07:30:42 -07:00
Valentin Petrov
fe069c9570 Fixes the coll_allgather usage bug
One should use the correct module object when calling
      c_coll.coll_allgather. Otherwise there will be a segfault in the
      case, for example, when hcoll is used. In that case
      c_coll.coll_allgather = mca_coll_hcoll_allgather while
      c_coll.coll_gather_module = tuned.

Signed-off-by: Valentin Petrov <valentinp@mellanox.com>
2017-03-14 09:47:39 +02:00
Ralph Castain
330b11c8ab Merge pull request #3156 from rhc54/topic/tm
Update the TM module to support regex passing
2017-03-14 00:25:19 -07:00
Ralph Castain
b1a01d77ae Update the TM module to support regex passing
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-03-13 21:50:40 -07:00
Nathan Hjelm
9410574253 Merge pull request #3149 from hjelmn/btl_ugni_2_0
Improve multi-threaded RMA performance of the ugni btl
2017-03-13 16:28:41 -06:00
Ralph Castain
e4a35f2dbf Merge pull request #3152 from rhc54/topic/setup
Update launchers to get correct regex
2017-03-13 14:23:43 -07:00
Nathan Hjelm
d5aaeb74b6 btl/ugni: return a descriptor from sendi
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-03-13 14:56:54 -06:00
Nathan Hjelm
a19e7023d1 btl/ugni: always check local SMSG CQ
This commit removes the local operation count check from the local SMSG
completion queue. This check was leading to hangs due to an undocumented
feature of the ugni library. The local SMSG CQ is used to send credit
return messages back to the sender. The ugni library never checks for
the completion itself but relying on the SMSG user to periodically
check the CQ.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-03-13 14:56:54 -06:00