Ralph Castain
a239b4c3c3
Per discussion on the PMIx side, do a better job of detecting mismatches between location directives for OPAL and PMIx. Provide a more helpful error message and error out if we find a mismatch. If any OPAL values are set and the PMIx equivalent is not, then transfer it.
...
Do not clear PMIX_INSTALL_PREFIX from the daemon's launch environment
Fixes #3980
Closes #4007
Refs #3985
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-08-04 19:36:00 -07:00
Ralph Castain
f128b4c546
Fix incorrect usage of '==' in test comparisons
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-08-03 21:21:26 -07:00
Howard Pritchard
5ce07a6983
Merge pull request #3997 from hppritcha/topic/swat_compiler_warning
...
btl/ugni: swat compiler warning
2017-08-02 15:44:09 -06:00
Artem Polyakov
500c8be888
pmix: fix PMIx envar name for the installation prefix.
...
Signed-off-by: Artem Polyakov <artpol84@gmail.com>
2017-08-02 08:03:36 +03:00
Ralph Castain
f39ce67982
Merge pull request #3951 from rhc54/topic/hwloc2
...
Update to hwloc 2.0.0a
2017-08-01 15:18:31 -06:00
Ralph Castain
69612b3e2a
Merge pull request #3990 from rhc54/topic/p2
...
Move handling of OPAL_PREFIX to PMIX_PREFIX down into embedded PMIx integration code
2017-08-01 15:13:59 -06:00
Brian Barrett
c4ae36f971
Merge pull request #3869 from Zzzoom/find_freq_bogomips
...
opal: Get x86 TSC frequency from bogomips
2017-08-01 13:23:21 -07:00
Howard Pritchard
12a5aacdfd
btl/ugni: swat compiler warning
...
Signed-off-by: Howard Pritchard <hppritcha@gmail.com>
2017-08-01 12:21:57 -06:00
Ralph Castain
8f34fa4a56
Move the detection of OPAL_PREFIX and subsequent posting of PMIX_PREFIX to the internal integration code for PMIx so we only do this when running with the embeddied PMIx
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-08-01 08:24:27 -06:00
Sylvain Jeaugey
eee494fc8a
common/cuda: Fix near-hang when remote side has exited
...
Ignore errors caused by remote side having exited when closing CUDA IPC mappings.
openmpi/ompi#3244
Signed-off-by: Sylvain Jeaugey <sjeaugey@nvidia.com>
2017-07-31 10:34:45 -07:00
Boris Karasev
e20b581529
pmix: fixed immediate request
...
This commit fixes a hang when using external PMIx v1 module
Signed-off-by: Boris Karasev <karasev.b@gmail.com>
2017-07-28 15:53:48 +06:00
Gilles Gouaillardet
825116044e
hwloc/base: fix info message for opal_hwloc_base_binding_policy
...
if np > 2, the default binding is now "numa"
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-28 11:17:15 +09:00
Jeff Squyres
d954167ecf
Merge pull request #3881 from bharatpotnuri/master
...
master: btl/openib: Handle EOPNOTSUPP
2017-07-26 11:32:40 -04:00
Ralph Castain
6ebaed8c01
Restore support for user-provided cpulist
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-25 23:51:21 -07:00
Ralph Castain
7a83fdb9bb
Update to hwloc 2.0.0a with shmem support.
...
Update to support passing of HWLOC shmem topology to client procs
Update use of distance API per @bgoglin
Have the openib component lookup its object in the distance matrix
Bring usnic up-to-date
Restore binding for hwloc2
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-25 20:26:22 -07:00
Ralph Castain
6fe5b36b50
Merge pull request #3963 from rhc54/topic/hwfix
...
Restore binding support
2017-07-25 22:09:04 -05:00
Ralph Castain
96f07aebfa
Restore binding support
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-25 18:44:44 -07:00
Ralph Castain
0042c758f1
Update the tools support so it allows tools to access PMIx
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-25 17:10:08 -07:00
Ralph Castain
058e802b11
Add missing export directives
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-25 07:19:08 -07:00
George Bosilca
1ea8fab095
Make external symbols visible.
...
All symbols that need to be accessed from a MCA component must be marked
explicitly as visible using PMIX_EXPORT. This patch allows current trunk
to almost work on OsX. More on the devel mailing list.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
2017-07-25 01:14:22 -04:00
Ralph Castain
af85e48dd7
Silence Coverity warning, silence pmix_error_log of success
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-21 15:33:16 -07:00
Ralph Castain
492f98f8a5
Update to latest PMIx v2.1.0a
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-21 12:58:09 -07:00
Ralph Castain
f7e8780a42
Remove fortran support from platform file
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-20 21:02:30 -07:00
Ralph Castain
b225366012
Bring the ofi/rml component online by completing the wireup protocol for the daemons. Cleanup the current confusion over how connection info gets created and
...
passed to make it all flow thru the opal/pmix "put/get" operations. Update the PMIx code to latest master to pickup some required behaviors.
Remove the no-longer-required get_contact_info and set_contact_info from the RML layer.
Add an MCA param to allow the ofi/rml component to route messages if desired. This is mainly for experimentation at this point as we aren't sure if routing wi
ll be beneficial at large scales. Leave it "off" by default.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-20 21:01:57 -07:00
Ralph Castain
0e4e3af1db
Remove problem installation of hwloc 2.0
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-20 18:18:08 -07:00
Ralph Castain
7d8d877837
Remove build product and update .gitignore to avoid picking it up again
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-20 11:49:48 -07:00
Ralph Castain
8c30958879
Update to PMIx v2.1.0alpha
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-20 11:12:06 -07:00
Gilles Gouaillardet
593e4ce63f
hwloc: add hwloc2x
...
internal hwloc 2x is used with --with-hwloc=future
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-20 17:39:51 +09:00
Gilles Gouaillardet
60aa9cfcb6
hwloc: add support for hwloc v2 API
...
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-20 17:39:44 +09:00
Gilles Gouaillardet
9f29f3bff4
hwloc: since WHOLE_SYSTEM is no more used, remove useless
...
checks related to offline and disallowed elements
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-20 17:39:21 +09:00
Gilles Gouaillardet
1a34224948
hwloc: do not set the HWLOC_TOPOLOGY_FLAG_WHOLE_SYSTEM flag
...
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-20 17:39:16 +09:00
Ralph Castain
fca68b070b
Merge pull request #3934 from rhc54/topic/singleton
...
Fix the isolated pmix component. Cleanup the ess/singleton component …
2017-07-19 16:02:37 -05:00
Ralph Castain
543c16b28d
Fix the isolated pmix component. Cleanup the ess/singleton component - we shouldn't be automatically discovering the local topology as that is now done on-demand.
...
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-07-19 12:14:29 -07:00
Howard Pritchard
2fa0c4c6ec
pmix/s1: fix problems with ref counting in s1
...
s1 pmix component wasn't doing proper ref counting
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-07-18 15:59:28 -06:00
Josh Hursey
8688219091
Merge pull request #3775 from jjhursey/fix/mca_base_verbose-file
...
opal/mca: Fix mca_base_verbose file suffix processing
2017-07-18 10:14:42 -05:00
Howard Pritchard
771f51af12
Merge pull request #3917 from hppritcha/topic/remove_cr_config_master
...
configure: remove CR/FT related options
2017-07-17 16:12:07 -06:00
Nathan Hjelm
2060fcf8bb
mca/base: use the project name when registering pvars
...
References #3918 . Close when applied to v2.0.x, v2.x, and v3.0.x.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-07-17 15:57:50 -05:00
Howard Pritchard
45e2771162
configure: remove CR/FT related options
...
As part of the process for addressing removal of CR/FT related
code from master (and hence from the 3.0.0 release), it was agreed
at the OMPI devel F2F on 7/13/17 that we'd break this in to two
pieces:
1) remove the configure arguments (fewer changes)
2) remove all the CR/FT code, etc. in a subsequent bigger commit
that may not make it in to 3.0.0 in time.
By doing 1), the available configure options would not change
in a subsequent 3.0.x release if we end up not being able to do 2)
before 3.0.0 is released.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-07-17 13:48:59 -06:00
Nathan Hjelm
e5343c16c0
btl/vader: remove debug code that should not be in a release
...
References #3902 . Close when in master, v3.0.x, and v2.x.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-07-17 11:58:47 -05:00
Gilles Gouaillardet
6e35cfc19a
btl/sm: fix misc memory leak
...
as reported by Coverity with CID 1415105
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-16 13:02:55 +09:00
Jeff Squyres
5cf64e6555
btl/sm: effectively delete the SM BTL
...
If a user explicitly asks for the "sm" BTL, print a show_help message
saying that the SM BTL is dead, and the user should be using "vader".
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-07-15 09:33:08 -07:00
Artem Polyakov
0929c32cd8
Merge pull request #3893 from karasevb/yoda_spml_remove
...
Remove Yoda SPML
2017-07-15 08:47:31 -07:00
Gilles Gouaillardet
9124afbeae
pmix: do not invoke PMIX_INFO_CREATE() with a zero size
...
Thanks Lisandro Dalcin for the report
Fixes open-mpi/ompi#3854
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-14 15:00:05 +09:00
Boris Karasev
77c50efb95
Yoda SPML is removed
...
Signed-off-by: Boris Karasev <karasev.b@gmail.com>
2017-07-14 08:47:16 +03:00
Artem Polyakov
4d3e22e815
Merge pull request #3870 from hppritcha/topic/repair_s2_launch
...
pmix/s2: fix srun native launch for pmi2
2017-07-13 12:45:22 -05:00
Potnuri Bharat Teja
9154ade8b1
btl/openib: Handle EOPNOTSUPP
...
Updated openib BTL to handle EOPNOTSUPP as per
https://www.open-mpi.org/community/lists/devel/2016/04/18839.php
Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com>
2017-07-13 21:05:32 +05:30
Howard Pritchard
eeb91bc82b
pmix/s2: fix srun native launch for pmi2
...
recent changes that broke native launch on cray
using srun or aprun was also broke native launch
using pmi2.
This commit fixes this problem.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2017-07-12 17:45:52 -06:00
Jeff Squyres
ccf17808b6
Merge pull request #3258 from markalle/pr/symbol_name_pollution
...
symbol name pollution
2017-07-12 16:19:25 -05:00
Carlos Bederián
b5883a358b
Get x86 TSC frequency from bogomips
...
Signed-off-by: Carlos Bederián <bc@famaf.unc.edu.ar>
2017-07-12 17:31:25 -03:00
Gilles Gouaillardet
32606ad476
btl/tcp: fix heterogeneous support for put / large messages
...
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-07-12 10:27:45 +09:00