ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
24 602 коммитов 3 Ветки 0 Теги 130 MiB
f6e932c864
Граф коммитов

5054 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
4dad5de8ff Silence a couple of warnings - strncpy returns a char*, not an int 2016-01-16 09:44:52 -08:00
Jeff Squyres
60ffe713b8 common syms: whitelist bison-generated common symbols 
Bison generates some common symbols that we can't do anything about,
so whitelist them.
 2016-01-16 03:53:14 -08:00
Gilles Gouaillardet
1d38430e43 opal: replace opal_convert_jobid_to_string with opal_snprintf_jobid 2016-01-14 10:39:03 +09:00
Gilles Gouaillardet
4c43fb2a50 orte_rmaps_base_map_job: set OPAL_BIND_ALLOW_OVERLOAD when needed 2016-01-13 17:13:36 +09:00
Ralph Castain
332019b43a Silence warning 2016-01-10 09:59:36 -08:00
Nathan Hjelm
fab1eca536 grpcomm: fix bugs in grpcomm algorithms 
This commit fixes multiple issues in the bruck's and recursive
doubling grpcomm algorithms. The following changes are included:

 - Use the existing bitmap implementation instead of implementing a
   new one. There were bugs in the implementation that caused an
   overrun of the bitmap array.

 - Clean up the algorithms to eliminate errors.

 - Send as little extra data as possible in the bruck's
   algorithm.

The changes were testest with various numbers of ortes varying from 1
to 4096.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2016-01-07 10:12:08 -07:00
Ralph Castain
f53d3c7a18 Silence warning 2015-12-30 10:16:58 -08:00
Ralph Castain
0a6b8d2c14 Correctly handle connection terminations during finalize so mpirun doesn't hang. Cleanup some corner cases in the error notification system 2015-12-30 07:16:43 -08:00
Ralph Castain
1cdc1c121c Revert "Standardize the handling of shutdown in the OOB TCP component" 
This reverts commit open-mpi/ompi@12dccaa911.
 2015-12-30 07:05:40 -08:00
Ralph Castain
12dccaa911 Standardize the handling of shutdown in the OOB TCP component 2015-12-29 07:57:22 -08:00
rhc54
5dfb7ac396 Merge pull request #1266 from ggouaillardet/topic/misc_pmix_fixes 
Topic/misc pmix fixes
 2015-12-29 07:02:44 -08:00
Ralph Castain
810f2446b7 Add pmix120 component, update the error handling functions in the PMIx API. 
Update the configure logic for the new pmix120 component

ckpt

Get the pmix120 component to work - still not really registering or handling notifications, but infrastructure now operates

Cleanup some of the symbol scopes, and provide a more comprehensive rename.h file. Will pretty it up later - let's see how this works

Cleanup the rename files to use the pretty macros
 2015-12-28 23:15:44 +09:00
Gilles Gouaillardet
352b05a552 rmaps: warn if oversubscribing when manually setting the number of hosts 
This is a port of the v1.10 series one-off open-mpi/ompi-release@8c5ce45ab6
 2015-12-28 10:38:57 +09:00
Ralph Castain
8ab28cdc82 Fix a typo that causes segfaults on multi-node executions 2015-12-24 08:43:47 -08:00
rhc54
d7199dc75b Merge pull request #1255 from annu13/fixup 
Fixup
 2015-12-22 20:54:48 -08:00
annu13
43f44f31c1 moved code to job setup first before enabling comm 2015-12-22 14:30:59 -08:00
Howard Pritchard
39367ca0bf plm/alps: only use srun for Native SLURM 
Turns out that the way the SLURM plm works
is not compatible with the way MPI processes
on Cray XC obtain RDMA credentials to use
the high speed network.  Unlike with ALPS,
the mpirun process is on the first compute
node in the job.  With the current PLM launch
system, mpirun (HNP daemon) launches the MPI
ranks on that node rather than relying on
srun.

This will probably require a significant amount
of effort to rework to support Native SLURM
on Cray XC's.  As a short term alternative,
have the alps plm (which gets selected by default
again on Cray systems regardless of the launch system)
check whether or not srun or alps is being used on the
system.  If alps is not being used, print a helpful
message for the user and abort the job launch.

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
 2015-12-22 11:03:42 -08:00
rhc54
d9cd451a16 Merge pull request #1250 from rhc54/topic/rf 
Fix the default slot mapping in rank file mapper
 2015-12-21 10:57:52 -08:00
Ralph Castain
7cc5879bdd Fix the default slot mapping in rank file mapper 2015-12-21 09:47:27 -08:00
Ralph Castain
94ffe10808 Do not override any external settings for PMIx component selection 2015-12-21 08:36:12 -08:00
Jeff Squyres
53ca721ff4 configury: clean up .so version numbers 
Move .so version numbers to their appropriate project in the top-level
VERSION file.  Also add the project name to all .so version number
names.  Remove no-longer-used .so names.
 2015-12-18 12:50:23 -05:00
Ralph Castain
64b695669a Cleanup warnings in opal and orte layers when building optimized on Mac 2015-12-17 07:51:24 -08:00
Ralph Castain
3a56f0d34b Create the pmix external component. Fix a few places where opal/util/argv.h were required when building with an external pmix (go figure). 
NOTE: Building with external pmix *requires* that you also build with external libevent and hwloc libraries. Detect this at configure and error out with large message if this requirement is violated.

Closes #1204  (replaces it)
Fixes #1064
 2015-12-15 15:26:13 -08:00
Howard Pritchard
7a82174747 Merge pull request #1195 from hppritcha/topic/wlm_detect 
support Cray nativized slurm environment
 2015-12-15 07:58:53 -07:00
Jeff Squyres
3e308f41f7 rmaps base help: update binding error messages 
Due to user confusion, update the show-help messages displayed when
processor and/or memory binding fails.  Thanks to Dave Love
(@loveshack) for the initial suggestion.

Fixes open-mpi/ompi#1087
 2015-12-14 13:02:41 -05:00
Ralph Castain
03eb1a80bf Update the PMIx native component to release v1.1.1, with addition of one bug-fix commit beyond the official release 
Rename the pmix1xx component to pmix111 so it reflects the actual release it includes

Resolve the problem of PMIx being passed a bogus --with-platform argument when configuring the PMIx tarball code. There is no reason we should be passing --with-platform arguments to any internal subdirectory, so just leave that out when constructing the opal_subdir_args variable.

Update the PMIx code and continue attempting to debug direct modex

Fix a problem in the ORTE PMIx server - there was an early intent to optimize the direct modex by fetching data for all procs from the target job on the remote node, instead of fetching the data one proc at a time. However, this was never completely implemented, and so we would hang if we had multiple overlapping requests for data from more than one proc on the node.

Update PMIx to v1.1.2
 2015-12-12 18:46:38 -08:00
Ralph Castain
5e5adebf8e Port the changes from #782 to the master. Not everything applies here as the code in the 1.10 series is a little different. In addition, we asked for a few changes (e.g., using MPI_ERR_ARG instead of "13") that are incorporated here. 
Thanks to @jsharpe for the PR
 2015-12-12 12:40:34 -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
Jeff Squyres
00c5dc9449 rml oob: C99-ification of structure member assignment 2015-12-08 17:05:16 -08:00
Howard Pritchard
cb7c26ce96 plm/slurm: add support for cray native slurm 
Cray has added plugins to slurm to support
the Cray programming env (alpslli, cray pmi, etc).
Some of the workarounds needed with plm/alps
to avoid issues with Cray PMI getting mixed up
with orte launch system are also required in
a cray native slurm environment.

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
 2015-12-08 13:47:20 -06:00
Howard Pritchard
9548b8a9e8 plm/alps: add wlm detect infrastructure 
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
 2015-12-07 07:43:20 -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
Mark Santcroos
3119bc14b2 Merge branch 'master' into fix/alpsinfov3 2015-11-13 08:53:06 -05:00
Ralph Castain
986a8c1d48 If an executable isn't found, it's possible for the state machine to hit the grpcomm with a zero-node map before we actually terminate with error. Silence the annoying malloc warning about zero-byte requests. 
In a novm operation that only has the HNP, ensure the #nodes gets set

Clean up the error reporting
 2015-11-11 14:24:13 -08:00
Jeff Squyres
8bd356549a orte proc_info.h: use symbolic names 
This fix was actually applied in the v2.x branch first (as commit
open-mpi/ompi-release@a9b22afc1a).
 2015-11-10 13:39:21 -08:00
Mark Santcroos
299fd69c6d Merge branch 'master' into fix/alpsinfov3 2015-11-10 15:40:19 -05:00
rhc54
474a869b8d Merge pull request #1121 from dmt4/orterun-manpage-typos 
change -0bind-to and -bind-to to --bind-to in the manpages
 2015-11-10 11:24:08 -08:00
Dimitar Pashov
9f6e306064 change -0bind-to and -bind-to to --bind-to in the manpages 2015-11-10 17:44:53 +00:00
Ralph Castain
6a607d42a6 Prevent a segfault on tools if a connection attempt fails - tools don't open the opal/pmix framework and thus have no way of looking up a proc hostname 2015-11-10 09:11:34 -08:00
Mark Santcroos
5ec2b4d98c Fix some messages in the process. 2015-11-09 18:03:26 -05:00
Mark Santcroos
8ec89001b3 Merge branch 'master' into fix/alpsinfov3 2015-11-09 02:45:23 -05:00
Ralph Castain
9b0cdc0de2 Add support for -pernode and -npernode options to orte-submit 2015-11-08 11:34:18 -08:00
Ralph Castain
f1483eb2dc Need to delay registration of the waitpid callback until after the fork/exec of the child process. Fix the bit testing of process type so that the proper state component gets selected for HNP. 2015-11-06 21:35:24 -08:00
Ralph Castain
5f446570d8 Work on cleaning up memory leaks that are causing orte-dvm to eventually run out of memory. Still don't have everything plugged, but getting better. Sync to the PMIx master that includes removal of the pmix_common.h.in file that really didn't need to be generated, and update to the PMIx_server_init API. 2015-11-06 14:15:30 -08:00
Mark Santcroos
a40b4eb2ee Support ALPS_APPINFO_VERSION 3. 2015-11-06 09:53:41 -05:00
Ralph Castain
ec0cc4bf21 Ensure that we completely register an nspace prior to launching local procs as otherwise we may attempt to send it down before it is registered, leading to data corruption 2015-11-05 20:51:56 -08:00
Ralph Castain
68996d6858 Move the argv_free back to the correct place - I blame Jeff for suggesting it was wrong to begin with 2015-11-05 07:57:54 -08:00
Ralph Castain
169c44258d Fix missing check 2015-11-03 19:00:28 -08:00
Ralph Castain
fe0c995f6b Fix a couple of minor issues identified by Jeff 2015-11-03 17:30:51 -08:00
Ralph Castain
186c18be0e Add missing cmd line options to mpirun man page, update NEWS to contain that change 2015-11-01 09:19:08 -08:00
Ralph Castain
0523f60479 Remove debug from orte-submit help output 2015-11-01 09:19:07 -08:00
rhc54
1fe27bf1dd Merge pull request #1084 from rhc54/topic/dashhost 
Fix relative node syntax for dash-host option
 2015-10-31 21:24:39 -07:00
Ralph Castain
8bfbe7f16c Add a new MCA parameter for default_dash_host to offer a mirror of the default_hostfile 2015-10-31 19:09:54 -07:00
Ralph Castain
24419b6523 Fix relative node syntax for dash-host option 2015-10-31 19:00:46 -07:00
rhc54
b23f1f3578 Merge pull request #1080 from federeghe/bugfixes 
oob_tcp: fix peer->state wrong check
 2015-10-31 16:09:23 -07:00
Ralph Castain
22dc05194e Minor cleanup - explicitly NULL the last member of a function pointer module. Should default to that anyway, but this is cosmetically nicer. 2015-10-30 08:19:55 -07:00
Federico Reghenzani
6536a6a9f5 oob_tcp: fix peer->state wrong check 2015-10-29 16:43:58 +01:00
Ralph Castain
267ca8fcd3 Cleanup the PMIx direct modex support. Add an MCA parameter pmix_base_async_modex that will cause the async modex to be used when set to 1. Default it to 0 for now 
to continue current default behavior.

Also add an MCA param pmix_base_collect_data to direct that the blocking fence shall return all data to each process. Obviously, this param has no effect if async_
modex is used.
 2015-10-27 17:31:56 -07:00
rhc54
3ffbf08283 Merge pull request #1068 from marksantcroos/master 
Make odsl debug message consistent.
 2015-10-24 08:11:11 -07:00
Mark Santcroos
30aab75b86 Make message consistent. 2015-10-24 13:40:03 +02:00
Ralph Castain
6506b0a5e5 Resolve a race condition that prevented the sigchild callback from being registered before short-lived apps terminated 
Thanks to Mark Santcroos for the assistance in tracking it down.
 2015-10-23 21:02:31 -07:00
Nathan Hjelm
9602484568 Merge pull request #1040 from hjelmn/mtl_priority 
Change how cm's priority is calculated
 2015-10-19 14:18:36 -06:00
Nathan Hjelm
8b5810f7f7 mca/base: add priority output to mca_base_select 
The mca_base_select function uses returned priorities to select the
best component/module. This priority may be of use to the caller so
pass that information back in an optional argument. If the priority is
not needed pass NULL.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-10-19 12:32:41 -06:00
Ralph Castain
363f62a506 Fix singleton operations when running under a SLURM allocation. Sadly, SLURM's PMI will return success even if the PMI server isn't actually available. This leads to erroneous selection of pmix and ess components. So add a further requirement (namely, that we see a job_step envar) to the SLURM pmix components along with some modification of ess selection code to avoid the problem 2015-10-17 20:24:03 -07:00
Jeff Squyres
62351f442a help: remove stale help messages and files 
Found by contrib/check-help-strings.pl.
 2015-10-13 16:50:20 -04:00
Jeff Squyres
f9e9b69d93 Merge pull request #1001 from igor-ivanov/master 
orte/mca/rmaps: Improve orte_rmaps_dist_device help message
 2015-10-09 14:07:47 -04:00
Igor Ivanov
489f27f8e9 orte/mca/rmaps: Improve orte_rmaps_dist_device help message 
See: https://github.com/open-mpi/ompi/issues/953
 2015-10-09 17:58:07 +03:00
rhc54
232f97a80c Merge pull request #968 from JohnWestlund/master 
simplify use of sockaddr* structs to work around buffer overflow warning
 2015-10-07 17:42:19 -07:00
Howard Pritchard
d899320574 odls/alps: close the directory 
Close the /proc/self/fd dir after checking for open fds.

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
 2015-10-06 11:13:44 -07:00
Igor Ivanov
d379873443 oshmem: Add man.1 pages for oshmem tools 
This changes add man pages for oshrun, oshcc and oshfort as well as
depricated shmemrun, shmemcc and shmemfort.
 2015-10-05 15:41:28 +03:00
John Westlund
044fea8df7 re-order != comparison, OBJ_RELEASE mca_oob_tcp_addr_t on failure 2015-10-02 15:59:48 -07:00
John Westlund
6bfaa925ec simplify use of sockaddr* structs to work around buffer overflow warning 2015-10-02 14:26:52 -07:00
Ralph Castain
8f6855459d Cleanup some coverity warnings 2015-09-30 10:33:53 -07:00
Gilles Gouaillardet
0445484820 ras: remove orte_ras_proc_t and associated code 2015-09-30 08:52:52 +09:00
Gilles Gouaillardet
7cc14ee6f6 orte/rmaps: silence warning 2015-09-29 16:05:52 +09:00
Ralph Castain
fad5638596 Resolve the naming issue when direct-launched by PMIx-enabled RMs using a minimal-impact approach. Detect if we were launched via ORTE - if so, then use our standard methods for computing the jobid. If not, then just hash the nspace to create the jobid, and track the jobid <-> nspace correspondece down in the opal/mca/pmix/pmix1xx component. We then do the translation any time a function that passes process names is invoked. 2015-09-27 09:57:59 -07:00
Ralph Castain
0140ff048d Now that we have an "isolated" PLM component, we cannot just let rsh silently decline to run when it cannot find a launch agent - if we do, then we will -always- run on the local node. So if the user specifies a launch agent and we can't find it, then generate a pretty error message, report a fatal error back to the component select, and exit out. 
This required modifying the mca_component_select function to actually check the return code on a component query - it was blissfully ignoring it.

Also do a little cleanup to avoid bombarding the user with multiple error messages.

Thanks to Patrick Begou for reporting the problem
 2015-09-24 07:16:48 -07:00
Ralph Castain
749bd4e6fe Plug a few memory leaks identified by valgrind 2015-09-23 15:21:04 -07:00
Ralph Castain
f28448702a Eliminate malloc by utilizing /proc/self/fd - optimization 2015-09-22 07:24:54 -07:00
Ralph Castain
f872e99315 Fix orte-submit so it allows application procs to select the correct ess component. Protect orte_data_server from multiple calls to finalize. 2015-09-21 20:31:57 -07:00
Howard Pritchard
ef6cf50687 Merge pull request #917 from hppritcha/topic/alps_warning_swat 
oob/alps: swat compiler warning
 2015-09-21 16:17:30 -06:00
Howard Pritchard
8d7e759b85 oob/alps: swat compiler warning 
swat some alps related compiler warnings when using --enable-picky

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
 2015-09-21 14:24:26 -07:00
Ralph Castain
92ae386a34 As Jeff proposed, change the check to looking for the filename's first character to be a digit 2015-09-21 08:22:58 -07:00
rhc54
13def2a69b Merge pull request #911 from rhc54/topic/cleanup 
Cleanup the odls "close file descriptor" commit to conform to OMPI co…
 2015-09-20 07:01:39 -07:00
Howard Pritchard
1367a442b6 Merge pull request #910 from hppritcha/topic/odls_alps_use_907_stuff 
odls/alps: do smarter close of fds in child
 2015-09-20 07:37:55 -06:00
Ralph Castain
c167acc5a7 Cleanup the odls "close file descriptor" commit to conform to OMPI coding standards and remove memory leaks 2015-09-19 20:46:36 -07:00
Howard Pritchard
a31cc21bea odls/alps: do smarter close of fds in child 
Use a modified variant of #907.  Thanks to plesn
for noticing this.

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
 2015-09-19 14:17:05 -07:00
Piotr Lesnicki
1dd5487fae odls: close only used file descriptors at fork/exec 2015-09-18 16:44:57 +02:00
Ralph Castain
1b7930ad52 Silence some warnings and address Coverity issues 2015-09-16 07:58:22 -07:00
Ralph Castain
8b88ea9b13 Fix singletons by removing stale code 2015-09-16 00:58:05 -07:00
Ralph Castain
c1bbbb5e2f Remove the last involvement of the OOB system from the MPI layer, remove the no-longer-needed usock/oob component, and have procs no longer open the RML, OOB, ROUTED, and GRPCOMM frameworks as PMIx now provides all required app-mpirun cmds 2015-09-15 13:08:35 -07:00
Ralph Castain
22d7c0081a Fix the no-disconnect test by resolving a segfault on free - opal_dss.unload will return the remaining unpacked portion of a buffer. As such, it cannot return the pointer to that info as it might be partway inside of a malloc'd region. So copy the data out of the buffer. 2015-09-11 13:01:35 -07:00
Ralph Castain
dc5796b8a1 Revert "Revert "Fix the handling of cpusets so we get the correct cpuset for each local peer. Add the ability to indicate that a modex request is "optional" so we don't call the server if we don't find the value. Take advantage of that to allow the MPI layer to decide that the lack of locality info indicates non-local"" 
Fix the locality computation by correctly computing the vpid of the local peer

This reverts commit open-mpi/ompi@6a8fad49e5.
 2015-09-11 08:29:51 -07:00
Ralph Castain
6a8fad49e5 Revert "Fix the handling of cpusets so we get the correct cpuset for each local peer. Add the ability to indicate that a modex request is "optional" so we don't call the server if we don't find the value. Take advantage of that to allow the MPI layer to decide that the lack of locality info indicates non-local" 
This reverts commit f94f3cda21.
 2015-09-11 02:01:25 -07:00
Ralph Castain
f94f3cda21 Fix the handling of cpusets so we get the correct cpuset for each local peer. Add the ability to indicate that a modex request is "optional" so we don't call the server if we don't find the value. Take advantage of that to allow the MPI layer to decide that the lack of locality info indicates non-local 2015-09-10 10:25:30 -07:00
rhc54
f6b6b9a9ca Merge pull request #877 from rhc54/topic/s1s2 
Cleanup s1 and s2 components
 2015-09-08 19:20:59 -07:00
Ralph Castain
1cdb86b8c7 Cleanup s1 and s2 components, and ensure that mpirun and orteds only use non-direct-launch pmix components. 2015-09-08 18:37:09 -07:00
Ralph Castain
459f169e06 Fix segfault upon job error 
Silence some unnecessary error-logs
 2015-09-08 14:03:06 -07:00
Jeff Squyres
bc9e5652ff whitespace: purge whitespace at end of lines 
Generated by running "./contrib/whitespace-purge.sh".
 2015-09-08 09:47:17 -07:00
Ralph Castain
e6add86e4f Deal with connect/accept between two jobs from different mpirun's. Somewhat optimize connect/accept by using MPI bcast to distribute the participants instead of another PMIx lookup. Cleanup some Coverity issues. 2015-09-07 09:19:24 -07:00