1
1
Граф коммитов

25147 Коммитов

Автор SHA1 Сообщение Дата
Jeff Squyres
17202e5177 Merge pull request #1733 from jsquyres/pr/hwloc1113-fix
hwloc1113: add missing file to Makefile.am
2016-05-31 13:59:08 -04:00
Jeff Squyres
5cfee95ea4 hwloc1113: add missing file to Makefile.am
Lack of this file causes a failure when you run autogen.pl on a
distribution tarball.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2016-05-31 09:57:50 -07:00
rhc54
93ff4ce36d Merge pull request #1731 from rhc54/topic/timeout
Provide ETIMEDOUT as the mpirun exit code if the timeout limit was hit
2016-05-31 08:41:21 -07:00
Ralph Castain
0cd0ccb7fd Provide ETIMEDOUT as the mpirun exit code if the timeout limit was hit 2016-05-31 07:45:31 -07:00
Gilles Gouaillardet
1bbc5fadee ompi/win: silence an other warning 2016-05-31 13:18:39 +09:00
Gilles Gouaillardet
c41321b9e5 ompi/win: silence warning 2016-05-31 13:03:20 +09:00
rhc54
0965cb3d41 Merge pull request #1730 from rhc54/topic/pmixext
Patch from Gilles - modify detection of PMIx version for external libraries
2016-05-30 18:50:12 -07:00
Ralph Castain
7b115a9e0b Patch from Gilles - modify detection of PMIx version for external libraries 2016-05-30 14:30:10 -07:00
George Bosilca
d2abff583e Fix race condition during BTL TCP tear-down.
bot🏷️bug
bot:assign:@hjelmn
2016-05-30 10:47:14 -05:00
rhc54
876257469e Merge pull request #1728 from rhc54/topic/sim
Enable simulation of large-scale clusters
2016-05-29 21:29:16 -07:00
Ralph Castain
3913595e10 Enable simulation of large-scale clusters by allowing multiple daemons/node. Specifying the ras_base_multiplier parameter to be greater than 1 will cause ORTE to replicate each allocated node by that factor. A daemon will be spawned for each replica, thus letting ORTE function as if it were on a much larger cluster.
Note that this cannot be used for MPI performance testing. It is really only useful for ORTE scaling tests. It also only works with the rsh/ssh launcher.
2016-05-29 18:56:18 -07:00
rhc54
a93c01d4f4 Merge pull request #1724 from rhc54/topic/timeout
Add a timeout cmd line option and an option to report state info upon timeout to assist with debugging Jenkins tests
2016-05-28 08:36:41 -07:00
Ralph Castain
ebe159acef Add a timeout cmd line option and an option to report state info upon timeout to assist with debugging Jenkins tests
If requested, obtain stacktraces for each application process and report it to stderr upon timeout

stack traces: minor improvements

- Also include the hostname and PID of the each process for which
  we're sending the stack traces (vs. just including the ORTE process
  name)
- Send a specific error message if we couldn't find "gstack" in the
  $PATH (e.g., on OS X)
- Send a sepcific error message if gstack fails to run
- Print a message that obtaining the stack traces may take a few
  seconds so that users don't wonder what's happening

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>

help-orterun.txt: minor tweaks

Trivial update: show "--timeout" (instead of "-timeout") in the help
message, just to encourage the use of double-dash options.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>

trivial: stacktrace -> stack trace

Trivial word smything.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2016-05-28 08:36:25 -07:00
Jeff Squyres
59f4a765b3 Merge pull request #1656 from hpcraink/pr/make_manpage
In case, we do not build Fortran, Fortran 2008 or CXX, the regexp in …
2016-05-28 11:02:12 -04:00
Jeff Squyres
e126d2cd18 Merge pull request #1584 from bgoglin/master
Update hwloc to v1.11.3
2016-05-28 11:01:54 -04:00
Nathan Hjelm
d8fd3a411a Merge pull request #1725 from hjelmn/request_fixes
ompi/request: fix bugs in MPI_Wait_some and MPI_Wait_any
2016-05-27 13:47:49 -06:00
Nathan Hjelm
0591139f49 ompi/request: fix bugs in MPI_Wait_some and MPI_Wait_any
This commit fixes two bugs in MPI_Wait_any:

 - If all requests are inactive then the sync wait would hang forever
   because no requests are attached to the sync.

 - The request pointer was pointing to the request before the completed
   request which caused the wrong request to be freed or marked inactive.

MPI_Wait_some had a similar issue if all the requests were pending.

These issues were identified by MTT.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-27 12:36:10 -06:00
Nathan Hjelm
3974987ba3 Merge pull request #1723 from hjelmn/warning_fixes
win: fix warnings
2016-05-27 12:26:04 -06:00
Nathan Hjelm
0adfb328e1 win: fix warnings
Signed-off-by: Nathan Hjelm <hjelmn@me.com>
2016-05-27 10:14:02 -06:00
rhc54
e5ee7adbe0 Merge pull request #1722 from rhc54/topic/pmixext
Enable PMIx external support for both 1.1.4 and 2.0 versions
2016-05-27 08:59:09 -07:00
Ralph Castain
55923eacd3 Stealing some pieces of Josh Hursey's PR #1583 and modifying a bit, allow the opal/pmix external component to handle both PMIx 1.1.4 and PMIx 2.0 versions. Automatically detect the version of the target external library and adjust the only two APIs that changed (PMIx_Init and PMIx_Finalize)
Rename temp vars in .m4 to avoid conflict with Travis
2016-05-27 08:06:31 -07:00
Nathan Hjelm
d25b846c01 Merge pull request #1704 from hpcraink/pr/configure_framework
Fix configure for FreePGI on OSX
2016-05-26 17:01:08 -06:00
Nathan Hjelm
8c9292d5d1 Merge pull request #1721 from hjelmn/xrc_fix
btl/openib: fix XRC WQE calculation
2016-05-26 17:00:31 -06:00
Nathan Hjelm
56bdcd0888 btl/openib: fix XRC WQE calculation
Before dynamic add_procs support was committed to master we called
add_procs with every proc in the job. The XRC code in the openib btl
was taking advantage of this and setting the number of work queue
entries (WQE) based on all the procs on a remote node. Since that is
no longer the case we can not simply increment the sd_wqe field on the
queue pair. To fix the issue a new field has been added to the xrc
queue pair structure to keep track of how many wqes there are total on
the queue pair. If a new endpoint is added that increases the number
of wqes and the xrc queue pair is already connected the code will
attempt to modify the number of wqes on the queue pair. A failure is
ignored because all that will happen is the number of active send work
requests on an XRC queue pair will be more limited.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-26 15:58:31 -06:00
Aurelien Bouteiller
49bd28d0ac Merge pull request #1714 from hjelmn/scif_exclusivity
btl/scif: reduce default exclusivity
2016-05-26 17:53:11 -04:00
Nathan Hjelm
f19c647f21 Merge pull request #1718 from hjelmn/config_fix
config: fix typo in mxm configury
2016-05-26 13:19:23 -06:00
Joshua Ladd
1a5fd6bf83 Merge pull request #1719 from ICLDisco/ucx_request_fix
Removal of ompi_request_lock from pml/ucx.
2016-05-26 15:09:57 -04:00
Thananon Patinyasakdikul
60d0fbf683 Removal of ompi_request_lock from pml/ucx. 2016-05-26 12:36:58 -04:00
Nathan Hjelm
8c2086995d config: fix typo in mxm configury
A 1 was missing when setting $1_LDFLAGS leading to erroneous items in
the wrapper cflags.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-26 10:28:07 -06:00
Nathan Hjelm
87ea9be863 Merge pull request #1715 from hjelmn/ugni_overhead
btl/ugni: reduce overhead of progress function
2016-05-26 10:17:00 -06:00
Gilles Gouaillardet
46710ba151 travis: fix a typo and create bogus directories to avoid compiler warnings 2016-05-26 15:28:10 +09:00
George Bosilca
90f294096e Remove more references to the request mutex.
Regarding BFO it should be mentionned that this component is currently
unmaintained, and that despite my efforts I could not make it compile
(it would not compile before this patch either).
2016-05-25 23:27:06 -04:00
Nathan Hjelm
5d322170a0 Merge pull request #1716 from hjelmn/request_fixes
Request fixes
2016-05-25 18:14:03 -06:00
Nathan Hjelm
9d439664f0 pml/yalla: update for request changes
This commit brings the pml/yalla component up to date with the request
rework changes.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-25 15:42:53 -06:00
Nathan Hjelm
8445c885ce pml/cm: update for request changes
This fixes a hang caused by the request refactor work. The cm pml was
not updated and was hanging is most cases.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-25 15:35:32 -06:00
Nathan Hjelm
dbfab94ede atomic/mxm: rename symbol that is a duplicate of one in atomic/ucx
This fixes an error when building with --enable-static.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-25 15:34:40 -06:00
Nathan Hjelm
99627319f0 btl/ugni: reduce overhead of progress function
This commit reduces the overhead of calling the ugni progress
function. It does the following:

 - Check for new connections once every eight calls.

 - Do not call remote smsg progress unless we are connected to at
   least one remote peer.

 - Do not call rdma progress unless at least one rdma fragment is
   outstanding.

 - Check endpoint wait list size before obtaining a lock.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-25 14:27:34 -06:00
Nathan Hjelm
5caf12cd9b btl/scif: reduce default exclusivity
This commit reduces the default exclusivity so that btl/scif is not
used for send/recv over other shared memory transports.

Fixes open-mpi/ompi#1712

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-25 14:25:07 -06:00
Nathan Hjelm
8e1d59aea8 Merge pull request #1708 from hjelmn/c__fix
request: fix compilation error
2016-05-25 10:48:02 -06:00
Nathan Hjelm
ef11ba9394 request: fix compilation error
The request.h header is unfortunately included files in the C++
bindings. C++ does not allow assigning from void * to another
pointer without a cast. This commit adds the cast. We can clean this
up when the C++ bindings are deleted.

Fixes open-mpi/ompi#1707

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-25 09:52:23 -06:00
Joshua Ladd
ce783a9ebf Merge pull request #1706 from vspetrov/coll_hcoll_req_type_bugfix
coll/hcoll: bugfix: initialize req_type field
2016-05-25 10:56:33 -04:00
Valentin Petrov
5ff6372886 coll/hcoll: bugfix: initialize req_type field
If left uninitialized then segfault is possible in MPI_Waitall in
    the case the field by chance equals OMPI_REQUEST_GEN.
2016-05-25 15:38:01 +03:00
Rainer Keller
3727cba9bb Fix compilation for FreePGI on OSX
Our checks and the ones of libevent are somewhat flawed.
If adding multiple "-framework" to CXXFLAGS or CFLAGS, we strip
the keyword from the command-line, not good.
libevent however assumes plain gcc without testing properly
that the compiler supports -Wno-deprecated-declarations.
2016-05-25 09:12:39 +02:00
George Bosilca
2b868c4952 Fix MPI datatype args.
Compensate for the datatype ID that we add to the array.
2016-05-24 23:36:54 -04:00
Jeff Squyres
dd9a819a1c odls_default: do not opal_output() while creating a process!
It is verbotten to use opal_output() after the fork() but before the
exec()!  It results in all manner of undefined behavior.  For example,
on some OS X systems, if you run a trivial "hello world" MPI program
with a high level of ODLS verbosity:

```sh
$ mpirun -np 3 --mca odls_base_verbose 100 ./hello_c
```

You will see a bunch of output from the mpirun ODLS base, but then it
*may* hang in odls_default_module.c:do_child() -- after the fork() but
before the exec() -- while trying to opal_output() some debugging
statements.

The solution is to remove these extraneous opal_output() statements.
Indeed, the ODLS base is already outputting the same information that
these opal_output() statements are trying to emit, anyway.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2016-05-24 21:28:57 -04:00
Nathan Hjelm
461ca1203b Merge pull request #1703 from hjelmn/grdma_cuda_fix
rcache/grdma: fix typo in cuda code
2016-05-24 18:51:22 -06:00
bosilca
b90c83840f Refactor the request completion (#1422)
* Remodel the request.
Added the wait sync primitive and integrate it into the PML and MTL
infrastructure. The multi-threaded requests are now significantly
less heavy and less noisy (only the threads associated with completed
requests are signaled).

* Fix the condition to release the request.
2016-05-24 18:20:51 -05:00
Nathan Hjelm
af52dad8f8 rcache/grdma: fix typo in cuda code
Fixes open-mpi/ompi#1702

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-24 15:56:39 -06:00
Nathan Hjelm
1d3110471c Merge pull request #1697 from hjelmn/acc_order
win: add support for accumulate_ordering info key
2016-05-24 14:34:05 -06:00
Nathan Hjelm
5126da5377 win: add support for accumulate_ordering info key
This commit adds support for the MPI-3.1 accumulate_ordering info
key. The default value is rar,war,raw,waw and is supported using an
MCA variable flag enumerator.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2016-05-24 11:13:30 -06:00