1
1
Граф коммитов

3839 Коммитов

Автор SHA1 Сообщение Дата
Gilles Gouaillardet
3300d7cc00 pmix: rename pmix_munge_module 2015-12-28 23:16:27 +09:00
Ralph Castain
a5b95a0939 Continue work on error notification system 2015-12-28 23:15:59 +09:00
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
c757c5c612 pmix/external: Fix error handler usage 2015-12-28 23:15:17 +09:00
Gilles Gouaillardet
1157329732 configury: misc pmix112 fixes 2015-12-28 23:15:16 +09:00
Gilles Gouaillardet
d416c7fd8a pmix/external: no more circular dependencies if not building shared DSO 2015-12-28 23:14:03 +09:00
Gilles Gouaillardet
4f8c0c5f0d event/external: add missing include file 2015-12-28 17:22:59 +09:00
Gilles Gouaillardet
fec973efda configury: test portability
replace test ... -o ... with test ... || test ...
and test ... -a ... with test ... && test ...
2015-12-28 13:58:45 +09:00
Nathan Hjelm
700a21022a Merge pull request #1260 from artpol84/openib_proc_account_fix
Openib proc accounting fix
2015-12-27 15:19:52 -07:00
Artem Polyakov
a20826e6b4 Fix vader resource leak.
This nasty bug was nicely masked. It was causing `mca_btl_vader_component.vader_frags_user`
overflow and as the result rear hangs of ompi-test-suite.
2015-12-28 00:41:45 +06:00
Gilles Gouaillardet
2d9aa38e6a btl/openib: fix heterogeneous support 2015-12-25 16:31:35 +09:00
Artem Polyakov
3031affdb7 Fix openib process accounting if procs was dynamically added. 2015-12-24 17:56:35 +06:00
Artem Polyakov
400af6c52d openib addproc improvements:
1. finer grained locks;
2. separate srq creation from cq adjustments.
2015-12-24 17:56:35 +06:00
Artem Polyakov
41c325f15a Shift common code for calculating a port count and btl_rank in openib
into the static function
2015-12-24 17:56:35 +06:00
Gilles Gouaillardet
0b3e3c6817 opal/runtime: add missing #include <unistd.h>
Thanks Marco Atzeri for contributing the original patch
2015-12-24 14:41:56 +09:00
Gilles Gouaillardet
f0e3e16f49 pmix/base: add missing #include <unistd.h>
Thanks Marco Atzeri for contributing the original patch
2015-12-24 14:41:52 +09:00
Gilles Gouaillardet
66d9c2daea rcache/vma: add missing #include "opal/util/output.h"
Thanks Marco Atzeri for contributing the original patch
2015-12-24 14:41:49 +09:00
Gilles Gouaillardet
5fa63f086a btl/tcp: add missing #include <unistd.h>
Thanks Marco Atzeri for contributing the original patch
2015-12-24 14:41:46 +09:00
Gilles Gouaillardet
15ed7ad9f5 btl/sm: add missing #include <unistd.h>
Thanks Marco Atzeri for contributing the original patch
2015-12-24 14:41:41 +09:00
Gilles Gouaillardet
65a081ae6a mca/base: add missing #include "opal/util/output.h" and <unistd.h>
Thanks Marco Atzeri for contributing the original patch
2015-12-24 14:41:33 +09:00
Gilles Gouaillardet
42313acd58 btl/usnic: add missing #include <alloca.h> 2015-12-24 14:33:58 +09:00
Gilles Gouaillardet
38a8826136 opal/datatype: #include <alloca.h> when needed and nowhere else 2015-12-24 14:33:58 +09:00
Nathan Hjelm
84d890b7e7 Merge pull request #1248 from artpol84/openib_proc_init_race
Openib dynamic add proc race conditions
2015-12-22 21:48:05 -07:00
Artem Polyakov
08ad8357a8 Fix local process accounting in openib when dynamic add_proc is on. 2015-12-22 22:44:46 +06:00
Artem Polyakov
3c2f6d5560 Protect openib_btl->device data with explicit opal_mitex locks. 2015-12-22 18:33:26 +06:00
Gilles Gouaillardet
607d7c7545 btl/sm: rename file after file descriptor has been closed.
Thanks George for spotting this.
2015-12-22 13:56:53 +09:00
Artem Polyakov
e06bffe213 Fix ib_proc locking 2015-12-21 18:52:31 +06:00
Artem Polyakov
3eb4756a17 Force locking regardles to the opal_using_threads() setting. 2015-12-21 18:52:31 +06:00
Artem Polyakov
11b72d9add Make important fields of ib_proc volatile. 2015-12-21 18:52:31 +06:00
Artem Polyakov
86c0c3ec52 Provide additional information: whether ib_proc was newly created or
it was already existing.
2015-12-21 18:52:31 +06:00
Artem Polyakov
9325bd3d69 Protect device initialization 2015-12-21 18:52:31 +06:00
Artem Polyakov
0f77bc7ea7 Perform endpoint initialization atomically. 2015-12-21 18:52:31 +06:00
Artem Polyakov
afaf9c9ea6 Shift ib_proc initialization to the separate function. 2015-12-21 18:52:31 +06:00
Artem Polyakov
3c9fd567b6 Fix openib race condition when direct modex is used.
The problem was in mca_btl_openib_proc_create. This function may be called
from several places simultaneously:
* from the main thread when somebody wants to do `MPI_Send()` (for example) for
the first time;
* from udcm if the counterpart peer is trying to connect and `mca_btl_openib_get_ep()`
is called.

In this case one of the threads may add an uninitialized proc structure
to the `mca_btl_openib_component.ib_procs` and the other will read it and
treat as initialized.

This commit turns ib_proc initialization into a single atomic operation.
2015-12-21 18:52:30 +06:00
Gilles Gouaillardet
db4f483653 btl/sm: fix race condition
write to file and then rename, so when the file is open for read, its content is known to have been written.

Fixes open-mpi/ompi#1230
2015-12-21 16:37:51 +09:00
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
rhc54
978c54880d Merge pull request #1238 from rhc54/topic/cleanup
Cleanup warnings in opal and orte layers when building optimized on Mac
2015-12-17 09:37:48 -08:00
Ralph Castain
64b695669a Cleanup warnings in opal and orte layers when building optimized on Mac 2015-12-17 07:51:24 -08:00
Nathan Hjelm
e77199fd4f Merge pull request #1235 from ggouaillardet/topic/ibv_exp_fixes
btl/openib: do not mix exp and non exp verbs
2015-12-17 08:36:09 -07:00
Gilles Gouaillardet
994a627f82 btl/openib: do not mix exp and non exp verbs 2015-12-17 16:45:43 +09:00
Artem Polyakov
0951a34e95 Fix openib memory registration limit calculation if cutoff = 0. 2015-12-17 13:45:19 +06:00
Gilles Gouaillardet
75d16cfb27 Fix a few places where opal/util/argv.h were required when building pmix components (go figure) 2015-12-17 16:19:25 +09:00
Jeff Squyres
2b9341a38a usnic: fix embarrissing typo 2015-12-15 19:01:19 -08:00
rhc54
7a9106e74d Merge pull request #1226 from rhc54/extpmix
Create the pmix external component.
2015-12-15 17:33:14 -08:00
Jeff Squyres
944d5061a6 usnic: sendto() can return EPERM if we send too fast
If we send too fast, sendto() can run out of resources and return
EPERM.  So delay a little and try again.
2015-12-15 15:31:29 -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
Jeff Squyres
ab1bbca5b9 usnic: improve error message
When sendto() fails, it would be helpful to see the errno value.
2015-12-15 15:04:25 -08:00
Jeff Squyres
c1a6beac8d usnic: fix error message
There were too many "%s" instances.  Re-order the output so that we
show file, line, and then the error message.
2015-12-15 14:48:38 -08:00
Nathan Hjelm
c98086f028 Merge pull request #1223 from hjelmn/ib_use_srq
btl/openib: use only SRQ on ib by default
2015-12-15 14:04:19 -08:00
Nathan Hjelm
00da520fd5 Merge pull request #1222 from hjelmn/vader_fix
btl/vader: do not attempt to munmap opal/shmem pointer
2015-12-15 09:06:50 -08:00