ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
22 532 коммитов 3 Ветки 0 Теги 130 MiB
d251fa1525
Граф коммитов

5357 Коммитов

Автор SHA1 Сообщение Дата
George Bosilca
420ae98dfe Remove all unnecessary whitespaces and make sure we close the module 
correctly.
 2015-03-05 13:00:13 -05:00
Gilles Gouaillardet
d6ae0a5e05 sharedfp/sm: fix misc memory leaks 
as reported by Coverity with CIDs 1196785, 1196787 and 1269896
 2015-03-05 16:33:32 +09:00
Gilles Gouaillardet
5b2122381b ompio: fix misc memory leaks 
as reported by Coverity with CIDs 72127, 72145, 72146, 72177, 72179,
72186, 731276, 731278, 1269888, 1269890
 2015-03-05 16:22:19 +09:00
Gilles Gouaillardet
ceeb0844b6 dpm: fix misc memory leaks 
as reported by Coverity with CIDs 1196737 and 1269850
 2015-03-05 14:20:09 +09:00
Gilles Gouaillardet
e75b1e6435 fs/base: fix misc memory leak 
as reported by Coverity with CID 72202
 2015-03-05 14:20:08 +09:00
Gilles Gouaillardet
9f13425980 fbtl/posix: fix misc memory leaks 
as reported by Coverity with CIDs 72125, 72126, 1269899 and 1269900
 2015-03-05 14:20:08 +09:00
Gilles Gouaillardet
838cd51644 pubsub: fix misc memory leak 
as reported by Coverity with CID 710627
 2015-03-05 14:20:08 +09:00
Gilles Gouaillardet
d0dded1e05 topo/base: fix misc memory leaks 
as reported by Coverity with CIDs 1269901 ans 1269902
 2015-03-05 14:20:08 +09:00
Gilles Gouaillardet
d1b2f043ff fix misc memory leaks 
as already reported by Coverity with CIDs
71818, 71819, 72250, 715767, 1196749 and 1274002
 2015-03-05 13:58:05 +09:00
Howard Pritchard
53fd425a6a romio: patches from Rob Latham for issue #255 
Patches supplied by Rob Latham which fix issue #255.

See
http://git.mpich.org/mpich.git/commit/4e80e1d2b9
http://git.mpich.org/mpich.git/commit/5a10283bf7fd

Signed-off-by: Howard Pritchard <hppritcha@gmail.com>
 2015-03-02 15:33:49 -08:00
Yohann Burette
375611505b mtl/ofi: Fix synchronous Recv from ANY_SRC. 2015-03-02 08:57:46 -08:00
Alex Mikheev
168c83ed95 OMPI/MXM: add out of band barrier at the end of del_procs 
mxm shutdown requires out of band barrier
 2015-03-02 12:56:02 +02:00
Gilles Gouaillardet
464dbccb7b ompi/bcol: fix misc memory leaks 
as reported by Coverity with CIDs 715758-715762,715765 and 715767
 2015-03-02 12:48:02 +09:00
Howard Pritchard
2b5c52fb05 romio: add comment per Rob Latham's request 2015-02-28 11:05:03 -07:00
Jeff Squyres
8d04215741 coll: trivial spelling fix 
s/Algoritm/Algorithm/g
 2015-02-27 18:20:17 -08:00
Rolf vandeVaart
30e9dd5066 Look in extra rdma array to find bml. This is needed with recent BML changes. Only affects CUDA-aware code. 2015-02-27 09:02:21 -05:00
Gilles Gouaillardet
71ac1331f1 coll/tuned: remove unused variables 2015-02-27 17:26:48 +09:00
Gilles Gouaillardet
b179a17018 coll/base: add function prototypes 2015-02-27 17:26:36 +09:00
Gilles Gouaillardet
ce2020d255 coll/base: fix error reporting 
and silence CID 1271639
 2015-02-27 17:04:26 +09:00
George Bosilca
ced44e12da Update copyright. 2015-02-26 15:54:58 -05:00
George Bosilca
47e6e15e02 Typo in a rebase. 2015-02-26 15:54:19 -05:00
George Bosilca
d126c2e6f8 Fix few COVERITY reported issues. 2015-02-26 15:53:42 -05:00
George Bosilca
44d590b8fd Fix a small problem with the handling of requests in MPI_Alltoall. 2015-02-26 15:52:44 -05:00
George Bosilca
3f757bc8cb Add a constructor for mca_coll_base_comm_t. 2015-02-26 15:52:36 -05:00
George Bosilca
d6e69ecab3 Do not preallocate any requests. They are instead automatically 
preallocated on the first collective that needs them.
Remove the ompi_coll_tuned_preallocate_memory_comm_size_limit MCA
parameter.
 2015-02-26 15:52:27 -05:00
George Bosilca
0445670bb9 Fix the automatic handling of communicator associated requests. 
If the array doesn't exist, or if it's size is not adequate then
we reallocate it. Otherwise just keep using the same array of requests.
 2015-02-26 15:52:18 -05:00
George Bosilca
67d01bd8cd Redirect most of the basic module functions to base. 2015-02-26 15:52:00 -05:00
George Bosilca
211f05fb09 Complete the dismantle of the tuned module. 2015-02-26 15:50:55 -05:00
George Bosilca
aa019e239e Rename the base header file containing the prototypes of the collective 
functions.
 2015-02-26 15:50:29 -05:00
George Bosilca
8fbcdf685d Split the tuned framework in two. Move all the functions down in the 
base, so that they can now be used by all modules. Keep the decision
functions in tuned.
 2015-02-26 15:46:13 -05:00
George Bosilca
004f65a865 Fix issue with the error reporting as suggested by Gilles. 2015-02-26 13:01:13 -05:00
Todd Kordenbrock
0cf45df1a0 osc-portals4: fix incomplete free list conversion 2015-02-26 10:53:45 -06:00
Todd Kordenbrock
b2696edeef mtl-portals4: fix incomplete free list conversion 2015-02-26 10:53:45 -06:00
Todd Kordenbrock
c73e4fd98b coll-portals4: fix incomplete free list conversion 2015-02-26 10:53:45 -06:00
Howard Pritchard
918540dd06 Merge pull request #430 from ggouaillardet/issue/404 
romio: always set err_ status in MPIO_DATATYPE_ISCOMMITTED
 2015-02-26 09:02:34 -07:00
Gilles Gouaillardet
05140df1e6 coll/tuned: regression fix 
fix the regression introduced in open-mpi/ompi@004160f8da
 2015-02-26 13:58:06 +09:00
Gilles Gouaillardet
74a46863ca romio: always set err_ status in MPIO_DATATYPE_ISCOMMITTED 
from the ROMIO sources :
/* Any MPI implementation that wishes to follow the thread-safety and
   error reporting features provided by MPICH must implement these
   four functions.  Defining these as empty should not change the behavior
   of correct programs */
the MPIO_DATATYPE_ISCOMMITTED macro now always set err_=0
this is an optimistic approach for Open MPI, but it is likely other upper
layers already checked the datatype was committed.
not setting err_ is incorrect since it can lead to some use of uninitialized
variable.

Fixes open-mpi/ompi#404
 2015-02-26 13:39:05 +09:00
Howard Pritchard
309dfd6d3d Merge pull request #427 from roblatham00/romio-extended-greq-fix 
fix build error if greq extensions not available
 2015-02-25 10:49:00 -07:00
Mike Dubman
a0afb7d96e Merge pull request #424 from miked-mellanox/topic/master_fix_yalla 
fixes issue #414
 2015-02-25 19:01:47 +02:00
Rob Latham
24a6f14257 fix build error if greq extensions not available 
Fall back to immediately completing the non-blocking request if we are
unable to use the extended generalized request routines.
 2015-02-25 10:52:22 -06:00
Jeff Squyres
c3381150de ob1: fix another PERUSE compile error 2015-02-25 05:53:12 -08:00
yosefe
0332ab4d8b Initialize pml_yalla bsend request status. 2015-02-25 15:33:26 +02:00
Nysal Jan K.A
ded408f485 Fix a crash while closing libnbc 
If the free list initialization fails in libnbc_open()
mca_coll_libnbc_component.active_requests remain uninitialized,
resulting in a crash while closing the component
 2015-02-25 17:26:28 +05:30
Nathan Hjelm
0ac2f08460 pml/ob1: fix peruse compile error 
Fixes #416
 2015-02-24 15:39:46 -07:00
Jeff Squyres
a85a392896 Merge pull request #422 from jsquyres/topic/coverity-fixes 
Some Coverity fixes
 2015-02-24 17:00:10 -05:00
Jeff Squyres
1c3cf068a4 nbc ireduce_scatter: ensure to check the correct return code 
This was CID 1196644 and 1196621
 2015-02-24 15:24:11 -05:00
Jeff Squyres
e9980654a8 nbc ireduce_scatter_block: ensure to check the correct return code 
This was CID 1196643 and 1196615
 2015-02-24 15:24:11 -05:00
Jeff Squyres
657d2cf77b io_ompi: check the correct returned value 
This was CID 731273
 2015-02-24 15:24:11 -05:00
Jeff Squyres
b35eb6fe10 nbc ireduce_scatter_block: ensure to check the correct return code 
This was CID 709594 and 709592
 2015-02-24 15:24:10 -05:00
Jeff Squyres
b0acef6f2d nbc_ireduce_scatter: ensure to check the proper return code 
This was CID 709229 and 709224.
 2015-02-24 15:24:10 -05:00
Jeff Squyres
fd815bba22 bcol basesmuma allreduce: remove redundant assignment 
This was CID 1196655 and 1196656
 2015-02-24 15:24:10 -05:00
Jeff Squyres
3758e3a242 topo base: ensure topo is != NULL before acting on it 
This was CID 1041229
 2015-02-24 15:24:09 -05:00
Jeff Squyres
1cf197d771 coll_basic_barrier: guard against opal_hibit() returning -1 
This was CID 1196606 and 1196607
 2015-02-24 15:24:08 -05:00
Nathan Hjelm
6dcdb3daf7 mtl/mxm: fix typo from opal_free_list_t update 2015-02-24 11:43:15 -07:00
igor-ivanov
0f44cdd779 Merge pull request #421 from igor-ivanov/pr/fix-oshmem-coverity 
oshmem: Fix set of coverity issues
 2015-02-24 21:40:06 +04:00
Nathan Hjelm
5ef24000c7 pml/yalla: fix typo in PML_YALLA_FREELIST_INIT 2015-02-24 10:08:54 -07:00
Nathan Hjelm
4443a79af8 mtl/mxm: fix opal_free_list_wait usage error 2015-02-24 10:05:45 -07:00
Nathan Hjelm
5f1254d710 Update code base to use the new opal_free_list_t 
Use of the old ompi_free_list_t and ompi_free_list_item_t is
deprecated. These classes will be removed in a future commit.

This commit updates the entire code base to use opal_free_list_t and
opal_free_list_item_t.

Notes:

OMPI_FREE_LIST_*_MT -> opal_free_list_* (uses opal_using_threads ())

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-02-24 10:05:45 -07:00
Nathan Hjelm
ed78553512 Update opal_free_list_t usage to reflect new class interface. 
Please verify your components have been updated correctly. Keep in
mind that in terms of threading:

OPAL_FREE_LIST_GET -> opal_free_list_get_st
OPAL_FREE_LIST_RETURN -> opal_free_list_return_st

I used the opal_using_threads() variant anytime it appeared multiple
threads could be operating on the free list. If this is not the case
update to _st. If multiple threads are always in use change to _mt.
 2015-02-24 10:05:44 -07:00
Igor Ivanov
3e2dd782ea oshmem: Fix set of coverity issues 
Signed-off-by: Igor Ivanov <Igor.Ivanov@itseez.com>
 2015-02-24 19:03:10 +02:00
Jeff Squyres
b70fa3e2cb osc_sm: Fix valgrind warning 
Many thanks to Lisandro Dalcin for contributing this patch.

Fixes open-mpi/ompi#202.
 2015-02-24 03:36:17 -08:00
Jeff Squyres
c7d8563d8d fcoll_two_phase: fix trivial compiler warning 2015-02-23 08:58:43 -08:00
Howard Pritchard
c9e81b54fb Merge pull request #412 from hppritcha/topic/owner_files 
add owner files to opa/ompi/orte mca directories
 2015-02-23 09:48:20 -07:00
Howard Pritchard
61fb62499a hcoll belongs to MLNX and is active 2015-02-23 09:14:03 -07:00
Mike Dubman
e02121af6f Merge pull request #411 from miked-mellanox/topic/fix_cov 
coverity fixes for yalla
 2015-02-23 12:01:05 +02:00
Mike Dubman
5b3b04b26e mxm: revert coverity fixes 
mxm fails on this commit: 780c93ee57
 2015-02-23 07:52:28 +02:00
Gilles Gouaillardet
004160f8da coll/tuned: silence CID 1269934 2015-02-23 13:45:23 +09:00
Howard Pritchard
bf89131f9e add owner files to opa/ompi/orte mca directories 
This commit adds an owner file in each of the component directories
for each framework.  This allows for a simple script to parse
the contents of the files and generate, among other things, tables
to be used on the project's wiki page.  Currently there are two
"fields" in the file, an owner and a status.  A tool to parse
the files and generate tables for the wiki page will be added
in a subsequent commit.
 2015-02-22 15:10:23 -07:00
Mike Dubman
00d416ba9d yalla: fix coverity errors 
dead code fix
 2015-02-22 13:57:45 +02:00
Igor Ivanov
010dce307a Fix set of coverity issues 
List of CIDs (scan.coverity.com):
oshmem:
1269787, 1269907, 1270161, 1270162, 1270977, 1270978
ompi:
1270170, 1270172, 1270173

Signed-off-by: Igor Ivanov <Igor.Ivanov@itseez.com>
 2015-02-20 17:45:46 +04:00
Ralph Castain
8b625c2756 Ensure the provider name isn't NULL before trying to strdup it 2015-02-19 08:38:23 -08:00
Jeff Squyres
6098b84294 libfabric: pass the appropriate LDFLAGS to libfabric components 
When compiling against an external libfabric, ensure to also pass the
appropriate -L flags so that the compiler/linker can find it.
 2015-02-19 05:35:38 -08:00
Rolf vandeVaart
dbd0064713 Fix bug in CUDA-aware and GDR introduced by refactoring 2015-02-18 17:44:28 -05:00
Gilles Gouaillardet
bda8058f39 coll/tuned: fix memory leaks and misc issues 
as reported by Coverity with CIDs
70132, 70265, 70267, 70268, 70322, 70400, 70580, 70615,
1269934, 1269944, 1269968, 1269982, 1269983
 2015-02-18 16:29:42 +09:00
Nathan Hjelm
030e9bc714 mtl/ofi: fix mca variable usage error 
The fi_fabric function appears to free the provider string passed in
in the fabric_attr. This causes MCA to free an invalid pointer when
the parameter is freed.

References #374
 2015-02-17 12:15:31 -07:00
Gilles Gouaillardet
7dabc7b3ab bml/r2: fix a typo 
reported by Coverity as CID 1270228
 2015-02-17 14:28:17 +09:00
Jeff Squyres
0bb1dfeca9 osc_base_obj_convert: remove unnecessary MEMCHECKER line 
Commit open-mpi/ompi@1a3597aam changed the type of the `convertor`
variable from `ompi_osc_base_convertor_t` (which contained an
`opal_convertor_t`) to an `opal_convertor_t`.  Hence, using memchecker
to ensure that the inner convertor of the `ompi_osc_base_convertor_t`
is considered initialized is now unnecessary.
 2015-02-16 07:27:44 -08:00
Gilles Gouaillardet
0d560ddf77 osc: fix typo 
this typo caused build failure when configure'd with --enable-memchecker
see http://mtt.open-mpi.org/index.php?do_redir=2234
 2015-02-16 10:09:08 +09:00
George Bosilca
a7a4d6335e Various cleanups. 2015-02-15 11:39:09 -05:00
Jeff Squyres
5da123da7c ofi: change mtl_ofi_provider to level 4 
Seems more appropriate; this is a variable that a tuner might want to
actually set.
 2015-02-13 14:46:05 -08:00
Nathan Hjelm
4bf7a207e9 bml/r2: add all rdma btls even if another btl has higher exclusivity 
Background: In order to support atomics each btl needs to provide support
for communicating with self unless the btl module can guarantee global
atomicity. Before this commit bml/r2 discarded any BTL with lower
exclusivity than an existing send btl. This would cause the BML to
discard any btl other than self.

The new behavior is as follows:

 - If an exisiting send btl has higher exclusivity then the btl will not be
   added to the send btl list for the endpoint.

 - If a btl provides RDMA support then it is always added to the rdma btl
   list.

 - bml_btl weight for send btls is now calculated across all send btls.

 - bml_btl weight for rdma btls is now calculated across all rdma btls.

With this change self should still win as the only send btl for loopback
without disqualifying other btls (ugni, openib) for atomic operations.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-02-13 11:46:37 -07:00
Nathan Hjelm
3847025540 pml/ob1: when using btl_get try to register the entire region before attempting to break the get into multiple rdma fragments 
A little background. Historically ob1 always registered the entire memory
region when the RGET protocol was in use. This changed when Mellanox
added support to fragment RGET using the btl_prepare_dst function. Now
that the BTL layer has changed to split out the limits of get/put there
is explicit fragmentation code in ob1. Before this commit the registration
was still done per RGET fragment.

This commit will attempt to register the entire region before creating
RGET fragments. If the registration is successfull then all RGET
fragments will use this registration otherwise they will each attempt
to register their own segment of the receive buffer. If that fails
enough times each fragment will give up and fall back on send/recv.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-02-13 11:46:37 -07:00
Nathan Hjelm
868e10caf2 pml/bfo: ompi ignore until updated for BTL 3.0 interface 
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-02-13 11:46:37 -07:00
Nathan Hjelm
c4a0e02261 pml/ob1: update for BTL 3.0 interface 
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-02-13 11:46:37 -07:00
Nathan Hjelm
9285e2c356 bml: update for BTL 3.0 interface 
This commit brings the bml framework up to date with BTL 3.0 interface.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-02-13 11:46:37 -07:00
Gilles Gouaillardet
b68e1cd579 ompio: fix memory leaks reported by Coverity scan 
and silence CID 72195
 2015-02-13 13:18:44 +09:00
Nathan Hjelm
0e822e03f7 osc/sm: always release the lock on MPI_Unlock 
When a lock was obtained with MPI_MODE_NOCHECK it was not correctly
release on unlock. This is an error.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
 2015-02-12 18:54:22 -07:00
Nathan Hjelm
dd8aaca295 Merge pull request #389 from hjelmn/osc_datatype_fix 
osc/base: fix accumulate on derived datatypes
 2015-02-12 12:01:05 -07:00
Jeff Squyres
27a783b1c3 coll_sm_barrier: remove dead code 
This was CID 1269978.
 2015-02-12 10:24:02 -08:00
Rolf vandeVaart
1f749b0224 Bump priority of coll cuda component so it is higher than self. 
Otherwise, get some odd interactions with coll self in CUDA-aware
builds.
 2015-02-12 12:29:03 -05:00
Nathan Hjelm
1a3597aa93 osc/base: fix accumulate on derived datatypes 
With certain datatypes the opal_datatype_unpack method for performing
the accumulate operation does not work. This commit modifies the
accumulate code in the osc base to use opal_convertor_raw instead.

Fixes #385
 2015-02-11 12:36:30 -07:00
Nathan Hjelm
a2bdfd99a2 osc/pt2pt: do not set active_incoming_frag_signal_count to 0 on fence completion 2015-02-11 12:34:04 -07:00
Jeff Squyres
c368699bde configury: whitespace cleanup 
No code changes.
 2015-02-09 20:49:04 -08:00
Jeff Squyres
0acdf34044 configury: --enable-dist ceased being useful long ago 
The --enable-dist flag is a relic from a prior iteration of the build
system.  It isn't necessary any more.
 2015-02-09 20:49:04 -08:00
Jeff Squyres
f38f2a159b pml_base: whitespace cleanup; no code changes 2015-02-06 11:27:50 -08:00
Jeff Squyres
46a1722dfc pml_base: fix errant show_help message 2015-02-06 11:27:50 -08:00
Howard Pritchard
45e7c7fd60 Merge pull request #378 from hppritcha/topic/mtl_query_cast_fix 
mtl/query: squash compiler warning
 2015-02-06 12:23:33 -07:00
Todd Kordenbrock
b5a0f3d347 osc-portals4: rename OPAL_ASSEMBLY_ARCH values from OMPI_* to OPAL_* 2015-02-04 16:08:55 -06:00
Todd Kordenbrock
762b05bcda mtl-portals4: fix mismatch between format and type-size 2015-02-04 15:35:03 -06:00
Todd Kordenbrock
87759a1b1e mtl-portals4: fix signedness mismatch warning 2015-02-04 15:35:03 -06:00