1
1
Граф коммитов

283 Коммитов

Автор SHA1 Сообщение Дата
Nathan Hjelm
033894b493 Merge pull request #541 from hjelmn/c99_components
C99 component initialization
2015-04-20 10:45:39 -06:00
Devendar Bureddy
3dbd95fa73 OSHMEM: enable mpi collective by default 2015-04-20 19:39:36 +03:00
Nathan Hjelm
c4a61969c0 oshmem: use C99 subobject naming for component initialization
This commit helps future-proof oshmem components by initializing each
component member by name.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-04-18 10:29:58 -06:00
Nathan Hjelm
3436f2917d Merge pull request #449 from hjelmn/mca_base_update
mca/base update
2015-04-16 08:41:48 -06:00
Howard Pritchard
18039b34b4 ompi/java: better error message if dlopen fails
The error message emitted by ompi/java when dlopen
fails is misleading and not very informative.

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2015-04-06 13:35:09 -06:00
Nathan Hjelm
b68d66bb9b MCA: Add the project/project version to the MCA base component
This commit adds support for project_framework_component_* parameter
matching. This is the first step in allowing the same framework name
in multiple projects. This change also bumps the MCA component version
to 2.1.0.

All master frameworks have been updated to use the new component
versioning macro. An mca.h has been added to each project to add a
project specific versioning macro of the form
PROJECT_MCA_VERSION_2_1_0.

Signed-off-by: Nathan Hjelm <hjelmn@me.com>
2015-03-27 10:59:04 -06:00
Jeff Squyres
a026456bef (orte|ompi|oshmem)*info tools: convert to opal_dl interface
Noe that this commit removes option:lt_dladvise from the various
"info" tools output.  This technically breaks our CLI "ABI" because
we're not deprecating it / replacing it with an alias to some other
"into" tool output.

Although the dl/libltdl component contains an "have_lt_dladvise" MCA
var that contains the same information, the "option:lt_dladvise"
output from the various "info" tools is *not* an MCA var, and
therefore we can't alias it.  So it just has to die.
2015-03-09 08:18:13 -07:00
Jeff Squyres
8d04215741 coll: trivial spelling fix
s/Algoritm/Algorithm/g
2015-02-27 18:20:17 -08:00
Alina Sklarevich
e4c4e7df5e Fix the calls to ibv_fork_init and remove btl_openib_want_fork_support.
In order to have an effect, ibv_fork_init should be called in the
beginning of the verbs initialization flow - before the calls to the
ibv_create_qp and ibv_create_cq verbs.
These functions are called from the oob/ud code and by the time the
other verbs components (btl openib, pml yalla, ...) call ibv_fork_init,
it's too late. This commit forces the call to ibv_fork_init (if it's
requested) right at the beginning of all the components that are using
verbs.
(ibv_fork_init() can be safely called multiple times)

This commit also removes the btl_openib_want_fork_support mca parameter
and adds a new mca parameter instead - opal_verbs_want_fork_support.
Through this new parameter, fork support may be requested for ALL
components.
The default value for this parameter is set to 1.

Before this commit the btl_openib_want_fork_support parameter didn't
provide fork support for the openib btl if its value was set to 1.
(because when openib called ibv_fork_init, it was already after the
calls to ibv_create_* in oob/ud and thereofre it failed).
2015-02-25 10:58:50 +02:00
Jeff Squyres
bd99b741d2 java: update to new mpool API 2015-02-24 13:31:23 -08:00
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
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
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
15be948d79 wrappers: *_EXTRA_INCLUDES does not exist any more
There were a few places where *_EXTRA_INCLUDES (and derivates) were
still being used.  This commit removes all of them.
2015-02-20 08:43:25 -08:00
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
Igor Ivanov
426d1ce146 oshmem: Fix set of coverity issues
List of CIDs (scan.coverity.com):
1269721, 1269725, 1269787, 1269907, 1269909, 1269910, 1269911, 1269912,
1269959, 1269960, 1269984, 1269985, 1270136, 1270157, 1269845, 1269875,
1269876, 1269877, 1269878, 1269884, 1269885, 1270161, 1270162, 1270175,
1269734, 1269739, 1269742, 1269743

Signed-off-by: Igor Ivanov <Igor.Ivanov@itseez.com>
2015-02-19 23:00:17 +04:00
Jeff Squyres
b27ba475ef oshmem java: really abort in bad situations
The previous condition just called assert(), which is a no-op in
non-debug builds.  Change it to print a message and then call abort()
to really actually above.

This was CID 1270155.
2015-02-14 07:07:01 -08:00
Nathan Hjelm
16ae7d97d1 spml/yoda: update for BTL 3.0 interface
This commit make spml/yoda compatible with BTL 3.0. This is meant as a
starting point only. More work will be needed to make optimial use of
the new interface.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-02-13 11:46:38 -07:00
Jeff Squyres
1e58920b4d *info param.c: use stack string buffers
Coverity identified that we treated the possibility that one of the
message buffers could be NULL in some places (because strdup() could
fail), but not in others.

So just use stack buffers that will never be NULL.

This was CID 1269914.
2015-02-12 10:24:02 -08:00
Jeff Squyres
3ac1d0dae5 *-info: add "lt_dladvise support" lines 2015-02-11 12:25:20 -08:00
Mike Dubman
6611f4ce38 OSHMEM: fix warnings 2015-02-09 20:49:03 -08:00
Bert Wesarg
83ab7e2d32 build: silent LN_S also for the SHMEM Fortran profile part
Effectively applies 173c046617 to a later
added Makefile.am.
2015-01-22 21:45:51 +01:00
Bert Wesarg
0d0a754c42 Remove VampirTrace. 2015-01-22 08:08:07 +01:00
Jeff Squyres
f4693c9afd oshmem: fix compiler warning
Compiler warning that ret was used before it was assigned.  Since this
part of the code is not performance-critical, just throw in an extra
assignment and be done with it.
2015-01-16 12:48:36 -08:00
Gilles Gouaillardet
661c35ca67 cleanup dead code caused by the removal of the --with-threads configure option 2015-01-16 19:13:59 +09:00
Mike Dubman
54a072caaa OSHMEM: fix infinite recursion and stack size violation
send reply before posting the receive request again to limit the recursion size to
number of receive requests.
send can call opal_progress which calls this function again. If recv req is started
stack size will be proportional to number of job ranks.
2015-01-04 16:31:19 +02:00
Gilles Gouaillardet
27aec2ef5b configury: disable f08 fortran bindings if the compiler does
not support c_funloc with TS 29113 subclause 8.1 aka
removed restrictions on ISO_C_BINDING module procedures.
2014-12-17 17:35:45 +09:00
Alex Mikheev
c76261da07 OSHMEM: atomic mxm: fix mkey conversion
Correctly return mxm_empty_mem_key when shmem mkey is empty
2014-12-16 16:34:42 +02:00
Alex Mikheev
71ebbca26d OSHMEM: spml ikrit: fix spelling in help file 2014-12-16 16:18:38 +02:00
Alex Mikheev
3f7ed56548 OSHMEM: spml ikrit: fix mxm disconnect flow
Add out of band barrier before performing mxm disconnect.
It will make sure that every pe is ready to disconnect. Otherwise
bad things may happen.
2014-12-16 15:07:17 +02:00
Ralph Castain
66a860c1e2 Okay, okay - use the oshmem word here 2014-12-12 08:42:54 -08:00
Ralph Castain
064a241145 Don't install the Java shmem tools if we aren't building Java support. Thanks to Paul Hargrove for noticing. 2014-12-12 08:25:38 -08:00
Alex Mikheev
428add390e OSHMEM: spml ikrit: add skew to connect/disconnect
Each pe connects/disconnects starting from itself instead of pe=0. This
will distribute network traffic in a more friendly way.
2014-12-03 15:36:45 +02:00
Alex Mikheev
8de50d8420 OSHMEM: spml ikrit: add call to mxm_mq_destroy()
Make valgrind happy by calling mxm_mq_destroy() on module
close.
2014-12-01 12:36:46 +02:00
Nathan Hjelm
d495d49b1c Merge pull request #273 from open-mpi/topic/yoda_rdma_flags
OSHMEM: spml yoda: use flags to check if btl is RDMA capable
2014-11-16 12:04:04 -07:00
Alex Mikheev
fbb9dc5b1e OSHMEM: spml ikrit valgrind fix
always initialize request flags
2014-11-16 17:24:16 +02:00
Alex Mikheev
3443c1d5e5 OSHMEM: spml yoda: use flags to check if btl is RDMA capable 2014-11-16 17:20:20 +02:00
Gilles Gouaillardet
2177f9ec3e fix missing copyright, no code change 2014-11-13 14:56:09 +09:00
Gilles Gouaillardet
cd6e3ecb07 oshmem/yoda: fix a typo in mca_spml_yoda_get_completion 2014-11-13 14:53:32 +09:00
Ralph Castain
780c93ee57 Per the PR and discussion on today's telecon, extend the process name definition as a two-field struct of uint32_t's down to the OPAL layer. This resolves issues created by prior commits that impacted both heterogeneous and SPARC support. This also simplifies the OMPI code base by removing the need for frequent memcpy's when transitioning between the OMPI/ORTE layers and OPAL.
We recognize that this means other users of OPAL will need to "wrap" the opal_process_name_t if they desire to abstract it in some fashion. This is regrettable, and we are looking at possible alternatives that might mitigate that requirement. Meantime, however, we have to put the needs of the OMPI community first, and are taking this step to restore hetero and SPARC support.
2014-11-11 17:00:42 -08:00
Alex Mikheev
097b469f61 OSHMEM: sshmem verbs: fix shared_mr detection
It seems that 5ce2f10067
changed default flag values but it did not modify detection code.
2014-11-10 13:34:04 +02:00
Alex Mikheev
7327b13823 OSHMEM: sshmem mmap: removed unused help topics 2014-11-05 16:39:20 +02:00
Alex Mikheev
1f2ab43ba9 OSHMEM: spml ikrit: remove empty lines in helpfile 2014-11-04 11:26:09 +02:00
Alex Mikheev
d06fb85350 OSHMEM: fixes 'improve mxm transport sanity check'
The code that actually checks for valid transport combos
somehow did not make it to the original commit:

74ab30b738
2014-11-04 11:07:22 +02:00
Alex Mikheev
e1cf6f37ba OSHMEM: spml ikrit: disable rdmap op DCI pool
Instead use single pool for both rdma and send receive ops.
2014-11-03 10:01:07 +02:00
Jeff Squyres
6b1acf77cc configury: add OPAL/ORTE lib prefixes in oshmem wrapper data files
These 2 files were missed in
open-mpi/ompi@894acb0aa8.
2014-10-28 17:17:21 -07:00
Gilles Gouaillardet
eef7590e58 wrappers: add the $(EXEEXT) extension to the installed symbolic links 2014-10-28 16:42:51 +09:00
Mike Dubman
5ce2f10067 OSHMEM: integrate review comments from open-mpi/ompi-release#7 2014-10-24 17:21:46 +03:00
Jeff Squyres
c22e1ae33b configury: new OPAL_SET_LIB_PREFIX/ORTE_SET_LIB_PREFIX macros
These two macros set the prefix for the OPAL and ORTE libraries,
respectively.  Specifically, the OPAL library will be named
libPREFIXopen-pal.la and the ORTE library will be named
libPREFIXopen-rte.la.

These macros must be called, even if the prefix argument is empty.

The intent is that Open MPI will call these macros with an empty
prefix, but other projects (such as ORCM) will call these macros with
a non-empty prefix.  For example, ORCM libraries can be named
liborcm-open-pal.la and liborcm-open-rte.la.

This scheme is necessary to allow running Open MPI applications under
systems that use their own versions of ORTE and OPAL.  For example,
when running MPI applications under ORTE, if the ORTE and OPAL
libraries between OMPI and ORCM are not identical (which, because they
are released at different times, are likely to be different), we need
to ensure that the OMPI applications link against their ORTE and OPAL
libraries, but the ORCM executables link against their ORTE and OPAL
libraries.
2014-10-22 10:32:19 -07:00
Jeff Squyres
01fd96bfa5 Revert "Provide a mechanism by which an upstream project can rename
the OPAL and ORTE libraries. This is required by projects such as ORCM
that have their own ORTE and OPAL libraries in order to avoid library
confusion. By renaming their version of the libraries, the OMPI
applications can correctly dynamically load the correct one for their
build."

This reverts commit 63f619f871.
2014-10-22 10:32:11 -07:00