1
1
Граф коммитов

18642 Коммитов

Автор SHA1 Сообщение Дата
Nathan Hjelm
c5596548b2 MPI-3: Add support for neighborhood collectives
Blocking versions are simple linear algorithms implemented in coll/basic. Non-
blocking versions are from libnbc 1.1.1. All algorithms have been tested with
simple test cases.

cmr=v1.7.4:reviewer=jsquyres

This commit was SVN r29265.
2013-09-26 21:55:08 +00:00
Dave Goodell
a42fa78da7 usnic: SEGV in OSU benchmarks
Prevent frag from being freed out from under us in the case
the PML callback routine calls usnic_free().  We accomplish this
by delaying decrement of sf_bytes_to_ack until after the callback is
performed, since sf_bytes_to_ack == 0 is condition of freeing the frag.

Fixes Cisco bug CSCuj45094.

Authored-by: Reese Faucette <rfaucett@cisco.com>

cmr=v1.7.3

This commit was SVN r29264.
2013-09-26 21:48:04 +00:00
Jeff Squyres
e4e3e411fc Next generation of MPI Java bindings.
Includes all MPI functions supported by Open MPI, including MPI-3
functions (as of about 2 weeks ago).  Many changes compared to the
prior generation of Java bindings; not much is left from the prior
generation, actually.  The changes include (but are not limited to):

 * Add support for more than just a subset of MPI-1 functions
 * Use typical Java case for symbol names
 * Support Java Direct buffers (giving darn-near "native C"
   performance)
 * Support "type struct" better than the prior generation
 * Make more of an effort for the Java bindings to be a thin layer
   over the back-end C bindings
 * ...and more

A proper README with more information about what is supported, how to
use these bindings, etc. will be committed shortly.

This commit was SVN r29263.
2013-09-26 21:44:39 +00:00
Jeff Squyres
6c53711ac8 Provide Java MPI_Op callbacks via an intercept routine (just like how
we do MPI::Op C++ callbacks).

This commit was SVN r29262.
2013-09-26 21:36:44 +00:00
Jeff Squyres
7941c81caa The TargetConditionals.h check is specific to Java -- move it to
ompi_setup_java.m4.

This commit was SVN r29261.
2013-09-26 21:34:00 +00:00
Mike Dubman
7c6ff00da5 Add caching of FCA communicators
developed by Dinar, reviewed by miked/yossi.
cmr:v1.7.3:reviewer=jsquyres:subject=add caching of FCA communicators.

This commit was SVN r29256.
2013-09-26 17:48:07 +00:00
Rolf vandeVaart
d67e3077f5 Add a check for the CUDA 6.0 version of the cuda.h header file.
This commit was SVN r29250.
2013-09-26 12:46:06 +00:00
Joshua Ladd
ba17053470 Fixing OSHMEM compiler warnings when --enable-mpi-thread-multiple is set
This commit was SVN r29249.
2013-09-26 01:21:17 +00:00
Ralph Castain
dee8336f68 Do not use modex recv to fetch the locality as this will automatically force retrieval of hostnames, which we are trying to avoid. Instead, use the database API to fetch that info.
cmr:v1.7.3:reviewer=hjelmn

This commit was SVN r29248.
2013-09-25 21:36:25 +00:00
Joshua Ladd
d1af0e2041 Removing a silly check in the critical path
This commit was SVN r29247.
2013-09-25 21:34:57 +00:00
Ralph Castain
6522963b9c Flag that a daemon has been launched when it reports back to the HNP so we avoid re-launching it on spawns against dynamic allocations
cmr:v1.7.3:reviewer=jsquyres

This commit was SVN r29245.
2013-09-25 16:58:19 +00:00
Joshua Ladd
82e092db1b Adding interface changes in hcoll component to support non-blocking collectives in libhcoll. This was added by Elena Elkina and reviewed by Josh Ladd.
cmr:v1.7.3:reviewer=jladd:subject=Add support for non-blocking collectives in hcoll

This commit was SVN r29244.
2013-09-25 16:14:59 +00:00
Ralph Castain
9aeba777fa Ensure we don't enter into an infinite loop looking for the PML modex key if it isn't present. The PMI implementation will load ALL modex keys when the first key is queried, so the hash db component can safely return "not found" if a subsequent key isn't present. The PML modex_recv needs to assume everything is okay if the modex recv fails to return a value.
cmr:v1.7.3:reviewer=jladd:subject=Prevent infinite loop when PML modex not found

This commit was SVN r29243.
2013-09-25 16:04:00 +00:00
Joshua Ladd
008a2af2d8 Cleaning up a bit
This commit was SVN r29240.
2013-09-24 22:09:36 +00:00
Joshua Ladd
ac421fc1b9 Fixing OSHMEM 64-bit compiler warnings/picky errors in MXM enabled components.
This commit was SVN r29239.
2013-09-24 21:48:57 +00:00
Jeff Squyres
eee8e9bb92 Update gitignore per the latest values of svn:ignore
This commit was SVN r29238.
2013-09-24 20:41:45 +00:00
Rolf vandeVaart
667c66941b Remove redundant (and possibly erroneous) alignment code in rcache. It is already handled by users of the rcache.
This was per RFC http://www.open-mpi.org/community/lists/devel/2013/09/12927.php and discussed in developers meeting.

This commit was SVN r29233.
2013-09-24 17:23:50 +00:00
Rolf vandeVaart
3b5e0736a3 Adjust verbosity levels upward.
This commit was SVN r29232.
2013-09-24 14:35:48 +00:00
Ralph Castain
23c8848157 Only connect the first time thru the Torque launch, remove stale code
cmr:v1.7.3:reviewer=jsquyres

This commit was SVN r29227.
2013-09-22 23:53:57 +00:00
Ralph Castain
63da76ad5f Silence warnings about pointer casting
This commit was SVN r29226.
2013-09-22 19:21:29 +00:00
Ralph Castain
400c68ed0f Fix a segfault when a topology file is given to use in place of the one detected by mpirun itself. In that situation, the rmaps framework replaces the opal_hwloc_topology structure - but since that occurs *after* mpirun has set the node->topology field, we lose that definition. So don't set the node->topology field until after the rmaps framework has been opened.
Does not need to go to 1.7 branch as that ordering is different.

-This line, and those below, will be ignored--

M    orte/mca/ess/hnp/ess_hnp_module.c

This commit was SVN r29225.
2013-09-21 19:47:41 +00:00
Nathan Hjelm
01839db11b MCA/base: When encounter a duplicate file value don't free the filename.
Stale code.

cmr=v1.7.3:reviewer=rhc

This commit was SVN r29224.
2013-09-21 18:53:36 +00:00
Mike Dubman
08cc4ebecf fix for solaris which does not have bzero
cmr:v1.7.3:reviewer=jsquyres

This commit was SVN r29223.
2013-09-21 07:26:02 +00:00
Nathan Hjelm
bc31773523 Fix bug in db/pmi when a stored byte object has a NULL pointer.
cmr=v1.7.3:reviewer=samuel

This commit was SVN r29215.
2013-09-20 15:38:36 +00:00
Jeff Squyres
758cd25fff Move the MCA / MPI_T level of the LAMA component down to 5 (from 9).
This commit was SVN r29214.
2013-09-20 15:23:27 +00:00
Ralph Castain
34fbec1f49 Sadly, the connection priorities being defined at time of variable instantiation were being overridden just before registering the param. Thus, changes people made to the relative priority of the cpc methods were being lost. Fix it be removing the duplicate initializiation, letting the value defined at instantiation be the one actually used.
cmr:v1.7.4:reviewer=hjelmn

This commit was SVN r29212.
2013-09-19 19:45:00 +00:00
Ralph Castain
fe9a744289 Add missing include file on Solaris environments
Refs trac:3763

This commit was SVN r29211.

The following Trac tickets were found above:
  Ticket 3763 --> https://svn.open-mpi.org/trac/ompi/ticket/3763
2013-09-19 18:43:13 +00:00
Ralph Castain
7bc20866fd C standard stipulates that we have to cast the function to another of the same type to avoid unexpected behavior. We aren't using the function in this case, but Nick correctly points out that we should follow the standard regardless.
Refs trac:3755

This commit was SVN r29210.

The following Trac tickets were found above:
  Ticket 3755 --> https://svn.open-mpi.org/trac/ompi/ticket/3755
2013-09-19 18:42:21 +00:00
Rolf vandeVaart
804545278f Per discussion on devel list, delete unused registration cache.
http://www.open-mpi.org/community/lists/devel/2013/08/12803.php

This component was .ompi_ignored on December 17, 2006 by gleb.  Now, it is time for it go....

This commit was SVN r29209.
2013-09-18 21:22:34 +00:00
Rolf vandeVaart
c9a33fad83 Fix some tabs. Add optional messsage to dump. Some minor format change to dump function.
This commit was SVN r29208.
2013-09-18 21:08:15 +00:00
George Bosilca
273d66d0f2 The MPI_Intercomm_create test was broken, as the remote peer was
always considered as being 1 (instead of count).

This commit was SVN r29207.
2013-09-18 16:47:54 +00:00
George Bosilca
85db48df0e Identification, tab vs. space.
This commit was SVN r29206.
2013-09-18 16:45:00 +00:00
Ralph Castain
f051500166 Sadly, there is no RTE-agnostic way to prune the modex entries, so we must send them all.
Refs trac:3766

This commit was SVN r29204.

The following Trac tickets were found above:
  Ticket 3766 --> https://svn.open-mpi.org/trac/ompi/ticket/3766
2013-09-18 14:09:23 +00:00
Joshua Ladd
d81186df9b More OSHMEM fixes for Sun C 5.12 compiler
This commit was SVN r29203.
2013-09-18 13:54:47 +00:00
Mike Dubman
2a5c342587 Modifications that are necessary in order to meet latest libhcoll API.
cmr:v1.7.3:reviewer=jladd

This commit was SVN r29202.
2013-09-18 12:22:02 +00:00
Ralph Castain
7de493fc02 Silence a warning about an address that can never be NULL - libevent needs to deal with the situation where the user may have compiled the code on a system where this function is present, but executes it on one where it isn't. Thus, a compile-time test isn't adequate.
Pushed upstream.

cmr:v1.7.3:reviewer=jsquyres

This commit was SVN r29201.
2013-09-18 02:03:01 +00:00
Ralph Castain
865a7028f8 Per patch from George, with a few minor cleanups. Correctly address the complete exchange of required wireup information in Intercomm_create so all procs in the resulting communicator know how to talk to each other.
Refs trac:29166

This commit was SVN r29200.

The following Trac tickets were found above:
  Ticket 29166 --> https://svn.open-mpi.org/trac/ompi/ticket/29166
2013-09-18 02:01:30 +00:00
Ralph Castain
99611ac1d2 Revert r29166 in favor of a better solution from George
This commit was SVN r29199.

The following SVN revision numbers were found above:
  r29166 --> open-mpi/ompi@497c7e6abb
2013-09-18 01:41:26 +00:00
George Bosilca
55273f1c98 Cleanup spaces, nothing else.
This commit was SVN r29197.
2013-09-18 00:07:58 +00:00
George Bosilca
9e6c3c0646 Save the error code.
This commit was SVN r29196.
2013-09-17 23:50:11 +00:00
George Bosilca
7b319a101d Fix the case where we build without Fortran support.
This commit was SVN r29194.
2013-09-17 20:45:46 +00:00
Nathan Hjelm
7929fb9dea Cleanup complex datatypes and update datatypes and operator code to use C99.
This commit changes the underlying opal complex datatypes to match the
C99 types: float _Complex, double _Complex, and long double _Complex. The
fortran and C++ types now are aliases to these basic types instead of
structure types. The operators in ompi/mca/op/base now work on only the
C99 types and the fortran types use these operators if the fortran type
matches a C complex type (this should almost always be the case.)

C99 is not is use in both the datatype and operator code and should make
the code both cleaner and much less fragile.

This commit was SVN r29193.
2013-09-17 17:49:42 +00:00
Joshua Ladd
601f9e37cd Fixing FORTRAN bindings in OSHMEM
This commit was SVN r29190.
2013-09-17 16:39:56 +00:00
Rolf vandeVaart
440632b57f Add a function that will dump out the contents of the memory registration cache.
Useful for debugging any rcache issues.

This commit was SVN r29189.
2013-09-17 15:40:32 +00:00
Jeff Squyres
74d1278f48 btl_usnic_util.c:ompi_btl_usnic_util_abort() also passes in the strerror().
This commit was SVN r29188.
2013-09-17 12:35:51 +00:00
George Bosilca
5f686a90d0 Fix several issues regarding MPI_IN_PLACE and different flavors
of MPI_Alltoall.
 - add support for MPI_IN_PLACE in the self collective component.
 - fix the extent usage in the tuned collective component.
 - correctly use the peer counts instead of local - add support for MPI_IN_PLACE in the self collective component.
 - fix the extent usage in the tuned collective component.
 - correctly use the peer counts instead of local.
Thanks to Fujitsu for the patch.

This commit was SVN r29187.
2013-09-17 11:35:18 +00:00
Reese Faucette
8f235e6977 usnic: wrong SG entry used to compute length for small put()s
This commit was SVN r29186.
2013-09-17 08:18:02 +00:00
Reese Faucette
651d61f1a3 Clean up debugging logging a bit.
MSGDEBUG2 now means "print a one-liner for all PML calls into BTL, and
also when BTL calls PML with a recv completion (not send completions)"
MSGDEBUG1 means print more internal gory detail
MSGDEBUG is gone, replaced by MSGDEBUG1

In the process also found that PUT_DEST style fragments could
potentially be leaked in usnic_free() since send_fragment tests were
being applied to see if it was eligible to be freed.

This commit was SVN r29185.
2013-09-17 07:29:40 +00:00
Reese Faucette
f35d9b50e3 Cisco CSCuj22803: fixes for Bsend
changes required to support MPI_Bsend().  Introduces concept of
attaching a buffer to a large segment that the PML can scribble into and
we will send from.  The reason we don't use a pinned buffer and send
directly from that is that usnic_verbs does not (yes) support num_sge>1
for regular sends.  This means the data gets copied twice, but that is
unavoidable.

changed the logic in handle_large_send to be more sensible

Incorporated David's review comments

This commit was SVN r29184.
2013-09-17 07:27:39 +00:00
Reese Faucette
25b5c84d0f Cisco CSCuj13135: Data corruption in MPI_Bsend_ator_c
Do not assume that the "size" passed to alloc_send() will be the same as
the size of the message the resulting fragment will hold when
usnic_send() is called.  This means usnic_send()/usnic_put() can never
trust any pre-computed size values, and are only allowed to look at the
lengths and pointers of the elements in the desc SG list.

This commit was SVN r29183.
2013-09-17 07:25:05 +00:00