1
1
Граф коммитов

7395 Коммитов

Автор SHA1 Сообщение Дата
George Bosilca
fd0e1b7261 If we detect an error on a request that has been already released
at the MPI level, we should call abort on MPI_COMM_WORLD.

Fixes ticket #1943.
cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r31982.
2014-06-10 16:24:13 +00:00
Alina Sklarevich
7b8ad47e93 MXM: fix env variable name to hint for thread usage in mxm
reviewed by MikeD
cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31968.
2014-06-09 06:40:32 +00:00
George Bosilca
f5ebd2faeb Fix the Fortran issue identified by Akan Sang Loon. The dist graph
is really special as the weights can be one of the following three
values (NULL, EMPTY or some legal value). As such, we need a complex
if to correctly convert the Fortran value to the corresponding C
value. Thus, always defining the c_ array is the simplest and most
straighforward approach.

cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r31955.
2014-06-05 17:10:48 +00:00
Ralph Castain
248a4b100f Per Artem, we don't know our VPID at the time of getting the initial timing mark, so just get it if timing is requested
This commit was SVN r31951.
2014-06-04 16:28:41 +00:00
Jeff Squyres
6cf49a3c57 Revert r31934: that MPI_Type_f2c.3 is already in the file list.
This commit was SVN r31950.

The following SVN revision numbers were found above:
  r31934 --> open-mpi/ompi@3ed4aaea99
2014-06-04 13:58:42 +00:00
Oscar Vega-Gisbert
c7b229f03e Java - slice methods: set buffer limit to its capacity before change its position
This commit was SVN r31943.
2014-06-03 21:32:27 +00:00
Jeff Squyres
3ed4aaea99 man pages: Add MPI_Type_f2c.3 file into the tarball
cmr=v1.8.2:reviewer=bosilca

This commit was SVN r31934.
2014-06-03 18:28:50 +00:00
Jeff Squyres
87f9f6815f use-mpi-tkr: fix ierr->ierror param names
Issue noted by Walter Spector on the user's mailing list.

Throwing to Craig Rasmussen for review.

cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r31933.
2014-06-03 15:26:02 +00:00
Jeff Squyres
82d4e33510 usnic: let connectivity client timeout if agent never appears
This would be a really, really weird case if it ever happens (i.e.,
you have usnics but the agent process failed somewhere in MPI_INIT
such that the agent never appears), but having an infinite loop
doesn't seem like a good idea.

(does not need to go to v1.8 because v1.8 still uses RML for
communication for the connectivity checker)

This commit was SVN r31932.
2014-06-02 18:14:35 +00:00
Jeff Squyres
af04d60098 usnic: do not enable the connecitivty agent if there are no usnics
Reviewed by Dave Goodell.

cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31931.
2014-06-02 18:09:55 +00:00
Gilles Gouaillardet
1a17a2a960 Fixes MEMCHECKER vs MPI_IN_PLACE in *alltoall*
cmr=v1.8.2:reviewer=bosilca

This commit was SVN r31924.
2014-06-02 04:43:48 +00:00
Gilles Gouaillardet
c930e44bad Fetch info from both opal_dstore_nonpeer and opal_dstore_peer
This conservative fixes tries to fetch info from both
opal_dstore_nonpeer and opal_dstore_peer.
This is required is task A spawns tasks B and C.
B was previously unable to find info from C, this caused locality
info not being set and a hang in coll/ml init.

no CMR is required since v1.8 uses a unique dstore

This commit was SVN r31923.
2014-06-02 02:34:30 +00:00
Gilles Gouaillardet
d1bcd103ac btl/openib : correctly handle eager rdma buffers in mca_btl_openib_del_procs
if eager rdma is used, endpoint reference_count is greater than one.
this commit is a temporary fix that OBJ_RELEASE the endpoint as much as needed.
thought this is likely correct, it can be suboptimal and hence needs to be reviewed

cmr=v1.8.2:reviewer=hjelmn

This commit was SVN r31922.
2014-06-02 02:23:52 +00:00
Ralph Castain
9305756276 Per RFC:
http://www.open-mpi.org/community/lists/devel/2014/05/14838.php

Remove stale component

This commit was SVN r31917.
2014-06-01 17:03:03 +00:00
Ralph Castain
8736a1c138 Per RFC:
http://www.open-mpi.org/community/lists/devel/2014/05/14822.php

Revamp the ORTE global data structures to reduce memory footprint and add new features. Add ability to control/set cpu frequency, though this can only be done if the sys admin has setup the system to support it (or you run as root).

This commit was SVN r31916.
2014-06-01 16:14:10 +00:00
Ralph Castain
cf2c7381d0 Replace the PML barrier with an RTE barrier for now until we can come up with a better solution for connectionless BTLs.
Refs trac:4643

This commit was SVN r31915.

The following Trac tickets were found above:
  Ticket 4643 --> https://svn.open-mpi.org/trac/ompi/ticket/4643
2014-06-01 16:08:56 +00:00
Oscar Vega-Gisbert
cc219218a7 Java: update javadoc's install locations
This commit was SVN r31914.
2014-06-01 15:40:14 +00:00
Alina Sklarevich
f8a664f5ec MXM: generate the jobid only for MXM versions under v2.0.
reviewed by miked
cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31910.
2014-06-01 13:29:24 +00:00
Ralph Castain
1107f9099e Per the RFC issued here:
http://www.open-mpi.org/community/lists/devel/2014/05/14827.php

Refactor PMI support

This commit was SVN r31907.
2014-06-01 04:28:17 +00:00
Jeff Squyres
a1485569b9 If we don't find an OPAL dstore key (via modex), it's not an error --
we just didn't find it.  So don't ORTE_ERROR_LOG it.

cmr=v1.8.2:reviewer=rhc

This commit was SVN r31906.
2014-05-31 12:02:28 +00:00
Rolf vandeVaart
489664b4a9 Remove debug code.
This commit was SVN r31901.
2014-05-28 16:06:16 +00:00
Rolf vandeVaart
570e313c9b Add collective module to handle CUDA aware buffers and reductions. Per RFC sent last week.
Reviewed by bosilca.

This commit was SVN r31894.
2014-05-27 21:24:43 +00:00
Nathan Hjelm
2614dfc4bf bcol/basesmuma: fix remaining memory leaks in basesmuma
We were still leaking 1) file descriptors for data files, and 2) some
control files. I fixed both of these leaks and everything is looking
good. This should fix the bug where we are running out of file
descriptors when running the loop_spawn test. I also too the
opportunity to refactor the code a bit to make the mapping/unmapping
simpler. This should help avoid these sorts of issues in the future.

Depends on #4678

cmr=v1.8.2:reviewer=manjugv

This commit was SVN r31893.
2014-05-27 18:40:41 +00:00
Ralph Castain
f840013b41 Correct typo spotted by Gilles
This commit was SVN r31892.
2014-05-27 17:14:01 +00:00
Mike Dubman
7b05b5c4c2 HCOLL: use proper parameter in progress unregister
fixed by Nadezhda, reviewed by Elena/MikeD

cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31891.
2014-05-26 09:03:30 +00:00
Gilles Gouaillardet
d04db1e213 Fix mmap flags in bcol_basesmuma_smcm_reg_mmap
if in_ptr is NULL, the MAP_FIXED flag cannot be passed to mmap

this caused a hang in topology/cart and topology/sub from ibm
test suite on trunk.

cmr=v1.8.2:reviewer=hjelmn

This commit was SVN r31890.
2014-05-26 07:18:31 +00:00
Gilles Gouaillardet
75616320b5 remove all message output from ompi/communicator/comm_helpers.c
Thanks George for pointing out.

cmr=v1.8.2:reviewer=bosilca:ticket=4676

This commit was SVN r31889.

The following Trac tickets were found above:
  Ticket 4676 --> https://svn.open-mpi.org/trac/ompi/ticket/4676
2014-05-26 01:54:24 +00:00
Gilles Gouaillardet
40f3b849eb Fix argument checks for [i]neighbor_alltoall{v|w}
This fixes a bug introduced in :
 - r31815 (trunk) 
 - r31853 (v1.8 branch)

cmr=v1.8.2:reviewer=bosilca

This commit was SVN r31888.

The following SVN revision numbers were found above:
  r31815 --> open-mpi/ompi@8bafe06c57
  r31853 --> open-mpi/ompi@bff944d766
2014-05-23 08:19:17 +00:00
Gilles Gouaillardet
de15b623b5 do not incrementing/decrementing the opal_progress_event_state.
Per Ralph :
"I noticed that we are incrementing and decrementing the opal_progress_event state.
However, this no longer has any impact whatsoever on the RML as that is running in
the independent ORTE event thread. So all this actually does is impact the MPI layer
by adding an unnecessary overhead."

Thanks Ralph for pointing this :-)

cmr=v1.8.2:reviewer=rhc:ticket=4671

This commit was SVN r31887.

The following Trac tickets were found above:
  Ticket 4671 --> https://svn.open-mpi.org/trac/ompi/ticket/4671
2014-05-23 05:02:31 +00:00
Gilles Gouaillardet
4843909a6b correctly allocate cart in mca_topo_base_cart_sub(...)
since r31716 mca_topo_base_comm_cart_2_2_0_t is an object
and must be allocated/freed with OBJ_NEW/OBJ_RELEASE.

this fixes topology/cart_sub_zero from the ibm test suite.

v1.8 does not use objects, so no cmr for this branch

This commit was SVN r31883.

The following SVN revision numbers were found above:
  r31716 --> open-mpi/ompi@e3df77548d
2014-05-23 03:07:46 +00:00
George Bosilca
66e91f3797 Remove a warning on p.
And correctly handle the reference count from OBJ_NEW in the error
path (thanks Gilles for noticing).

This commit was SVN r31877.
2014-05-22 05:57:21 +00:00
George Bosilca
6ed1ac032e Release the buffer in all error cases and add small code cleanups.
This commit was SVN r31876.
2014-05-22 05:17:35 +00:00
Gilles Gouaillardet
5e9347bfb4 Fix deallocation error in mca_common_sm_rml_info_bcast()
buffer is sent (num_local_procs-1) time, so it should be
OBJ_RETAIN'ed (num_local_procs-2) time

cmr=v1.8.2:reviewer=rhc

This commit was SVN r31874.
2014-05-22 04:53:42 +00:00
Jeff Squyres
af5399ab5b Gah; we need to have a unique barrier ID for the ompi_rte_barrier()
operation.  Ralph will fix shortly.

For the time being, put back the original code...

Refs trac:4669

This commit was SVN r31872.

The following Trac tickets were found above:
  Ticket 4669 --> https://svn.open-mpi.org/trac/ompi/ticket/4669
2014-05-22 00:31:39 +00:00
Jeff Squyres
c860f289ab usnic: re-order teardown sequence now that del_procs() is actually called
Now that the infrastructure is calling BTL del_procs() before the BTL
finalize(), the usnic BTL had to re-order some of its teardown
sequence to avoid assert() failing.

This is part of a larger conversation involving #4669.  Since
MPI_FINALIZE and MPI_COMM_DISCONNECT currently use an
oob/grpcomm-based barrier, the usnic BTL can ''absolutely know'' that
these endpoints and procs will no longer be used.  If the ORTE DPM
goes back to a PML-based barrier, the usnic BTL will need to grow more
complex teardown semantics (a la TCP socket FIN/ACK/FIN_WAIT states).

Refs trac:4669

This commit was SVN r31871.

The following Trac tickets were found above:
  Ticket 4669 --> https://svn.open-mpi.org/trac/ompi/ticket/4669
2014-05-21 23:59:17 +00:00
Jeff Squyres
5aa05f2f1d Potentially temporarily change the barrier in the ORTE DPM component
to be based on grpcomm (i.e., an out-of-band based barrier) rather
than the simplistic PML-based barrier that it currently uses.

This is pending a larger discussion with Nathan and George, but it
will allow the usnic BTL to stop assert()-failing in light of the
recent del_procs() change.

This commit was SVN r31870.
2014-05-21 23:09:01 +00:00
Mike Dubman
fad1063980 MXM: fix warning
reviewed by Yossi

cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31855.
2014-05-21 07:50:05 +00:00
George Bosilca
cc0239d52f Remove unused variables.
This commit was SVN r31846.
2014-05-21 01:34:26 +00:00
Jeff Squyres
b0a6e42f45 pml ob1: use the pre-computed size from the free lists
Based on a suggestion from George on #31806, use the pre-computed
sizes rather than duplicating the computation math (which may change
someday in the future).

cmr=v1.8.2:ticket=trac:4647

This commit was SVN r31841.

The following Trac tickets were found above:
  Ticket 4647 --> https://svn.open-mpi.org/trac/ompi/ticket/4647
2014-05-20 20:32:25 +00:00
George Bosilca
db9660264e Update the error message to pinpoint the right location.
Thanks Tim.

This commit was SVN r31839.
2014-05-20 20:08:42 +00:00
George Bosilca
685f051557 Move the allocator initialization from open to init. This clean
a memory leak. Similar changes shuld be applied to all the 
other PML that are copies of OB1. This patch is related to
#4653.

This commit was SVN r31838.
2014-05-20 19:34:18 +00:00
Gilles Gouaillardet
baf532087a Fix mca_coll_basic_alltoallw_inter()
Avoid sending/receiving zero size messages in order to be compliant
with the top-level modification

cmr=v1.8.2:ticket=4651:reviewer=bosilca

This commit was SVN r31836.

The following Trac tickets were found above:
  Ticket 4651 --> https://svn.open-mpi.org/trac/ompi/ticket/4651
2014-05-20 09:22:57 +00:00
George Bosilca
137874ec4d In fact btl_eager and btl_dma will just vanish upon destruction of the
bml_endpoint. No need to clean them bfore.

This commit was SVN r31835.
2014-05-20 08:53:22 +00:00
George Bosilca
85e3caaa17 Handle the del_procs correctly. The btl_send is the complete list
of existing BTL fo an endpoint, all the others are just partial list.
Thus, all the cleaning should first be done in the btl_send array,
and them in the other arrays (btl_eager and btl_rdma).

This commit was SVN r31834.
2014-05-20 08:46:57 +00:00
George Bosilca
1647664c43 Show the unreacheable message for the first unreacheable proc
and then stop.

This commit was SVN r31833.
2014-05-20 08:40:32 +00:00
Gilles Gouaillardet
c496131eef Fixes *alltoall* collectives at top
- fix bugs
 - silent warnings

cmr=v1.8.2:ticket=4651:reviewer=bosilca

This commit was SVN r31831.

The following Trac tickets were found above:
  Ticket 4651 --> https://svn.open-mpi.org/trac/ompi/ticket/4651
2014-05-20 05:32:57 +00:00
Gilles Gouaillardet
ef4548a215 bml/r2 : fix mca_bml_r2_del_procs()
cmr=v1.8.2:reviewer=hjelmn:ticket=trac:4645

This commit was SVN r31830.

The following SVN revision numbers were found above:
  r2 --> open-mpi/ompi@58fdc18855

The following Trac tickets were found above:
  Ticket 4645 --> https://svn.open-mpi.org/trac/ompi/ticket/4645
2014-05-20 02:55:47 +00:00
Nathan Hjelm
27d3e1ca25 bml/r2: fix a problem identified by Gilles
This commit fixes two issues:

 - This intent of the code @ bml_r2.c:486 is to prevent calling the
   btl_del_procs more than once for a given proc. Gilles correctly
   identified there was a problem in this code but r31786 we not the
   correct fix.

 - Fix a segmentation fault in r2 finalize revealed by the fact we
   actually call del_procs now.

cmr=v1.8.2:reviewer=ggouaillardet:ticket=trac:4645

This commit was SVN r31829.

The following SVN revision numbers were found above:
  r2 --> open-mpi/ompi@58fdc18855
  r31786 --> open-mpi/ompi@fc96b0a7b8

The following Trac tickets were found above:
  Ticket 4645 --> https://svn.open-mpi.org/trac/ompi/ticket/4645
2014-05-19 20:22:34 +00:00
Nathan Hjelm
a1d5ce0893 pml/ob1: as per past RFC bring the inline send optimization to
MPI_Isend.

I filed an RFC for this optimization some time back. It is a
relatively simple optimization. If the data associated with an
MPI_Isend can be put on the wire without allocating an MPI_Request
then do so. In this case we can legally return omp_request_empty
which will correctly indicate that the request is complete and that is
was not cancelled (these are the only requirements on send requests).

cmr=v1.8.3:reviewer=bosilca

This commit was SVN r31828.
2014-05-19 19:34:59 +00:00
Nathan Hjelm
22e59b056a bcol/basesmuma: fix leak in basesmuma code
Basesmuma was vallocing space for control data then mmapping over that
data. Nothing in the code suggests any need for mmapping a specific
address so I did the following to remove the leak:

 - Removed the valloc of the buffer space

 - ftruncate the mmaped file to ensure there is sufficient memory to
   allocate space for the control data.

Ideally this code should be using opal/shmem but that is a larger
change. Keeping it simple for now.

cmr=v1.8.2:reviewer=manjugv

This commit was SVN r31822.
2014-05-19 15:21:58 +00:00