1
1
Граф коммитов

8034 Коммитов

Автор SHA1 Сообщение Дата
George Bosilca
3af8dfd3e2 Fix a overwrite of the args buffer identified by Lisandro Dalcin. 2015-05-07 09:50:39 -04:00
Jeff Squyres
691b4ec1e5 romio314: whitespace cleanup
No code changes
2015-05-05 06:23:59 -07:00
Jeff Squyres
db257cdbc0 romio314: adhere to the prefix rule
Rename all files and symbols from "io_romio" to "io_romio314".  This
fixes --disable-dlopen builds (because they were missing
the mca_io_romio314_component symbol).
2015-05-05 06:23:59 -07:00
Devendar Bureddy
88eb1fa936 HCOLL: refactoring hcoll_init 2015-05-04 22:03:36 +03:00
Jeff Squyres
8127c24f30 romio314/Makefile.am: whitespace cleanup
No code changes.
2015-05-04 07:20:11 -07:00
Jeff Squyres
332bca7183 romio314/Makefile.am: name the component library properly 2015-05-04 07:20:11 -07:00
Howard Pritchard
74089f5e3a Merge pull request #558 from ggouaillardet/refresh/romio314
Refresh/romio314
2015-05-01 14:23:43 -06:00
Gilles Gouaillardet
ef4b6203a4 Correctly unpack datatype.
Alignment requirements were relaxed in open-mpi/ompi@33a3ace874
so correctly handle this when unpacking a datatype.
2015-05-01 17:08:15 +09:00
Gilles Gouaillardet
34128c1cad mca_topo_base_dist_graph_neighbors: do not fail if legitimate parameters are provided.
Per the MPI 3.0 standard (chapter 7, page 310) :
"If maxindegree or maxoutdegree is smaller than the numbers returned by
MPI_DIST_GRAPH_NEIGHBOR_COUNT, then only the first part of the full list is returned."
2015-05-01 13:31:05 +09:00
Gilles Gouaillardet
96e3cbe8fc Remove an incorrect assert.
Alignment requirements were relaxed in open-mpi/ompi@33a3ace874
and made a previous alignment check incorrect.
2015-05-01 12:53:12 +09:00
George Bosilca
33a3ace874 Minimize the alignments. We only do it when we need to pack
data that must be aligned (aka the displacement). All other
cases do not require special alignments, and are treated
normally.
Fix the comment regarding the alignment requirements.
2015-04-30 22:06:50 -04:00
George Bosilca
015d3f56cf Fix the INDEXED_BLOCK issue identified by IBM. 2015-04-30 14:43:19 -04:00
Gilles Gouaillardet
6b3126e69e ROMIO 3.1.4 refresh: add refresh notes 2015-04-30 19:02:20 +09:00
Gilles Gouaillardet
e1b6ab4f1d ROMIO 3.1.4 refresh: remove old romio 2015-04-30 19:01:23 +09:00
Gilles Gouaillardet
85e77079b4 ROMIO 3.1.4 refresh: use romio from mpich 3.1.4 2015-04-30 19:00:50 +09:00
Gilles Gouaillardet
92f6c7c1e2 ROMIO 3.1.4 refresh: apply post romio-3.1.4 patches 2015-04-30 18:56:53 +09:00
Gilles Gouaillardet
6400bc75ab ROMIO 3.1.4 refresh: patch romio for Open MPI 2015-04-30 18:53:55 +09:00
Gilles Gouaillardet
eacd434a02 ROMIO 3.1.4 refresh: import romio from mpich 3.1.4 tarball 2015-04-30 18:53:03 +09:00
Gilles Gouaillardet
e2e91142d5 ROMIO 3.1.4 refresh: prepare new romio directory 2015-04-30 18:52:22 +09:00
Gilles Gouaillardet
1ee58af8f5 update ROMIO .gitignore 2015-04-30 18:49:53 +09:00
Gilles Gouaillardet
697a866b6e ddt: correctly align next datatype description
This bug can be evidenced by the test/datatype/ddt_pack
test case on sparc architecture.
2015-04-30 15:04:54 +09:00
Ryan Grant
6ab91a6781 Merge pull request #561 from tkordenbrock/topic/mtl.fix.datatype.overflow
mtl-portals4: fix datatype overflow in ompi_mtl_portals4_long_isend()
2015-04-28 15:15:59 -06:00
Ryan Grant
cc3da91700 Merge pull request #562 from tkordenbrock/topic/mtl.expand.source.bits.to.24
mtl-portals4: expand the source field of the match bits to 24 bits
2015-04-28 14:31:06 -06:00
Rolf vandeVaart
91a8ec52ca Fix possible unintialized warnings 2015-04-28 16:25:35 -04:00
Todd Kordenbrock
8a4616f724 mtl-portals4: fix datatype overflow in ompi_mtl_portals4_long_isend()
The length parameter of ompi_mtl_portals4_long_isend() was declared
as "int", which may not be big enough depending on the platform and
compiler options used.  This commit changes the type to size_t to
prevent overflow.
2015-04-28 14:40:25 -05:00
Todd Kordenbrock
3e437f6184 mtl-portals4: expand the source field of the match bits to 24 bits
The source field was 16 bits which is not sufficient for many
current and future machines.  This commit expands the source field
to 24 bits and reduces the tag field from 32 bits to 24 bits.
2015-04-28 14:25:30 -05:00
Gilles Gouaillardet
18b75bd40d io/base: check the MCA version matches 2015-04-28 17:48:23 +09:00
Ralph Castain
3d46850c4d Per patch from Marco Atzeri, have the fortran wrapper links go directly to opal_wrapper to avoid breaks in the chain in some environments. 2015-04-25 17:09:06 -07:00
Yohann Burette
1be185ed87 mtl/ofi: Remove use of MR. 2015-04-24 15:55:21 -07:00
Nathan Hjelm
2716b8b1da osc/pt2pt: correct flush expected counts
Signed-off-by: Nathan Hjelm <hjelmn@me.com>
2015-04-24 13:34:21 -06:00
Nathan Hjelm
f1d09e55ec osc/pt2pt: silence warnings
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-04-23 15:35:47 -06:00
Nathan Hjelm
29b435a5a4 osc/pt2pt: fix bugs that caused incorrect fragment counting
This commit fixes a bug identified by MTT that occurred when mixing
passive and active target synchronization. The bugs fixed in this
commit are:

 - Do not update incoming fragment counts for any type of unbuffered
   control message. These messages are out-of-band and should not be
   considered towards the signal counts.

 - Complete a change from using received counts to expected counts for
   lock, unlock, and flush acks. Part of the change made it into
   master before the rest was ready. This was preventing wakeups in
   some cases.

 - Turn the passive_target_access_epoch module member into a
   counter. As long as at least one peer is locked we are in a
   passive-target epoch and not an active target one. This fix will
   ensure that fragment flags are set appropriately.

fixes #538

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-04-23 13:22:24 -06:00
Jeff Squyres
63c7520273 use-mpi-f08/Makefile.am: also link in libmpi_mpifh.la
Per mail from Macro Atzeri, we also need to link in libmpi_mpifh.la,
lest we exhaust relative offset addressing (e.g., in 32 bit builds).

See http://www.open-mpi.org/community/lists/devel/2015/04/17304.php.
2015-04-22 14:22:36 -07:00
Ryan Grant
1436417488 Merge pull request #547 from tkordenbrock/topic/mtl.add.logical.mode
mtl-portals4: add the option to use the Portals4 logical to physical mapping
2015-04-22 15:06:13 -06:00
Nathan Hjelm
7c95ecf859 mca/base: provide functions to determine if a framework is registered/open
This commit also fixes a problem with the lazy opening of topo
components. The topo framework incorrectly: 1) checked if the topo
framework was open by checking the length of the components list, and
2) called the framework open directly instead of using
mca_base_framework_open.

fixes #544

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-04-21 13:54:25 -06:00
Todd Kordenbrock
8e56002ec7 mtl-portals4: add missing return to portals4_init_interface() 2015-04-21 11:30:33 -05:00
Todd Kordenbrock
34c50fa963 mtl-portals4: move MD cleanup closer to failure
PtlMDRelease() was called if read_msg() returned a failure code.
This commit moves the PtlMDRelease() inside read_msg() so that it
doesn't get called in cases where the failure happens before or at
the PtlMDBind().
2015-04-21 11:30:33 -05:00
Todd Kordenbrock
422be76770 mtl-portals4: add a debug message for thread multiple mode 2015-04-21 11:30:33 -05:00
Todd Kordenbrock
35e5ffd001 mtl-portals4: add the option to use the Portals4 logical to physical table
This commit adds an MCA variable to select Portals4 logical
addressing, populates the logical-to-physical mapping table and
initializes the NI in this mode.
2015-04-21 11:30:33 -05:00
Yohann Burette
19607d2ce7 mtl/ofi: Remove memset() from progress path. 2015-04-20 14:12:39 -07:00
Yohann Burette
d2eda04801 mtl/ofi: Use fi_tinject for small messages. 2015-04-20 14:12:39 -07:00
Nathan Hjelm
033894b493 Merge pull request #541 from hjelmn/c99_components
C99 component initialization
2015-04-20 10:45:39 -06:00
Yohann Burette
ba1bc00df1 mtl/ofi: Remove FI_CANCEL. 2015-04-20 09:40:37 -07:00
Devendar Bureddy
19f5a3eff4 HCOLL: skip hcoll if enable_mpi_threads is true
reasons:
    1) default OCOMS is not configured with --enable-ocoms-multi-threads
    2) locking overheads
2015-04-20 19:39:49 +03:00
Devendar Bureddy
dd8e9fa176 HCOLL: enable by defaut 2015-04-20 19:39:30 +03:00
Nathan Hjelm
d251fa1525 pml/ob1: fix heterogenous build
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-04-20 09:27:00 -06:00
Howard Pritchard
3339274136 Merge pull request #542 from hppritcha/topic/coverity_714118
fcoll/two_phase: coverity fix
2015-04-20 05:42:12 -06:00
Howard Pritchard
de215addc6 fcoll/two_phase: coverity fix
fix CID 714118

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2015-04-18 14:34:48 -06:00
Nathan Hjelm
df75d0382f ompi: use C99 subobject naming for component initialization
This commit helps future-proof ompi components by initializing each
component member by name.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-04-18 10:29:58 -06:00
Yohann Burette
9392bb5ede mtl/ofi: Implement Probe/Mprobe/Mrecv using FI_PEEK/FI_CLAIM. 2015-04-17 16:42:13 -07:00