1
1
Граф коммитов

370 Коммитов

Автор SHA1 Сообщение Дата
Nathan Hjelm
195c892a9d Add support for MPI_Comm_dup_with_info, MPI_Comm_create_group, and
MPI_Comm_idup.

As part of this work I implemented a basic request scheduler in
ompi/comm/comm_request.c. This scheduler might be useful for more
than just communicator requests and could be moved to ompi/request
if there is a demand. Otherwise I will leave it where it is.

Added a non-blocking version of ompi_comm_set to support ompi_comm_idup.
The call makes a recursive call to comm_dup and a non-blocking version
was needed. To simplify the code the blocking version calls the nonblocking
version and waits on the resulting request if one exists.

cmr=v1.7.4:reviewer=jsquyres:ticket=trac:3796

This commit was SVN r29334.

The following Trac tickets were found above:
  Ticket 3796 --> https://svn.open-mpi.org/trac/ompi/ticket/3796
2013-10-02 14:26:40 +00:00
Jeff Squyres
114022146a Add a missing "const" to MPI_Unpack (addendum to r29266).
Refs trac:3785.

This commit was SVN r29278.

The following SVN revision numbers were found above:
  r29266 --> open-mpi/ompi@0b8fc13299

The following Trac tickets were found above:
  Ticket 3785 --> https://svn.open-mpi.org/trac/ompi/ticket/3785
2013-09-27 12:48:28 +00:00
Nathan Hjelm
0b8fc13299 MPI-3.0: update C bindings with const and consistent use of [] for
arrays.

The MPI 3.0 standard added const to all in buffers in the C bindings. This
commit adds the const keyword and in most cases casts const away. We will
eventually should go through and update the various interfaces (coll, pml,
io, etc) to take the const keyword. The group, comm, win, and datatype
interfaces have been updated with const.

cmr=v1.7.4:ticket=trac:3785:reviewer=jsquyres

This commit was SVN r29266.

The following Trac tickets were found above:
  Ticket 3785 --> https://svn.open-mpi.org/trac/ompi/ticket/3785
2013-09-26 21:56:20 +00:00
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
George Bosilca
85db48df0e Identification, tab vs. space.
This commit was SVN r29206.
2013-09-18 16:45:00 +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
Ralph Castain
497c7e6abb Fixes trac:2904
The intercomm "merge" function can create a linkage between procs that was not reflected anywhere in a modex, and so at least some of the procs in the resulting communicator don't know how to talk to some of the new communicator's peers.

For example, consider the case where:

1. parent job A comm_spawns a process (job B) - these processes exchange modex and can communicate

2. parent job A now comm_spawns another process (job C) - again, these can communicate, but the proc in C knows nothing of B

3. do an intercomm merge across the communicators created by the two comm_spawns. This puts B and C into the same communicator, but they know nothing about how to talk to each other as they were not involved in any exchange of contact info. Hence, collectives on that communicator now fail. 

This fix adds an API to the ompi/dpm framework that (a) exchanges the modex info across the procs in the merge to ensure all procs know how to communicate, and (b) calls add_procs to give the btl's a chance to select transports to any new procs.

cmr:v1.7.3:reviewer=jsquyres

This commit was SVN r29166.

The following Trac tickets were found above:
  Ticket 2904 --> https://svn.open-mpi.org/trac/ompi/ticket/2904
2013-09-15 15:00:40 +00:00
George Bosilca
2aed876be6 Fix the bug where MPI_is_thread_main returns true for all
threads when not in MPI_THREAD_MULTIPLE mode.
Thanks to Lisandro Dalcin for pointing it out.

This commit was SVN r29113.
2013-09-04 11:10:51 +00:00
Ralph Castain
c5f395d36a Silence unitialized var warnings
This commit was SVN r29035.
2013-08-16 21:37:35 +00:00
George Bosilca
710d3836d5 Use a recv convertor for the pack external case.
This commit was SVN r29003.
2013-08-06 09:09:42 +00:00
Nathan Hjelm
cb90a4a7fc Add simple algorithms to support MPI_IN_PLACE for MPI_Alltoall, MPI_Alltoallv, and MPI_Alltoallw.
Working on faster algorithms for tuned that will come at a later time.

cmr=v1.7.3:ticket=trac:2965

This commit was SVN r28952.

The following Trac tickets were found above:
  Ticket 2965 --> https://svn.open-mpi.org/trac/ompi/ticket/2965
2013-07-25 19:19:41 +00:00
Nathan Hjelm
c4c69b4ddf MPI-3: add support for large counts using derived datatypes
Add support for MPI_Count type and MPI_COUNT datatype and add the required
MPI-3 functions MPI_Get_elements_x, MPI_Status_set_elements_x,
MPI_Type_get_extent_x, MPI_Type_get_true_extent_x, and MPI_Type_size_x.
This commit adds only the C bindings. Fortran bindins will be added in
another commit. For now the MPI_Count type is define to have the same size
as MPI_Offset. The type is required to be at least as large as MPI_Offset
and MPI_Aint. The type was initially intended to be a ssize_t (if it was
the same size as a long long) but there were issues compiling romio with
that definition (despite the inclusion of stddef.h).

I updated the datatype engine to use size_t instead of uint32_t to support
large datatypes. This will require some review to make sure that 1) the
changes are beneficial, 2) nothing was broken by the change (I doubt
anything was), and 3) there are no performance regressions due to this
change.

Increase the maximum number of predifined datatypes to support MPI_Count

Put common get_elements code to ompi/datatype/ompi_datatype_get_elements.c

Update MPI_Get_count to reflect changes in MPI-3 (return MPI_UNDEFINED when the count is too large for an int)

This commit was SVN r28932.
2013-07-23 15:35:14 +00:00
George Bosilca
5fae72b9aa Add the MPI 2.2 MPI_Dist_graph functionality.
This patch reshape the way we deal with topologies completely. Where
our topologies were mainly storage components (they were not capable
of creating the new communicator), the new version is built around a
[possibly] common representation (in mca/topo/topo.h), but the functions
to attach and retrieve the topological information are specific to each
component. As a result the ompi_create_cart and ompi_create_graph functions
become useless and have been removed.

In addition to adding the internal infrastructure to manage the topology
information, it updates the MPI interface, and the debuggers support and
provides all Fortran interfaces.

This commit was SVN r28687.
2013-07-01 12:40:08 +00:00
Jeff Squyres
34fb0712c4 Per https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/256, we need
to set *flag=1 when source == MPI_PROC_NULL.

cmr:v1.7.2:reviewer=dgoodell
cmr:v1.6.5:reviewer=dgoodell

This commit was SVN r28621.
2013-06-12 21:38:07 +00:00
Ralph Castain
3e6e1046a3 fix a correctness issue by returning an error if waitall fails and invoking the mpi error handler
cmr:v1.7.2:reviewer=jsquyres

This commit was SVN r28533.
2013-05-16 15:04:37 +00:00
George Bosilca
2331000d63 Correctly handle the invalid status for null and inactive
requests. This patch fixes trac:3475.

CMR v1.6, v1.7

This commit was SVN r28431.

The following Trac tickets were found above:
  Ticket 3475 --> https://svn.open-mpi.org/trac/ompi/ticket/3475
2013-05-01 12:55:24 +00:00
Jeff Squyres
eeb1d83c1d Don't assign the status if MPI_STATUS_IGNORE is passed in. Thanks to
Lisandro Dalcin for finding the issue.

This commit was SVN r28430.
2013-05-01 12:32:58 +00:00
Jeff Squyres
975ff4675e rc was used before it was initialized. Looks like this minor bug was
introduced in r28241.

This commit was SVN r28316.

The following SVN revision numbers were found above:
  r28241 --> open-mpi/ompi@9d4a26f47d
2013-04-09 21:49:15 +00:00
Nathan Hjelm
9d4a26f47d Update OMPI frameworks to use the MCA framework system.
Notes:
  - This commit also eliminates the need for an available components list in use
    in several frameworks. None of the code in question was making use of the
    priority field of the priority component list item so these extra lists were
    removed.
  - Cleaned up selection code in several frameworks to sort lists using opal_list_sort.
  - Cleans up the ompi/orte-info functions. Expose the functions that construct the
    list of params so they can be used elsewhere.

patches for mtl/portals4 from brian

missed a few output variables in openib

This commit was SVN r28241.
2013-03-27 21:17:31 +00:00
Ralph Castain
a4b6fb241f Remove all remaining vestiges of the Windows integration
This commit was SVN r28137.
2013-02-28 17:31:47 +00:00
Brian Barrett
312f37706e In talking about this with Jeff and Ralph, we don't actually need
ompi_show_help, because opal_show_help is replaced with an 
aggregating version when using ORTE, so there's no reason to
directly call orte_show_help.

This commit was SVN r28051.
2013-02-12 21:10:11 +00:00
Joshua Ladd
70ad711337 Backing out the Open SHMEM project
This commit was SVN r28050.
2013-02-12 17:45:27 +00:00
Mike Dubman
ff384daab4 Added new project: oshmem.
This commit was SVN r28048.
2013-02-12 15:33:21 +00:00
Brian Barrett
f42783ae1a Move the RTE framework change into the trunk. With this change, all non-CR
runtime code goes through one of the rte, dpm, or pubsub frameworks.

This commit was SVN r27934.
2013-01-27 23:25:10 +00:00
George Bosilca
4a581d276d Fix the issues with the MPI_Op and the Fortran90 types.
This commit was SVN r27707.
2012-12-19 11:08:18 +00:00
Ralph Castain
6aac54b02e Revert r27510, r27509, and r27508.
Not sure what happened here, but the resulting trunk wouldn't even configure. After spending time fixing that problem, I found it wouldn't compile due to multiple syntax errors that had been introduced in both the OPAL and OMPI layer. This raised questions as to the completeness of the work.

Given that the author is departing, I pinged Jeff about it and we agreed to revert this for now. Hopefully, it can either be fixed by the author prior to actual departure, or someone else can pick it up (now that it is in the history) and fix it.

This commit was SVN r27511.

The following SVN revision numbers were found above:
  r27508 --> open-mpi/ompi@12c3c743de
  r27509 --> open-mpi/ompi@79e4a8ca38
  r27510 --> open-mpi/ompi@1ad5ff625a
2012-10-27 16:43:45 +00:00
Shiqing Fan
1ad5ff625a Don't know why these lines are modified, probably by subversion. Although it is harmless, change it back as what it was.
This commit was SVN r27510.
2012-10-27 03:04:45 +00:00
Shiqing Fan
79e4a8ca38 Revert two files. Wrong changes were made due to a failed update.
This commit was SVN r27509.
2012-10-27 02:57:22 +00:00
Shiqing Fan
12c3c743de Per the MemPin RFC, submit the component source files, and update the memchecker macros.
This commit was SVN r27508.
2012-10-27 02:48:20 +00:00
Eugene Loh
25ad84b925 Ensure that MPI_Status objects have proper alignment:
- fix the Fortran layer to use new macros to convert Fortran-to-C status
- change the C internals to pull out old OMPI_SET_STATUS* macros

Also, change name of "status" argument in topo_test_f.c to "topo_type".

This commit was SVN r27403.
2012-10-04 14:39:51 +00:00
Eugene Loh
11f8dcb538 Update Mprobe/Improbe man page for MPI_PROC_NULL case and
mprobe.c/improbe.c for MPI_STATUS_IGNORE/MPI_PROC_NULL case.

This commit was SVN r27386.
2012-10-01 05:18:21 +00:00
Jeff Squyres
b08eaae9a2 The MPI spec says that we are supposed to return an empty status in
the (count==0) case.

This commit was SVN r27132.
2012-08-24 16:11:29 +00:00
George Bosilca
113b45b4e6 Add support for the new MPI_Type_create_hindexed_block (MPI 3.0).
This commit was SVN r26962.
2012-08-07 12:48:30 +00:00
Ralph Castain
89f3754559 Bring in the rest of MPI-3 Ticket #125 changes, including changes to mpi.h, function param lists, and associated documentation.
This commit was SVN r26944.
2012-08-03 01:09:59 +00:00
Ralph Castain
8b595ddf84 Complete changes specified in MPI-3 Ticket #125. Still holding on proposed change to MPI_Init as it remains under discussion (and is just plain uuuugly). No action taken on changing string arrays as that part of #125 was not included in the final pdf text.
This commit was SVN r26939.
2012-08-02 16:31:02 +00:00
Brian Barrett
7f2296eb39 need count for memcheck...
This commit was SVN r26812.
2012-07-19 21:28:47 +00:00
Ralph Castain
071c5fc85c Silence warnings
This commit was SVN r26809.
2012-07-19 14:29:47 +00:00
Brian Barrett
40fd943fa0 Need to pass the right module in for ireduce_scatter_block.
This commit was SVN r26787.
2012-07-12 13:24:20 +00:00
Brian Barrett
eb3c6546c2 Remove shortcut optimizations from non-blocking collectives. We need to call
into the collectives module even in the shortcutable cases to get the requests
created.

This commit was SVN r26784.
2012-07-11 16:03:55 +00:00
Josh Hursey
28681deffa Backout the ORCA commit. :(
There is a linking issue on Mac OSX that needs to be addressed before this is able to come back into the trunk.

This commit was SVN r26676.
2012-06-27 01:28:28 +00:00
Josh Hursey
542330e3a7 Commit of ORCA: Open MPI Runtime Collaborative Abstraction
This is a runtime interposition project that sits between the OMPI and ORTE layers in Open MPI.

The project is described on the wiki:
  https://svn.open-mpi.org/trac/ompi/wiki/Runtime_Interposition

And on this email thread:
  http://www.open-mpi.org/community/lists/devel/2012/06/11109.php

This commit was SVN r26670.
2012-06-26 21:42:16 +00:00
Jeff Squyres
d7787b625e Fix memchecker usage in these functions -- these variables are not
used.  They're just copy-n-paste errors from the MPI_Reduce_scatter
variants. 

This commit was SVN r26669.
2012-06-26 21:24:45 +00:00
Jeff Squyres
dd70f84e3d Fix some linker problems with MPI_[Ir|R]educe_scatter_block:
* Ensure the profiling version is linked in
 * Ensure that PMPI variants are prototyped in mpi.h
 * Ensure weak symbols are set correctly

This commit was SVN r26649.
2012-06-25 15:26:14 +00:00
Brian Barrett
b9e8e4aeb9 * Initial merge of the non-blocking collectives interface. No implementation of
the back-end yet, coming real soon now, need to solve some tag issues first.

This commit was SVN r26641.
2012-06-22 20:54:12 +00:00
George Bosilca
3704bfccbe The sum of the local and remote recvcounts is identical (per MPI standard
requirement) so the recvcounts array is bounded by the number of
participants in the local group.

This commit was SVN r26496.
2012-05-25 04:27:19 +00:00
George Bosilca
5ff74babc6 Correctly use remote size where necessary. Completements the
patches for ticket #3105.

This commit was SVN r26490.
2012-05-24 13:41:51 +00:00
Jeff Squyres
4bda5da63a Similar to r26488, also update the param error checking for
REDUCE_SCATTER, SCATTERV, and GATHERV.  Thanks to the mpi4py community
for reporting the issue.

This commit was SVN r26489.

The following SVN revision numbers were found above:
  r26488 --> open-mpi/ompi@42793aa10f
2012-05-24 13:26:44 +00:00
Jeff Squyres
42793aa10f As reported by the mpi4py community (Bennet Fauber, Lisandro Dalcin,
Jonathan Dursi), fix a bug in the parameter verification of
MPI_ALLGATHERV: ensure to use the ''remote'' group size when checking
the recvcounts array.

This commit was SVN r26488.
2012-05-24 10:25:21 +00:00
Jeff Squyres
4b047c4257 Fix typos in the memchecker sections of MPI_IMRECV and MPI_MRECV, and
define the missing function memchecker_message().

Thanks to MTT for catching these errors.

This commit was SVN r26314.
2012-04-23 21:10:19 +00:00
Jeff Squyres
253444c6d0 == Highlights ==
1. New mpifort wrapper compiler: you can utilize mpif.h, use mpi, and use mpi_f08 through this one wrapper compiler
 1. mpif77 and mpif90 still exist, but are sym links to mpifort and may be removed in a future release
 1. The mpi module has been re-implemented and is significantly "mo' bettah"
 1. The mpi_f08 module offers many, many improvements over mpif.h and the mpi module

This stuff is coming from a VERY long-lived mercurial branch (3 years!); it'll almost certainly take a few SVN commits and a bunch of testing before I get it correctly committed to the SVN trunk.

== More details ==

Craig Rasmussen and I have been working with the MPI-3 Fortran WG and Fortran J3 committees for a long, long time to make a prototype MPI-3 Fortran bindings implementation.  We think we're at a stable enough state to bring this stuff back to the trunk, with the goal of including it in OMPI v1.7.  

Special thanks go out to everyone who has been incredibly patient and helpful to us in this journey:

 * Rolf Rabenseifner/HLRS (mastermind/genius behind the entire MPI-3 Fortran effort)
 * The Fortran J3 committee
 * Tobias Burnus/gfortran
 * Tony !Goetz/Absoft
 * Terry !Donte/Oracle
 * ...and probably others whom I'm forgetting :-(

There's still opportunities for optimization in the mpi_f08 implementation, but by and large, it is as far along as it can be until Fortran compilers start implementing the new F08 dimension(..) syntax.

Note that gfortran is currently unsupported for the mpi_f08 module and the new mpi module.  gfortran users will a) fall back to the same mpi module implementation that is in OMPI v1.5.x, and b) not get the new mpi_f08 module.  The gfortran maintainers are actively working hard to add the necessary features to support both the new mpi_f08 module and the new mpi module implementations.  This will take some time.

As mentioned above, ompi/mpi/f77 and ompi/mpi/f90 no longer exist.  All the fortran bindings implementations have been collated under ompi/mpi/fortran; each implementation has its own subdirectory:

{{{
ompi/mpi/fortran/
  base/               - glue code
  mpif-h/             - what used to be ompi/mpi/f77
  use-mpi-tkr/        - what used to be ompi/mpi/f90
  use-mpi-ignore-tkr/ - new mpi module implementation
  use-mpi-f08/        - new mpi_f08 module implementation
}}}

There's also a prototype 6-function-MPI implementation under use-mpi-f08-desc that emulates the new F08 dimension(..) syntax that isn't fully available in Fortran compilers yet.  We did that to prove it to ourselves that it could be done once the compilers fully support it.  This directory/implementation will likely eventually replace the use-mpi-f08 version.

Other things that were done:

 * ompi_info grew a few new output fields to describe what level of Fortran support is included
 * Existing Fortran examples in examples/ were renamed; new mpi_f08 examples were added
 * The old Fortran MPI libraries were renamed:
   * libmpi_f77 -> libmpi_mpifh
   * libmpi_f90 -> libmpi_usempi
 * The configury for Fortran was consolidated and significantly slimmed down.  Note that the F77 env variable is now IGNORED for configure; you should only use FC. Example:
{{{
shell$ ./configure CC=icc CXX=icpc FC=ifort ...
}}}

All of this work was done in a Mercurial branch off the SVN trunk, and hosted at Bitbucket.  This branch has got to be one of OMPI's longest-running branches.  Its first commit was Tue Apr 07 23:01:46 2009 -0400 -- it's over 3 years old!  :-)  We think we've pulled in all relevant changes from the OMPI trunk (e.g., Fortran implementations of the new MPI-3 MPROBE stuff for mpif.h, use mpi, and use mpi_f08, and the recent Fujitsu Fortran patches).

I anticipate some instability when we bring this stuff into the trunk, simply because it touches a LOT of code in the MPI layer in the OMPI code base.  We'll try our best to make it as pain-free as possible, but please bear with us when it is committed.

This commit was SVN r26283.
2012-04-18 15:57:29 +00:00