boost/mpi
2
0
Fork 0
mirror of https://github.com/boostorg/mpi.git synced 2026-02-25 16:32:22 +00:00
Code Issues Projects Releases Wiki Activity
198 Commits 74 Branches 96 Tags
40c3f3ec2efc8d20abcab2c0f3b87647d7f7b281
Commit Graph

198 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Alain Miniussi
40c3f3ec2e Propagated some changes made in serilization related with archive version and pfto removal. 2015-04-14 17:10:39 +02:00
K. Noel Belcourt
b4e86408c2 Enable new test now that missing source has been added. 2014-11-21 10:53:42 -07:00
Alain Miniussi
1144a111eb test file for stl container broadcast. 2014-11-20 21:45:47 +01:00
Noel Belcourt
cf5785e040 Merge pull request #11 from ventrebd/fix-#10444-mpi-archive-failure 
Fix #10444: MPI archive failure

No, I wasn't sure enough of the impact of this change to want it in 1.57.  I just merged it now, we'll have 3 months to make sure it doesn't cause any problems.  I apologize for the delay.
 2014-10-31 14:37:41 -06:00
Noel Belcourt
b636b6dbe7 Fix broken develop testers. 
Comment out missing test files committed by
Alain Miniussi so develop testers that run
MPI testers will cycle again.
 2014-10-31 13:17:08 -06:00
Alain Miniussi
2cf0ab9258 Merge pull request #22 from aminiussi/sendrecv 
Add a basic sendrecv that should be a direct transcription of the C interface
 2014-10-29 00:50:42 +01:00
Alain Miniussi
3884466f5c Forgot to test the non mpi primitive type. 2014-10-29 00:47:37 +01:00
Alain Miniussi
fad1e0737b Add a basic sendrecv that should be a direct transcription of the C interface. 2014-10-28 14:47:37 +01:00
Alain Miniussi
34c546b9cd Merge pull request #20 from aminiussi/nulltopo 
MPI_COMM_NULL is not a valid argument for MPI_Topo_test.
But it is used to indicated leftover processes when creating graph communicator or cartesians grids (for exampl, when you create a 2X2 grid from a 5 process communicator).

So we need to deal with it as a special case.
 2014-10-07 10:13:37 +02:00
Alain Miniussi
943023a6e2 Merge branch 'develop' of github.com:boostorg/mpi into nulltopo 2014-10-07 09:27:24 +02:00
Alain Miniussi
932457abd9 Merge pull request #21 from aminiussi/obsolete 
upgrade MPI 1.1 stuff to MPI 2.x stuff
 2014-10-07 09:23:47 +02:00
Alain Miniussi
f2e5edf320 The version 3 of the MPI standard is out now. 
Upgraded really old (1.1) API to 2.x.
 2014-10-07 00:39:54 +02:00
Alain Miniussi
33aa0744d1 When calling MPI_Topo_test, explicitly deal with MPI_COMM_NULL as it is not a legal value. Although it is what is return for leftovers process when creating cartesian grid or graphs. 2014-10-06 18:54:29 +02:00
Alain Miniussi
abd82f0eda Merge pull request #14 from aminiussi/compound_recv_wait 
Wait_any is missing some compound recv with intel MPI.
In practice, this change should only affect implementations which, when stacking 2 send or 2 recv will only execute the second one when the first one is completed. As a result, when testing a compound recv on compound send, we get a catch22. This merge fixes that problem.
 2014-09-16 14:08:44 +02:00
Alain Miniussi
1d498e48d5 Merge branch 'develop' into compound_recv_wait 2014-09-16 13:47:01 +02:00
Alain Miniussi
544b0017a5 Merge pull request #12 from aminiussi/test_any_bug 
Only the first request was tested in test_any. We now test all of them.
 2014-09-16 13:41:56 +02:00
Alain Miniussi
4f784b2347 Merge branch 'develop' into test_any_bug 2014-09-16 13:34:25 +02:00
Alain Miniussi
f104e41b84 Merge pull request #13 from aminiussi/nonblocking_test_split 
split nonblocking_test in two functions, one to run all tests and one to run simgle tests.29292
 2014-09-16 12:02:10 +02:00
Alain Miniussi
d7736702ac Merge branch 'develop' into nonblocking_test_split 2014-09-16 11:45:21 +02:00
Alain Miniussi
b4576d5811 Merge pull request #17 from aminiussi/develop 
Python section was missing [endsec]
 2014-09-16 11:35:59 +02:00
Alain Miniussi
6bc906ed0f Python section was missing [endsec] 2014-09-16 11:32:17 +02:00
Alain Miniussi
3392433732 Merge pull request #15 from aminiussi/map-c-update 
Propagate change made in mpi-forum documentation.
A mpi-1.1 directory was inserted on the web-site.
 2014-09-16 11:04:58 +02:00
Alain Miniussi
fcc098aa59 Small typo, we had double slashes in external URL. 
Did not hurt, but it nicer this way.
 2014-09-15 15:47:52 +02:00
Alain Miniussi
befe62d920 The documentation at mpi-forum moved from http://www.mpi-forum.org/docs/mpi-11-html to http://www.mpi-forum.org/docs/mpi-1.1/mpi-11-html. Changed the documentation to reflect that. 
In short term, we might want to move to MPI 2.x documentation.
 2014-09-15 14:48:43 +02:00
Alain Miniussi
ed4852863a In wait_any, we need to test for compound recv of serialized object that just got the size but not the body (that is, request[1] == null & handler != null). 
Note that in:
if (current->m_requests[0] != MPI_REQUEST_NULL &&
        (current->m_requests[1] != MPI_REQUEST_NULL ||
         current->m_handler)) {
      if (optional<status> result = current->test())
        return std::make_pair(*result, current);
    }

I suspect the 3 line test should be removed and every thng handled in request::test (which is supposed to tell wheter we have a complete request).
But I have no idea why it was done like that in the first place.
 2014-09-10 13:37:11 +02:00
Alain Miniussi
6fc0bc3d94 There are 2 function in nonblocking_test. Which makes debuging less convenient with no benefit. 2014-09-10 11:32:59 +02:00
Alain Miniussi
6c597245a2 test_any was actually nly testing the firt request due to a typo in the for loop. 2014-09-10 11:03:11 +02:00
Ventre, Brian D.
463e897f75 Fix #10444: MPI archive failure 
Add "const" to save_override for archive::class_id_type and archive::version_type.  WIthout it, they do not participate in overload resolution when called with a const argument, and the default template function is called.  This leads to saving 4 bytes for archive::version_type, but only loading 1 -- all subsequent load operations are then off by 3 bytes.
 2014-09-04 09:07:03 -04:00
K. Noel Belcourt
fe6e8b77ac Only resize vector on root rank we're reducing to. 
As Steve Nowlen (drnuke@lanl.gov) pointed out, we can
avoid the vector resize on all ranks except the root we're
reducing to.
 2014-09-01 21:07:53 -06:00
Noel Belcourt
d2d9017d77 Merge pull request #10 from danieljames/metadata 
Create metadata file.
 2014-08-23 14:10:10 -06:00
Daniel James
7014b82451 Add metadata file. 2014-08-18 15:07:26 +01:00
Noel Belcourt
e6dd3db0f6 Cleanup unused variables and typedefs. 2014-07-31 12:45:35 -06:00
Noel Belcourt
c6f67b85a1 Merge pull request #9 from gnzlbg/develop 
adds missing explicit conversion to bool of optional

Looks good, thanks.
 2014-07-31 12:16:44 -06:00
gnzlbg
c051d05ba2 fixes explicit operator bool of optional in c++>11 2014-07-31 20:11:06 +02:00
Noel Belcourt
76c79c90f8 Merge pull request #6 from juliohm/develop 
Add scatterv and gatherv collective operations

Tested with clang-5.1 on Mavericks and gcc-4.8.2 on Linux.  I'll make some wordsmithing, column 80, and a couple of other small changes after the commit.  Thanks Julio, nice work.
 2014-07-12 15:21:23 -06:00
Júlio Hoffimann
5351d17158 Add tests for scatterv and gatherv 2014-06-17 13:25:34 -03:00
Júlio Hoffimann
e70a7f937f Fix indentation in scatterv.hpp and gatherv.hpp 2014-06-17 13:23:53 -03:00
Júlio Hoffimann
9ae243ba59 Add hyperlinks to scatterv and gatherv in the docs 2014-06-16 16:15:53 -03:00
Júlio Hoffimann
f1ec2fcb22 Make scatterv and gatherv into collectives.hpp 2014-06-16 15:24:34 -03:00
K. Noel Belcourt
fd54b706c6 Merge branch 'master' into develop 
Conflicts:
	include/boost/mpi/packed_iarchive.hpp
	include/boost/mpi/packed_oarchive.hpp
 2014-05-09 19:56:18 -06:00
Júlio Hoffimann
1f651b8cb1 Add scatterv and gatherv collective operations 2014-05-03 14:26:06 -03:00
Noel Belcourt
1ef558e384 Merge pull request #3 from danieljames/remove-old-files 
Remove files incorrectly restored by git conversion.

Looks reasonable, thanks again.
 2014-04-27 14:38:14 -06:00
Noel Belcourt
f1e9c300c7 Merge pull request #4 from danieljames/master-cherry-pick 
Master cherry pick

Looks good Daniel, thanks for putting this together.  I don't see any problem with applying these changes.
 2014-04-27 14:35:29 -06:00
Douglas Gregor
f538c6d6bf Make MPI non-blocking semantics more clear 
[SVN r56203]

Conflicts:
	doc/mpi.qbk
 2014-04-01 22:05:48 +01:00
Matthias Troyer
92e837bd4c proposed fix for ticket 2399 
[SVN r55449]
 2014-04-01 22:04:48 +01:00
Matthias Troyer
dc3b1e5831 Fixed ticket 2723 
[SVN r55445]
 2014-04-01 22:04:34 +01:00
Daniel James
a6e603d50c Remove files incorrectly restored by git conversion. 
These directories didn't exist in trunk. In some repos, old deleted
files seem to have been kept in the git conversion. I think this is
something to do with the old CVS to subversion conversion making the
history a bit complicated.
 2014-04-01 21:59:40 +01:00
K. Noel Belcourt
a7a2501242 Fix problems caused by serialization changes. 2014-02-21 20:46:26 -07:00
Noel Belcourt
a55c9318bc Merge pull request #2 from jhunold/visibility 
Add visibility support for Boost.MPI. Fixes #2114
 2014-01-29 20:45:52 -08:00
Noel Belcourt
8a2913c42e Merge pull request #1 from danieljames/develop 
Update path for new location of Boost Build.
 2014-01-27 21:58:11 -08:00