ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
2 617 коммитов 3 Ветки 0 Теги 130 MiB
7646e14987
Граф коммитов

2617 Коммитов

This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
Jeff Squyres
7646e14987 Forgot to update the profiling prototypes 
This commit was SVN r2798.
 2004-09-22 14:49:29 +00:00
Jeff Squyres
201611ad3d Add <errno.h> for linux 
This commit was SVN r2797.
 2004-09-22 08:08:10 +00:00
Brian Barrett
3795754dc8 * random fixups from last commit 
This commit was SVN r2796.
 2004-09-22 05:34:41 +00:00
Sayantan Sur
7a772d6cff Changed some descriptor management code. 
This commit was SVN r2795.
 2004-09-21 21:10:13 +00:00
Brian Barrett
782d3af2b9 * Move to using a lazy selection for pcms so that we can have multiple PCM 
sets running at once - requires an additional step in spawning to get a
  handle (that will contain multiple pcms when we support multi-cell)
* change the selection logic of the pcms to not care about setting threads,
  but instead to select based on the selected thread level, since it
  would be a little late by the time we did the selection for pcms.
* started the long process of cleaning up the rsh pcm so that it
  actually kills processes and things.  Still doesn't do anything useful,
  but getting to the point where that might be possible

This commit was SVN r2794.
 2004-09-21 20:27:41 +00:00
Jeff Squyres
208b8410ba Bunches of fixes for attributes, particularly in terms of language 
interoperability

This commit was SVN r2793.
 2004-09-21 19:52:19 +00:00
Edgar Gabriel
e1406a1d5d adding the code to manage the communicator cid allocation in multithreaded scenarios. The code seems to work in a single threaded scenario (and thus should not generate any errors hopefully), the multithreaded case still to be tested. 
This commit was SVN r2792.
 2004-09-21 18:39:06 +00:00
Edgar Gabriel
479f6c5e6b ups, now its correct 
This commit was SVN r2791.
 2004-09-21 10:41:50 +00:00
Edgar Gabriel
1ed7d779aa fixing comm_compare for inter-communicators 
This commit was SVN r2790.
 2004-09-21 10:31:26 +00:00
Jeff Squyres
2137897b31 Oops! That change should not have been committed yet -- backing out... 
This commit was SVN r2789.
 2004-09-21 00:29:03 +00:00
Jeff Squyres
f08dab6638 Set *flag = 0 very early so that it's always 0, even if we return in 
error

This commit was SVN r2788.
 2004-09-21 00:26:19 +00:00
Jeff Squyres
92a3bff649 Ensure to set the "fortran" flag on the errorhandler that was just 
created

This commit was SVN r2787.
 2004-09-21 00:25:37 +00:00
Jeff Squyres
8724f59dbb Fix more fortran values 
This commit was SVN r2786.
 2004-09-21 00:25:08 +00:00
Jeff Squyres
17011989b4 - Unify the back end of the 3 errors_are_fatal functions 
- Make the unified back-end able to send output via fprintf(stderr,
  ...) (if we're before MPI_INIT or MPI_FINALIZED) and ompi_output(0,
  ...) if we're between MPI_INIT and MPI_FINALIZED

This commit was SVN r2785.
 2004-09-20 23:01:40 +00:00
Vishal Sahay
c9bf181096 Fix a prototype error 
This commit was SVN r2784.
 2004-09-20 22:25:27 +00:00
Brian Barrett
75e6f7dac5 * remove the can_spawn functions from the pcm. when there was one pcm at 
a time and no pcmclient, this made sense.  Now, the selection logic will
  implicitly do this for us.

This commit was SVN r2783.
 2004-09-20 20:12:04 +00:00
Brian Barrett
41e17e2758 * rename pack.{c,h} to bufpack.{c,h} because there was already a pack.c in 
src/mpi/c and you can't have two object files with the same name in
  the same library

This commit was SVN r2782.
 2004-09-20 19:55:01 +00:00
Jeff Squyres
017df2797b Update fortran constants for op's 
This commit was SVN r2781.
 2004-09-20 19:22:00 +00:00
Brian Barrett
79a2d8b0a1 * remove a whole bunch of unnecessary #include "runtime/runtime.h"s. I 
am sure that some will have to be added back in as functionality is
  added (like for comm_spawn.c), but for now might as well compile
  faster :)

This commit was SVN r2780.
 2004-09-20 18:42:01 +00:00
Brian Barrett
2dc55f12da * add more selection criteria to for the pcm selection code 
* remove the ns param switch - always use the ns at this point
* clean up some of the evil rms code that wasn't multi-pcm safe.  still
  have somme work on this front

This commit was SVN r2779.
 2004-09-20 18:25:00 +00:00
George Bosilca
efc09dfc94 increase timeout 
This commit was SVN r2778.
 2004-09-20 17:29:29 +00:00
George Bosilca
d46cbcbf61 fix timeout 
This commit was SVN r2777.
 2004-09-20 17:27:38 +00:00
George Bosilca
b7794b2dbc Following Brian remarque I remove all runtime.h include from the datatype related files. 
This commit was SVN r2776.
 2004-09-20 17:19:33 +00:00
Sayantan Sur
28c4de3d9b Finalization of arbitrary message implementation. 
Cleanups left.

This commit was SVN r2775.
 2004-09-20 15:54:32 +00:00
Jeff Squyres
a9cfc982f5 Complete the rename: mpirun2 is now mpirun. 
As a temporary workaround for those who can't help themselves from
typing "mpirun2", a sym link is created in $prefix/bin from mpirun2 to
mpirun.  We should remove this before release!

This commit was SVN r2774.
 2004-09-20 09:28:19 +00:00
Jeff Squyres
a7ca0c60dc Remove old mpirun script -- it doesn't work anymore 
This commit was SVN r2773.
 2004-09-20 09:22:14 +00:00
Jeff Squyres
f579cb08c1 Correct 2 minor typos 
This commit was SVN r2772.
 2004-09-20 07:03:01 +00:00
Vishal Sahay
d843dddf18 The last few f77 bindings 
This commit was SVN r2771.
 2004-09-20 06:28:41 +00:00
Sayantan Sur
b86c488a36 FIN !!! 
Still working on reporting send_progress for arbitrary messages.

This commit was SVN r2770.
 2004-09-20 06:07:16 +00:00
Sayantan Sur
83dc2da533 Arbitrary size message implementation. 
Implemented start->ack->put.

still working on fin.

This commit was SVN r2769.
 2004-09-20 05:37:27 +00:00
Sayantan Sur
42300f7489 Separated a macro to set send descriptor length without 
refilling the entire descriptor.

This commit was SVN r2768.
 2004-09-19 15:13:07 +00:00
George Bosilca
b4c49e057a Remove a compiler warning by defining the signal_cb function as static. 
And remove a segfault on ompi_info by insuring the the event library is initialized
before we add events inside.

This commit was SVN r2767.
 2004-09-19 14:34:41 +00:00
George Bosilca
50a8bee21a Remove some compilation warnings. Change the type of the argument from int* to unsigned int*. 
This commit was SVN r2766.
 2004-09-19 13:59:51 +00:00
George Bosilca
81bf995668 Correctly go to the next element. 
This commit was SVN r2765.
 2004-09-19 13:58:30 +00:00
George Bosilca
4534354a17 Correctly initialize the Fortran MPI2 predefined datatypes/ 
This commit was SVN r2764.
 2004-09-19 13:57:35 +00:00
George Bosilca
acaa209c02 First implementation ... and I hope the last :). 
This commit was SVN r2763.
 2004-09-19 13:56:39 +00:00
George Bosilca
7707eedb51 Some additions for MPI2. At the page 297 in the MPI2 standard several new predefined datatypes are introducted. 
This commit was SVN r2762.
 2004-09-19 10:16:00 +00:00
George Bosilca
335968c235 Additional Fortran types. In the MPI2 standard at the page 297 the binding C/C++/Fortran are defined for additional Fortran datatypes. 
This commit was SVN r2761.
 2004-09-19 09:43:37 +00:00
George Bosilca
41ee24f624 Update the functions to fit the prototypes. 
This commit was SVN r2760.
 2004-09-19 09:42:44 +00:00
Jeff Squyres
0713bb4955 More ddt updates 
This commit was SVN r2759.
 2004-09-19 08:08:54 +00:00
Jeff Squyres
f559b54df4 Update some more mpif.h values -- still need to do something about 
MPI_REQUEST_NULL, though.

This commit was SVN r2758.
 2004-09-19 07:39:44 +00:00
George Bosilca
6788ad6681 The new datatype engine. With this commit the engine is completly imported in OpenMPI. 
Better memory handling, faster it adapt it's behaviour to the low level driver and to the
datatype representation. However, only the TCP driver has been modified to fit the new engine.

For Fortran the MPI_DATATYPE_NULL is now registered at index 0 in the translation table.

This commit was SVN r2757.
 2004-09-19 07:20:24 +00:00
George Bosilca
c3b73e3b46 In the begining the parent is set to comm_null. If there are no communications a segfault is generated 
when ompi_mpi_finalize is called.

This commit was SVN r2756.
 2004-09-19 07:13:58 +00:00
George Bosilca
57b1659943 New datatype engine. 
This commit was SVN r2755.
 2004-09-19 07:12:29 +00:00
George Bosilca
33aa1c1e87 Adapt to the new datatype representation. 
This commit was SVN r2754.
 2004-09-19 07:06:46 +00:00
George Bosilca
eb5e5191fa Adapt to the new datatype engine and correct some minor bugs. 
This commit was SVN r2753.
 2004-09-19 07:05:58 +00:00
George Bosilca
3482ab96d6 Update the definition of the Fortran MPI_DATATYPE_NULL. 
This commit was SVN r2752.
 2004-09-19 03:50:30 +00:00
Jeff Squyres
c85f931ae5 - query the daemon in a persistent universe to find out how many app 
slots we have to fill in MPI_UNIVERSE_SIZE
- in a non-persistent universe, just use the size of COMM_WORLD for
  MPI_UNIVERSE_SIZE.  Note the comment in the file, however, I don't
  think that this is sufficient -- e.g., reserve 8 CPUs in a batch
  environment and then run "./master" -- UNIVERSE_SIZE should still be
  8, even though we don't have a persistent universe.  So more
  integration is necessary here.
- removed some debugging printfs and added a forgotten error condition

This commit was SVN r2751.
 2004-09-18 10:04:31 +00:00
Jeff Squyres
1f98abaced Make all the .o files in the ROMIO distribution (in the romio io 
compnent) adhere to the MCA name prefix rule through some symbolic
linking trickery.  See comments in romio-dist/mpi-io/Makefile.am or
any of the adion Makefile.am's for details.

This commit was SVN r2750.
 2004-09-18 10:00:34 +00:00
Brian Barrett
2660d42f90 * fix typo that broke things in the no signals in the event library case. 
This commit was SVN r2749.
 2004-09-17 17:27:08 +00:00