ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
3 459 коммитов 3 Ветки 0 Теги 130 MiB
f0d86178bc
Граф коммитов

3459 Коммитов

This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
George Bosilca
f0d86178bc Remove useless file (all functions moved in others .c and .h files). 
This commit was SVN r3869.
 2004-12-22 06:51:40 +00:00
George Bosilca
4acb2b853c Start the cleanup process. I try to remove all useless operations from the critical path. 
Start of the pool of memory that will be used for unexpected messages (just the framework at this moment).

This commit was SVN r3867.
 2004-12-22 00:14:31 +00:00
George Bosilca
92e94f5921 Decrease the creation time for a convertor. I target here specifically the contiguous data or/and the data with a short description. Each convertor 
have now a limited stack attached. If we handle contiguous data then we will use this stack, avoiding the free/malloc for the stack management. In 
all others cases the convertor work as before a stack containing the required number of elements will be allocated. This small modification
decrease the latency for GM by nearly 0.7 micro-sec as reported by NetPipe.

This commit was SVN r3866.
 2004-12-22 00:12:26 +00:00
Tim Woodall
fb203852d3 I/O forwarding framework 
This commit was SVN r3865.
 2004-12-21 22:16:09 +00:00
George Bosilca
d4434bac91 The eager protocol is defined directly in number of bytes not in number of "minimum fragments". 
In the case of a zero length message avoid any convertor creation on both sides.

This commit was SVN r3857.
 2004-12-21 06:39:06 +00:00
George Bosilca
d70a22b057 Start to add a MCA parameter to fix the limit between eager and rendez-vous protocol (gm_max_eager_fragments). 
This commit was SVN r3856.
 2004-12-21 06:23:42 +00:00
Jeff Squyres
7f5c731cdf Fix what looks like bit-rot. 
This commit was SVN r3850.
 2004-12-20 17:12:34 +00:00
Ralph Castain
b860872343 Fix the module declaration - thanks to Jeff for catching the warning. 
This commit was SVN r3849.
 2004-12-20 14:38:55 +00:00
Ralph Castain
832f4b566c Add a new function to the name server that takes a base and offset and returns the corresponding vpid. 
Fix the soh code.

This commit was SVN r3848.
 2004-12-20 01:48:37 +00:00
Ralph Castain
45f77e2fc1 Take a quick break and build type checking into the buffer packing system. Add a new constant OMPI_PACK_MISMATCH - return it whenever the caller asks for a different type than that of the next item in the buffer. 
This commit was SVN r3847.
 2004-12-18 20:18:41 +00:00
Jeff Squyres
18a2960f59 Remove dereferences to (void*) 
This commit was SVN r3846.
 2004-12-18 01:38:22 +00:00
Brian Barrett
43d22fce30 * Fix for ticket #1040. Make sure that we have enough space in the 
buffer for the SIOCGIFCONF ioctl to complete successfully.  Also, use
  the sa_len member of if ifreq's ifr_addr member, if available, for
  computing offsets of the ifreq structures.

  Since this has the potential to break people, setting the env
  variable OMPI_orig_if will result in the old code being used.  This
  will be removed once the new code survives a couple days in the
  wild.

This commit was SVN r3845.
 2004-12-17 16:27:37 +00:00
George Bosilca
5d50b025b8 Decrease the latency. Just keep everything minimal. 
This commit was SVN r3844.
 2004-12-17 09:27:10 +00:00
George Bosilca
17f37130d2 First fully working version of GM. Next step ... improuvements ... 
This commit was SVN r3843.
 2004-12-17 08:39:07 +00:00
Craig E Rasmussen
c134c911e3 Changed declared type of function pointers to void* from MPI_Fint* 
This commit was SVN r3842.
 2004-12-17 05:18:01 +00:00
Jeff Squyres
ae5180aa3b Ok, it's *not* enough to only have one of the files compilable. Need 
to make them all compilable.

#if 0'ed out one section in soh_base_open.c that needs to be fixed.

This commit was SVN r3841.
 2004-12-17 03:20:39 +00:00
Jeff Squyres
63474e0232 Have to have *one* compilable file or some platforms complain. 
Grrr...

This commit was SVN r3840.
 2004-12-17 00:26:16 +00:00
Jeff Squyres
c58a99b6ac Add a missing Makefile.am for soh, and add some svn:ignore's. 
Shame on me for not doing a sanity build :-\ -- thanks Brian for
pointing it out...

This commit was SVN r3839.
 2004-12-16 23:28:02 +00:00
Brian Barrett
bb7de89545 * Add configure tests for SA_RESTART signal constant and sa_len 
* protect use of SA_RESTART in event library

This commit was SVN r3838.
 2004-12-16 21:37:57 +00:00
Jeff Squyres
32d85dd7eb Properly add the soh framework in all the places it needs to be. Left 
the src/mca/soh/base/*.c files out of the default build because
they're uncompilable right now -- don't want to step on any of Ralph's
local changes.

This commit was SVN r3837.
 2004-12-16 18:52:33 +00:00
Jeff Squyres
1fc56e8b9c Fix minor typo 
This commit was SVN r3836.
 2004-12-16 18:43:54 +00:00
Jeff Squyres
a5b61fa75c Remove svn:executable from source files 
This commit was SVN r3835.
 2004-12-16 17:49:50 +00:00
Brian Barrett
6d969db21b * Code for bug 1036 - add special CFLAGS to enable the restrict keyword 
for compilers that don't support it out of the box but can with
  the right flag

This commit was SVN r3834.
 2004-12-16 16:34:16 +00:00
Brian Barrett
d3d52a4aed * Changes to make MPI_Comm_spawn_multiple work properly 
This commit was SVN r3833.
 2004-12-16 15:42:02 +00:00
Craig E Rasmussen
cfdeb89420 Fixed declaration of function pointers by declaring as external and removing intent. 
This commit was SVN r3832.
 2004-12-16 04:58:22 +00:00
Craig E Rasmussen
88eb530a32 Fixed tests of function pointer parameters 
This commit was SVN r3831.
 2004-12-16 04:56:14 +00:00
Craig E Rasmussen
81afaad919 Set assignment of void types (pointers) to VOID_VAL 
This commit was SVN r3830.
 2004-12-16 04:54:15 +00:00
George Bosilca
58099fdca7 Correctly compute the stack count for contiguous datatypes. At the position 1 
we should always have the amount in bytes that still have to be copied on the last
basic datatype (count included).

This commit was SVN r3829.
 2004-12-16 02:15:08 +00:00
George Bosilca
a61a24f545 added support for em64t (cf. Torsten Hoefler). 
This commit was SVN r3828.
 2004-12-16 02:13:29 +00:00
Craig E Rasmussen
354d78b0b8 Fixed comment errors on some MPI_File_* methods. 
This commit was SVN r3827.
 2004-12-15 21:11:51 +00:00
Mitch Sukalski
543338c070 more fixes 
This commit was SVN r3826.
 2004-12-15 20:28:10 +00:00
Brian Barrett
c0985fad73 * turns out that there won't be a process status if the process died 
before it hit ompi_rte_init (duh!).  So don't print a ton of messages
  complaining about something that isn't really a problem

This commit was SVN r3825.
 2004-12-15 19:26:00 +00:00
Brian Barrett
e0483a5c26 * Registry no longer barfs when it gets pounded at startup, so remove 
sleep hack.
* Fix some places where processes would not get cleaned up if an error
  occured during startup

This commit was SVN r3824.
 2004-12-15 19:22:30 +00:00
Brian Barrett
adf9c7cb81 * fix book keeping error that caused badness when 1) processes were started 
on multiple nodes and 2) the processes unexpectedly quit

This commit was SVN r3823.
 2004-12-15 18:52:15 +00:00
Ralph Castain
087f9a361b Since the &*(&(&* subversion won't let me commit the whole thing - getting the $#@$# "won't delete locally modified directory nonsense again - I've committed this separately. I'll have to blow away my existing tree and checkout another to verify things still work. 
Grrrrrr...

This commit was SVN r3822.
 2004-12-15 18:38:34 +00:00
Ralph Castain
8d94778f1e Create a new state-of-health monitoring framework and move the bproc monitoring code to it. Put in an .ompi-ignore to prevent it from being compiled for now. Lots more definition needs to be done here - design document to follow when I can get to it. 
This commit was SVN r3821.
 2004-12-15 18:36:21 +00:00
Brian Barrett
43b565f110 * add missing header file 
This commit was SVN r3820.
 2004-12-15 03:04:04 +00:00
Craig E Rasmussen
24ab817703 1. Fixes for procedures with function pointer arguments. 
2. Fixed for void* when method is not templated (not an array).

This commit was SVN r3819.
 2004-12-14 23:25:55 +00:00
Craig E Rasmussen
f9c8c89f1b 1. Fixed interface declaration for procedures taking function pointer 
arguments.
2. Some void* arguments are Fortran integers, rather than array memory.
   This is needed for void* when methods don't have the template='yes'
   tag.

This commit was SVN r3818.
 2004-12-14 23:23:50 +00:00
Brian Barrett
1901882225 * first take at a cntl-c handler for mpirun. This does not do everything 
that we want, but will do a reasonable job at cleaning up the job
  if SIGINT (cntl-c) or SIGTERM are received between spawning of processes
  and the death of all the processes.

  If you see strange errors out of mpirun, please let me know.  I'm sure
  there are a couple race conditions.  I'm going to clean the code up
  tonight to try to reduce some of them.

This commit was SVN r3817.
 2004-12-14 21:21:14 +00:00
Brian Barrett
f6b8ac67ee * yeah, so it turns out that for an unsigned int, 0 - 1 doesn't equal -1. 
Since there are no elements in the list, this function doesn't need to
  do anything anyway.  So just short circuit

This commit was SVN r3816.
 2004-12-14 21:06:49 +00:00
Mitch Sukalski
c551d387a1 more IB header fixes 
This commit was SVN r3815.
 2004-12-14 20:05:12 +00:00
Brian Barrett
2a8f636d6e * and one more dumb mistake bites the dust 
This commit was SVN r3814.
 2004-12-14 18:27:05 +00:00
Brian Barrett
dd7c53dfa3 * fix dumb error causing segfault in mpirun 
This commit was SVN r3813.
 2004-12-14 18:04:39 +00:00
Mitch Sukalski
19dbfd376b IB header fixes 
This commit was SVN r3812.
 2004-12-14 17:56:51 +00:00
Tim Woodall
60e43d6738 set progress to be non-blocking for non-threaded build 
This commit was SVN r3811.
 2004-12-14 16:35:38 +00:00
Tim Woodall
572df54b2a enable event processing for oob w/ non-thread build 
This commit was SVN r3810.
 2004-12-14 16:34:10 +00:00
Jeff Squyres
7abc1f0f3d Remove the outdated src/os tree -- all this atomic stuff is now in 
src/include/sys (and has been for a long time).

This commit was SVN r3809.
 2004-12-14 16:15:15 +00:00
Brian Barrett
2b9f9f67a4 * if we are on broken linux threads, make sure that forks and waitpids all 
happen in the event thread so that the process starters can actually 
  reap the dead processes when required
* Added logic to ompi_mpi_abort to kill both local and remote groups

This commit was SVN r3808.
 2004-12-14 15:47:31 +00:00
Jeff Squyres
ef5edc5dbe Update the nightly unimplemented report to skip the MPI-2 one-sided 
functions

This commit was SVN r3807.
 2004-12-14 13:57:49 +00:00