ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
3 928 коммитов 3 Ветки 0 Теги
Граф коммитов

3928 Коммитов

This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
Jeff Squyres
17ec412c57 Add another flag to the communicator: PML_ADDED. This flag is only 
set after we call mca_pml.pml_comm_add().  This allows us, in the
destructor, to only call mca_pml.pml_comm_del() if the corresponding
pml_comm_add() was called.  In this way, if an error occurs during the
construction of a communicator (e.g., attributes failing to copy, as
in the Intel test MPI_Keyval3_c), we simply propagate the error up the
stack so that an MPI exception can be generated -- not cause a failed
assert() in teg's pml_del_comm().

This commit was SVN r5032.
 2005-03-26 01:49:14 +00:00
Jeff Squyres
485adebcd5 Ensure that we invoke fortran ops with a function prototype that uses 
MPI_Fint's, and that we convert the C int's to MPI_Fint's before
invocation.

This commit was SVN r5031.
 2005-03-25 20:43:19 +00:00
Jeff Squyres
4ccd171f8a Fix a comment 
This commit was SVN r5030.
 2005-03-25 20:42:17 +00:00
Jeff Squyres
df8c736993 Ensure that we invoke fortran errorhandlers with a function prototype 
that uses MPI_Fint's, and that we convert the C int's to MPI_Fint's
before invocation.

This commit was SVN r5029.
 2005-03-25 20:25:28 +00:00
Jeff Squyres
b5082f09c2 Oops -- meant to remove that "if" block yesterday. 
This commit was SVN r5026.
 2005-03-25 18:40:37 +00:00
Jeff Squyres
b8f88e8c1c Fix some bugs in the version number scheme, and make the proces a bit 
more obvious (and documented)
- VERSION now has lots of comments in it
- separate out whether we want the SVN r number in the version and
  what the SVN r number is
- clearly describe the process for building the full version number
  string 
- make ompi info have a separate line item specifically for the SVN r
  number, regardless of what the version number is
- update "make dist" process to substitute in the SVN r version

This commit was SVN r5022.
 2005-03-25 03:19:46 +00:00
Jeff Squyres
c57377a965 - reduce kruft in finding out which thread level to use (still need to 
check initial values against Brian's new stuff)
- remove some additional kruft

This commit was SVN r5021.
 2005-03-25 03:06:06 +00:00
Jeff Squyres
c33fb9a5b0 Some changes to the ompi_obj_t back-end functionality (front-end 
interface is unchanged):
- the first call to OBJ_CONSTRUCT and OBJ_NET on a given class is now
  thread safe 
- fix memory leaks for the vtable constructor / destructors;
  orte_finalize() makes a call to free all the memory allocated for
  that stuff.  After ompi_class_finalize(), any calls to OBJECT
  functionality will be met with a swift, painful SEGV death!

This commit was SVN r5017.
 2005-03-24 22:13:05 +00:00
Jeff Squyres
428a38a29a Update logic for debug streams 
This commit was SVN r5016.
 2005-03-24 20:23:15 +00:00
Jeff Squyres
718dd96e6c Remove unused function 
This commit was SVN r5015.
 2005-03-24 20:22:01 +00:00
Jeff Squyres
cdd2808d89 Only prototype function if we need it 
This commit was SVN r5014.
 2005-03-24 20:21:32 +00:00
Jeff Squyres
b24edd6054 Don't finalize before we're finished 
This commit was SVN r5013.
 2005-03-24 20:17:06 +00:00
Tim Woodall
805095986c - mods to support daemon command line parameters 
- check return value correctly when posting non-blocking recvs
- use any values that have been set in the global structs as the
  defaults when registering mca parameters - this prevents any
  values that have been set in the structs from the command line
  parser from being overwritten

This commit was SVN r5011.
 2005-03-24 15:45:44 +00:00
Jeff Squyres
1f6b5be155 Clarify the comments on what the scripts are used for. 
This commit was SVN r5010.
 2005-03-24 15:07:32 +00:00
Jeff Squyres
3f5541349a Add UC copyright 
This commit was SVN r5009.
 2005-03-24 12:43:37 +00:00
Ralph Castain
d5ca02ae6e Add columns for responsible person and for stress test. 
This commit was SVN r5006.
 2005-03-23 23:10:39 +00:00
Jeff Squyres
6432cbfe72 Fix two minor memory bugs 
This commit was SVN r5005.
 2005-03-23 22:42:51 +00:00
Jeff Squyres
eb18a870c4 Update svn:ignore 
This commit was SVN r5004.
 2005-03-23 22:22:10 +00:00
Jeff Squyres
caad8c0463 Update svn:ignore 
This commit was SVN r5003.
 2005-03-23 22:20:02 +00:00
Jeff Squyres
7b35b47288 Only check for subscriptions if there are any to check. 
This commit was SVN r5002.
 2005-03-23 20:32:49 +00:00
Jeff Squyres
ee6b23f2fb - removed unused variables 
- fixed incorrect allocation of slot information (can't take the size
  of the fifo until after it has been created!)

This commit was SVN r5001.
 2005-03-23 18:46:29 +00:00
Jeff Squyres
997dac244e Make sure we free the right things 
This commit was SVN r5000.
 2005-03-23 18:45:26 +00:00
Ralph Castain
e2597395e8 Enable the compound command capability to reduce message loads during startup. 
Add two files (one .xls and one .pdf) that track the status of unit test development. Comments/revisions welcomed.

This commit was SVN r4999.
 2005-03-23 17:50:12 +00:00
Ralph Castain
f88ec60379 Check to see if slots_max is zero before limiting it (per Tim) to support oversubscription. 
This commit was SVN r4998.
 2005-03-23 17:43:24 +00:00
Jeff Squyres
864012cef7 Ensure to check the right thing when freeing cb's 
This commit was SVN r4997.
 2005-03-23 17:42:27 +00:00
Jeff Squyres
b27ea0c413 - fix some signed/unsigned comparisons 
- print out that the test is *supposed* to generate a warning

This commit was SVN r4996.
 2005-03-23 13:37:46 +00:00
Jeff Squyres
08328e3d79 Add missing #include 
This commit was SVN r4995.
 2005-03-23 13:04:20 +00:00
Ralph Castain
bb33402058 Complete the name service unit tests. 
This commit was SVN r4994.
 2005-03-22 20:48:55 +00:00
Ralph Castain
875da82a6f Begin cleanup of the name service unit tests. Fix an error in create_cellid. 
This commit was SVN r4993.
 2005-03-22 19:56:45 +00:00
Ralph Castain
7935e05309 Complete checkout/debug of registry unit tests. All checks out okay now, including memory leakage. 
This commit was SVN r4992.
 2005-03-22 18:58:36 +00:00
Ralph Castain
f6bb79ceef Fix test to properly clear all memory. 
This commit was SVN r4991.
 2005-03-22 18:46:07 +00:00
Ralph Castain
a97d3228b1 Cleanup memory leak in delete_itag. Update unit test to find it. 
This commit was SVN r4990.
 2005-03-22 18:40:12 +00:00
Brian Barrett
3b9753079d * Add note explaining why you can't use OMPI_DEBUG_ZERO (which was my 
first reaction when I saw the commit).  Also, use
  ifconf.ifc_len instead of ifc_len.  Shouldn't matter, but be careful.

This commit was SVN r4989.
 2005-03-22 16:24:32 +00:00
Ralph Castain
abd1048646 Cleanup some of the test areas. 
Add a memset to if.c so that purify and valgrind stop complaining about uninitialized memory.

This commit was SVN r4988.
 2005-03-22 15:37:58 +00:00
Brian Barrett
9b2b3ec078 * Make the sigchld test work again 
This commit was SVN r4987.
 2005-03-22 15:24:45 +00:00
Brian Barrett
897983ea30 * Fix regression in ompi_argv_delete exposed with the unit tests 
* Update a bunch of the unit tests to either be disabled (someone who
  isn't me and knows that code needs to fix them) or to work properly

This commit was SVN r4986.
 2005-03-22 14:41:45 +00:00
Brian Barrett
4b6aecf82d * Fix up the test directory so everything uses Automake's "make check" to 
build / run.  Only things that actually build / run right now are the
  asm and class tests.  The mca tests probably will with a static build
  but that hasn't been verified

This commit was SVN r4979.
 2005-03-22 04:25:01 +00:00
Brian Barrett
aa70a35fea * Sync trunk to r4977 of the tim branch 
This commit was SVN r4978.

The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
  r4977
 2005-03-22 00:31:17 +00:00
Jeff Squyres
ce43580e01 Include <string.h> 
This commit was SVN r4947.
 2005-03-20 14:07:13 +00:00
Jeff Squyres
2746eaa4d4 Allow orte_ symbols 
This commit was SVN r4946.
 2005-03-20 14:05:49 +00:00
Brian Barrett
12c6a00d85 * don't try to push out op stuff 
This commit was SVN r4945.
 2005-03-19 22:09:28 +00:00
Brian Barrett
c854ae0edb * more merges from the trunk, mainly to make "make dist" work correctly. 
This commit was SVN r4944.
 2005-03-19 22:04:32 +00:00
Brian Barrett
40d6ec1271 * Use openmpi-default-hostfile instead of hosts (since there will be a 
/etc/hosts if you install like an RPM would, with sysconfdir as /etc)

This commit was SVN r4939.
 2005-03-19 00:03:31 +00:00
Brian Barrett
48c81e0d59 * Real quick merge from the tim branch to get the last of the hostfile 
changes so that I can remove the references to the hosts file in the
  trunk.

This commit was SVN r4938.
 2005-03-19 00:01:13 +00:00
Brian Barrett
30af9a7b90 * More changes from the tim branch. Still has problems with ABORTed procs, 
but now tells you when it can't find orted.  Also includes memory leak
  plugs, bproc fixes, and gm repairs.

This commit was SVN r4937.
 2005-03-18 23:58:36 +00:00
Brian Barrett
a1d11d9b54 * sync with close to the trunk, but right before the gm changes that we 
actually need.  More, interesting, stuff coming soon.

This commit was SVN r4933.
 2005-03-18 23:40:08 +00:00
Brian Barrett
77c65d69cc * Merge changes from tim branch from r 4821 to 4892. Tree can now run 
MPI and non-ORTE applications for RSH on one node with or without
  threads.  I think we're approaching convergence with the tim branch

This commit was SVN r4895.
 2005-03-18 03:43:59 +00:00
Brian Barrett
e89b6ba7f3 * if we don't have progress threads, need to push the event library along 
while in condition variables.  This was lost during the merge, as I
  wasn't thinking when I cleaned that code up.

This commit was SVN r4828.
 2005-03-14 21:57:34 +00:00
Brian Barrett
6822a519bb * results from initial merge of the tim branch into the trunk. Compiles and 
ompi_info works, but that's all that has been tested.

This commit was SVN r4827.
 2005-03-14 20:57:21 +00:00
Jeff Squyres
97cb8ac2fe Removed mistaken commit 
This commit was SVN r4720.
 2005-03-08 15:50:31 +00:00