1
1

5683 Коммитов

Автор SHA1 Сообщение Дата
Rainer Keller
192625d2a1 - Once again: uninteresting cleanup to get diff smaller.
This commit was SVN r7178.
2005-09-04 20:54:19 +00:00
Galen Shipman
1af92a1de2 tried to acquire wrong lock..
This commit was SVN r7177.
2005-09-04 13:26:46 +00:00
Galen Shipman
ba9342085e Changes to support rcache in mpool
This commit was SVN r7176.
2005-09-04 04:20:55 +00:00
Galen Shipman
dc769e3560 Initial checking of rcache.. svn issues required mult checkins..
This commit was SVN r7175.
2005-09-04 04:19:31 +00:00
Galen Shipman
2314919b73 Initial checkin of rcache
This commit was SVN r7174.
2005-09-04 04:17:00 +00:00
Brian Barrett
492aeecd11 * track the actual allocations within the ompi and opal free_lists, so that
we can free all allocated memory when the free_list is destructed.  Fixes
  a whole bunch of memory leaks...

This commit was SVN r7173.
2005-09-03 19:46:44 +00:00
Jeff Squyres
bc72a7722b Updates:
- bcast now works properly for root!=0 and multi-fragment messages
- destroy mpool when communicator is destroyed
Still need to implement:
- "in use" flags for groups of fragments so that "wrapping around" in
  the data segment doesn't overwrite not-yet-read data
- ensure that shared memory isn't removed before all processes have
  finished with it (e.g., during COMM_FREE)

This commit was SVN r7172.
2005-09-03 11:49:46 +00:00
Jeff Squyres
e53fa399dc Bump the required version of Libtool up to 1.5.16
This commit was SVN r7171.
2005-09-03 10:11:30 +00:00
Graham Fagg
059d7f26f8 re-enable proper freeing of allocated structures now that I know these
are not causing an error.

This commit was SVN r7170.
2005-09-03 06:19:17 +00:00
Graham Fagg
35447d4c32 underfilling of a only segment and the doing a PML send and esp a recv on it can
cause flipping hard to find memory errors...

This commit was SVN r7169.
2005-09-03 06:14:03 +00:00
Graham Fagg
36eddb6609 checkpoint
This commit was SVN r7168.
2005-09-03 01:41:13 +00:00
Ralph Castain
12daecb826 More cleanup
This commit was SVN r7167.
2005-09-03 01:22:11 +00:00
Ralph Castain
4b5b3b4164 Properly handle the argv array and clean it up when done.
This commit was SVN r7166.
2005-09-03 00:15:21 +00:00
Jeff Squyres
99c3edab67 Added mca_mpool_base_module_destroy() function so that you can destroy
an mpool module before MPI_FINALIZE (i.e., ompi_mpi_finalize())

This commit was SVN r7165.
2005-09-03 00:07:34 +00:00
Jeff Squyres
1deea5ddb9 Minor mods:
- style and error code updates
- use strdup instead of malloc/strcpy

This commit was SVN r7164.
2005-09-03 00:06:20 +00:00
Jeff Squyres
42550d57d1 David made the good point that this is a TCL file, so we should name
it ".tcl".

This commit was SVN r7163.
2005-09-02 22:39:15 +00:00
Jeff Squyres
0add040f1b Ensure to ship openmpi-totalview.rc.
This commit was SVN r7162.
2005-09-02 22:10:46 +00:00
Jeff Squyres
fb4de5aec5 Totalview startup script that makes Totalview not display the source
code for orterun.  Users can either copy this file to $HOME/.tvdrc or
they can source it from their $HOME/.tvdrc:

source /path/to/openmpi-totalview.rc

This commit was SVN r7161.
2005-09-02 22:07:24 +00:00
Josh Hursey
78da530fd2 Fix a bug that Tim highlighted in which orted coredumps when an orterun is
CTRL-C'd. 
We were calling orte_finalize recursively which caused a segv when it tried to 
use a freed framework (orte_rmgr in this case).

I added a status flag to orte_universe_info to indicate where we are in the code.
This was needed to determine if we should call orte_abort or not when shutting
down in the tcp oob.

This commit was SVN r7160.
2005-09-02 21:07:21 +00:00
Ralph Castain
0f797fd40b Few more cleanups
This commit was SVN r7159.
2005-09-02 21:01:59 +00:00
Ralph Castain
4ed7752681 Continue cleaning up memory leaks during launch
This commit was SVN r7158.
2005-09-02 20:41:24 +00:00
Ralph Castain
f352890732 Cleaning up memory leaks for proxy operations.
This commit was SVN r7157.
2005-09-02 19:26:21 +00:00
Ralph Castain
4bd25e0292 Few minor memory leak cleanups
This commit was SVN r7156.
2005-09-02 18:50:01 +00:00
Tim Woodall
b65dc08ab1 counters need to be signed as we check for <0
This commit was SVN r7155.
2005-09-02 18:26:07 +00:00
Tim Woodall
dfe52fceef minor changes to thread locking
This commit was SVN r7154.
2005-09-02 16:27:01 +00:00
Brian Barrett
a9863e5ff5 * properly construct / destruct the condition variable in the free list
structures

This commit was SVN r7153.
2005-09-02 16:00:42 +00:00
Galen Shipman
589b1b8b5a Additional changes to add_proc and tokens
This commit was SVN r7152.
2005-09-02 15:18:36 +00:00
Tim Woodall
4fbf7965e3 check for rdma btls when leave pinned is set
This commit was SVN r7151.
2005-09-02 15:04:47 +00:00
Jeff Squyres
647f6a7648 Update data structures
This commit was SVN r7150.
2005-09-02 14:53:14 +00:00
Jeff Squyres
6ef4805729 Tree-based barrier and broadcast seem to be working. Now on to reduce
/ allreduce...

This commit was SVN r7149.
2005-09-02 12:57:47 +00:00
Jeff Squyres
a8f9df9951 Change the API slightly so that it is assumed that each process will
be called with a description of its memory segments to make local.  It
is a small enough API that changing to support a
one-process-does-all-assignment model is simple enough if we ever need
it.

This commit was SVN r7148.
2005-09-02 12:57:02 +00:00
Jeff Squyres
560f323048 Update to ensure that the macros AC_REQUIRE'd by AC_CHECK_HEADER are
always invoked properly.

This commit was SVN r7147.
2005-09-02 11:01:07 +00:00
Jeff Squyres
a7fbb0f95e Put in comments about why these assignments exist
This commit was SVN r7146.
2005-09-02 10:27:23 +00:00
Jeff Squyres
7e4f696501 Fix silly compiler warnings
This commit was SVN r7145.
2005-09-02 10:26:41 +00:00
Jeff Squyres
99db862e9b Fix silly compiler warning
This commit was SVN r7144.
2005-09-02 10:18:06 +00:00
Jeff Squyres
cf19709d89 Oops -- get the TV API prefix right.
This commit was SVN r7143.
2005-09-02 10:16:59 +00:00
Galen Shipman
a7a4da4502 Scale the SRQ based on the log base 2 of the number of peers,
this assumes that the peers have all been added via add_procs up front. 
Bad things will happen if add_procs is called again later on a new set of
 procs to fix this we need to modify the srq which may wreck things.. looking
 into this deeper.. 

This commit was SVN r7142.
2005-09-02 04:06:51 +00:00
Brian Barrett
827122fc62 * order libraries in the correct order so that opal_error test works
properly

This commit was SVN r7141.
2005-09-02 03:28:38 +00:00
Brian Barrett
7e9b777ead * there's no var_in variable for the component-specific variables, so don't
have on there.  Fixes static builds with GM.

This commit was SVN r7140.
2005-09-02 03:05:38 +00:00
Galen Shipman
c8a23106c0 More fixes for sq tokens,
Additional work on multi-rail support. 

This commit was SVN r7139.
2005-09-02 03:04:28 +00:00
Graham Fagg
25787c4ca7 added segmented binary
renamed split so we know its a split binary tree
(tested, chain/bin/splitbin all pass simple and ibm tests)

This commit was SVN r7138.
2005-09-02 02:19:59 +00:00
Graham Fagg
a7a20e1a57 corrected first seg recv for intermediate nodes (or else we lose a request)
This commit was SVN r7137.
2005-09-02 00:55:47 +00:00
Ralph Castain
66a215eae1 More memory cleanup...
1. Valgrind is good for something - chasing down memory leaks in registry led me to re-visit the dictionary functions and discover that I wasn't keeping track of the number of dictionary entries on each segment! Resulted in wasted time searching blank entries as well as leaked memory. This has now been fixed.

2. Fixed the orte_bitmap test. The init function for that class has been eliminated and the constructor adjusted to provide that functionality.

This commit was SVN r7136.
2005-09-02 00:26:58 +00:00
Graham Fagg
d78e9d0e9e get/handle extent correctly
last rank in comm is not always last in tree structure for nonzero roots

This commit was SVN r7135.
2005-09-01 22:56:47 +00:00
Ralph Castain
76e622a552 Clean up a few memory leaks - more to go...
This commit was SVN r7134.
2005-09-01 17:38:04 +00:00
Graham Fagg
f4de8776a6 ops
This commit was SVN r7133.
2005-09-01 17:12:28 +00:00
Ralph Castain
cb128ab87b Be a little more friendly and tell the user we couldn't reach their specified universe... :-)
This commit was SVN r7132.
2005-09-01 15:52:37 +00:00
Ralph Castain
d0f7dafc47 Revise the universe connection logic. Two cases are now handled:
1. user does NOT specify the universe name. For the default universe case, if we detect an existing default universe and cannot connect to it, we quietly create an alternative default name by adding the pid to the orte_default_universe name and move on - we no longer provide a warning message for this case.

2. user specified a universe name. If we detect an existing universe of that name and cannot connect to it, we consider this an error condition and abort.

This commit was SVN r7131.
2005-09-01 15:50:38 +00:00
Tim Woodall
636ab23fdb atomic increment/test
This commit was SVN r7130.
2005-09-01 15:09:50 +00:00
Ralph Castain
03e45e6723 Two quick additions:
1. Added OMPI_PROC_ARCH as a defined registry key and added the code so that the architecture info gets properly transmitted across all processes using the startup message.

2. Added an OMPI_MODEX_KEY definition and removed the hard-coded "modex" key from pml_modex_exchange

This commit was SVN r7129.
2005-09-01 15:05:03 +00:00