1
1
Граф коммитов

13250 Коммитов

Автор SHA1 Сообщение Дата
Rainer Keller
c32516c9a3 - Include errno.h, to get MTT for sun to run through
This commit was SVN r21143.
2009-05-04 09:13:16 +00:00
Edgar Gabriel
338b136c28 adding a feature which tries to reuse a block of cids assigned to a
communicator. This works, if all processes agree that all communicators
utilizing the cids in the block have been freed. If they don't, they assign a
new block of cid's.

This fixes the application scenario reported in the week, in fact the test
succefully creates 100,000 communicators without exceeding a cid of 20. The
fix also keeps the main property of the algorithm (namely using a single
Allreduce operation to get a new block) and did not modify the communicator
structure.

This commit was SVN r21142.
2009-05-02 18:03:57 +00:00
Brian Barrett
77cf736f48 Make max_contextid field match same type as cid in communicator.
Refs trac:1904

This commit was SVN r21141.

The following Trac tickets were found above:
  Ticket 1904 --> https://svn.open-mpi.org/trac/ompi/ticket/1904
2009-05-01 21:11:59 +00:00
Ralph Castain
7b420e32b6 Add some missing tests
This commit was SVN r21140.
2009-05-01 18:35:22 +00:00
Ralph Castain
0b9116b1e3 Don't really need all those if statements...duh. Cleanup the code a bit.
This commit was SVN r21139.
2009-05-01 17:11:44 +00:00
Ralph Castain
d98fc311e9 Restore the ability to specify a range of dynamic ports for use by the TCP OOB module. The range can now be specified as any combination of ranges (e.g., 1-5,8,10,21-30). The system will error out if you attempt to specify both static and dynamic ports.
This commit was SVN r21138.
2009-05-01 15:57:36 +00:00
Ralph Castain
f6da7d86a2 Propagate Brian's change so we abort if we run out of CIDs to the csum module
This commit was SVN r21137.
2009-05-01 15:09:44 +00:00
Ralph Castain
e1673778be Replace missing headers
This commit was SVN r21136.
2009-05-01 15:09:10 +00:00
Ralph Castain
dca0246e57 Add some logic so we don't try to kill procs that are already dead, check for file existence before trying to remove, etc.
Just eliminates some harmless (but disturbing to users) warning messages

This commit was SVN r21135.
2009-05-01 14:00:59 +00:00
Brian Barrett
7f898d4e2b * Make rdma the default. Somehow, the code didn't match what was supposed
to happen
* Properly error out (rather than cause buffer overflow) in case where
  the datatype packed description is larger than our control fragments.
  This still isn't standards conforming, but at least we know what
  happened.
* Expose win_set_name to external libraries (like the osc modules)
* Set default window name to the CID of the communcator it's using
  for communication

Refs trac:1905

This commit was SVN r21134.

The following Trac tickets were found above:
  Ticket 1905 --> https://svn.open-mpi.org/trac/ompi/ticket/1905
2009-04-30 22:36:09 +00:00
Brian Barrett
736debcffc Check during communicator creation that we didn't get assigned a CID we can't handle, so that the code aborts instead of hange.
Refs trac:1904

This commit was SVN r21133.

The following Trac tickets were found above:
  Ticket 1904 --> https://svn.open-mpi.org/trac/ompi/ticket/1904
2009-04-30 19:23:57 +00:00
Ralph Castain
dfb2146430 Perform the ziatest as a C program instead of a script - less trouble that way.
This commit was SVN r21132.
2009-04-30 18:43:26 +00:00
Josh Hursey
0deb009225 A bunch of improvements focused on Snapshot Coordination (SnapC) and File Management (FileM).
* Improved timing in SnapC Full Global Coordinator
 * Improved scalability of the SnapC Full protocol
 * Minor improvements to the error reporting mechanisms in SnapC and FileM
 * Improved the memory usage of the metadata routines - now the owner of the data is more explicit.
 * Added a FileM hint to indicate when files stored locally can be moved to/from a globally mounted file system using just the 'cp' command instead of the 'rcp/scp' command. Slightly improves performance, but not too drastically. Can be set using the following SnapC MCA parameter: {{{snapc_base_global_shared=1}}}
 * Implement the ability to throttle the number of outgoing connections in FileM. At larger scales this type of explicit throttling helps prevent overwhelming the HNP machine. Default: 10, set via MCA parameter: {{{filem_rsh_max_outgoing}}}
 * Add a few diagnostic/debugging features to SnapC and FileM.

This commit was SVN r21131.
2009-04-30 16:55:39 +00:00
Josh Hursey
38aca518bd Properly initialize this variable
This commit was SVN r21130.
2009-04-30 16:43:05 +00:00
Josh Hursey
13a3453e35 more copyright fixes - sorry
This commit was SVN r21129.
2009-04-30 16:41:50 +00:00
Josh Hursey
ab63ab6568 forgot to update the copyright
This commit was SVN r21128.
2009-04-30 16:39:54 +00:00
Josh Hursey
76812318bb Fix a potential NULL reference
This commit was SVN r21127.
2009-04-30 16:38:43 +00:00
Josh Hursey
1f42065950 Make sure the CRCPW wrapper does not try to reference a NULL value in MPI_Finalize(), due to the ordering of pml_finalize and comm_del.
Some of the PML interfaces are noops in BKMRK. Allow the CRCPW to detect and skip the call to these functions.

This commit was SVN r21126.
2009-04-30 16:36:50 +00:00
Josh Hursey
759c2b5596 Add a 'crs_blcr_dev_null' MCA parameter. This causes BLCR to checkpoint directly to /dev/null instead of to a file.
Though this is not useful in checkpointing an application, it can be a useful diagnostic.

This commit was SVN r21125.
2009-04-30 16:32:55 +00:00
Ralph Castain
a74b74a68c Now that we understand why it failed before (thanks George!), go back to relaying cmds before processing them on the orteds so we pickup some speed by executing cmds in parallel. Modify the termination system accordingly.
Cleanup the termination in orterun when abnormally commanded via ctrl-c. We can just directly terminate_orteds as the orteds always kill any lingering local procs before exiting - no need to do the two-step cha-cha.

This commit was SVN r21123.
2009-04-30 15:08:02 +00:00
Jeff Squyres
80a1ae45ba Add missing header
This commit was SVN r21122.
2009-04-30 11:36:35 +00:00
Shiqing Fan
ff0e51f686 Include a missing header.
This commit was SVN r21121.
2009-04-30 09:03:21 +00:00
Shiqing Fan
5b76f583e4 No RSH support on Windows.
This commit was SVN r21120.
2009-04-30 09:01:20 +00:00
Shiqing Fan
fbaa30bf61 Add a few log file definitions for Windows.
This commit was SVN r21119.
2009-04-30 08:59:46 +00:00
Ralph Castain
21a66c0eff Fix a couple of minor errors in slave launch support
This commit was SVN r21118.
2009-04-30 02:54:25 +00:00
Ralph Castain
ce206df568 Fix the danged ziatest - thx to Jeff, the mighty perl guru!
This commit was SVN r21116.
2009-04-29 17:58:35 +00:00
Rainer Keller
1c7dd5a4aa - Fix ‘ret’ may be used uninitialized in this function
As orte_util_nidmap_finalize does not return an error, initialize
    ret=ORTE_SUCCESS

This commit was SVN r21115.
2009-04-29 15:29:35 +00:00
Jeff Squyres
c01079e1f3 Fix some handling in error cases (avoid uninitialized variables,
etc.).  Yay compiler warnings for noticing these.  :-)

This commit was SVN r21114.
2009-04-29 14:40:12 +00:00
Ralph Castain
1fb92e2766 Cleanup the slave launch - completed debugging. w00t!
This commit was SVN r21113.
2009-04-29 14:37:33 +00:00
Ralph Castain
7194f1636f Complete the rewrite of rankfile mapper - ensure that all non-specified ranks are properly mapped when dealing with multiple app_contexts
This commit was SVN r21111.
2009-04-29 14:06:53 +00:00
Ralph Castain
e003e0447a Ensure we terminate cleanly when we do not launch
This commit was SVN r21110.
2009-04-29 14:06:12 +00:00
Ralph Castain
36671eb962 Replace missing header
This commit was SVN r21109.
2009-04-29 13:50:11 +00:00
Ralph Castain
adc73b0fe0 Continue replacing missing headers
This commit was SVN r21108.
2009-04-29 13:34:41 +00:00
Ralph Castain
e7b5cfa259 Replace missing include file
This commit was SVN r21107.
2009-04-29 13:19:46 +00:00
Jeff Squyres
ab7897861d Update svn:ignore
This commit was SVN r21106.
2009-04-29 10:32:56 +00:00
Shiqing Fan
e7b6445b32 Add a missed .windows file into the tarball.
This commit was SVN r21105.
2009-04-29 10:31:10 +00:00
Rainer Keller
71052deebb - Get rid of incompatible implicit declaration
Need #include string.h

This commit was SVN r21104.
2009-04-29 08:11:37 +00:00
Rainer Keller
1ef32928fd - get MTL up and running:
Nothing notable, except mtl_base_datatype.h -- Undo change from r21096:
   Yes, we should not include datatype_internal.h, but we did and we have to:
   we derefence desc, and get an incomplete type, otherwise.

This commit was SVN r21103.

The following SVN revision numbers were found above:
  r21096 --> open-mpi/ompi@221fb9dbca
2009-04-29 08:04:16 +00:00
Rainer Keller
1167935ea5 - svn propedit svn:ignore...
This commit was SVN r21102.
2009-04-29 06:59:17 +00:00
Rainer Keller
be008f2215 - In case of OPAL_ENABLE_FT == 1, we need to
#include "opal/runtime/opal_cr.h"

   Thanks to Caciano Machado for reporting.

This commit was SVN r21101.
2009-04-29 06:57:49 +00:00
Rainer Keller
4ffec30e94 - Guys, we don't have a configure-test for errno.h (should we?)
Anyway, keeps us from compiling on Jaguar.

This commit was SVN r21100.
2009-04-29 03:18:51 +00:00
Ralph Castain
5fa3b38d3c Revert r21097 as this results in multiple instantiations of global variables. Instead, fix the problem by including orte_globals.h in the orte_init.c.
Since I already had some changes in there, add in the rmaps rank_file changes - should work okay, but not fully tested.

This commit was SVN r21099.

The following SVN revision numbers were found above:
  r21097 --> open-mpi/ompi@88ae934c26
2009-04-29 02:13:14 +00:00
Rainer Keller
48e4574907 - Finally the script that does the checking and deletion of header
files...

This commit was SVN r21098.
2009-04-29 01:46:56 +00:00
Rainer Keller
88ae934c26 - So, this ain't right, but somewhere the orte-tools
do not detect the orte_name_wildcard and orte_name_invalid.

This commit was SVN r21097.
2009-04-29 01:44:43 +00:00
Rainer Keller
221fb9dbca ... Delayed due to notifier commits earlier this day ...
- Delete unnecessary header files using
   contrib/check_unnecessary_headers.sh after applying
   patches, that include headers, being "lost" due to
   inclusion in one of the now deleted headers...

   In total 817 files are touched.
   In ompi/mpi/c/ header files are moved up into the actual c-file,
   where necessary (these are the only additional #include),
   otherwise it is only deletions of #include (apart from the above
   additions required due to notifier...)

 - To get different MCAs (OpenIB, TM, ALPS), an earlier version was
   successfully compiled (yesterday) on:
   Linux locally using intel-11, gcc-4.3.2 and gcc-SVN + warnings enabled
   Smoky cluster (x86-64 running Linux) using PGI-8.0.2 + warnings enabled
   Lens cluster (x86-64 running Linux) using Pathscale-3.2 + warnings enabled

This commit was SVN r21096.
2009-04-29 01:32:14 +00:00
Rainer Keller
6c1cce8761 - For the upcoming header cleanup commit,
several header files (previously included by header-files)
   now have to be moved "upward".
   This is mainly system headers such as string.h, stdio.h and for
   networking, but also some orte headers.

This commit was SVN r21095.
2009-04-29 00:49:23 +00:00
Ralph Castain
f3cfe32b5d Update the slave launch and cleanup procedures. Track what files have been moved to the slave node to avoid attempting to copy them multiple times on top of each other. Cleanup any pre-positioned files, kill any lingering apps, and cleanup the session directory area upon termination of the daemon.
This commit was SVN r21094.
2009-04-29 00:11:19 +00:00
Ralph Castain
a405f04f1b Add missing header file
This commit was SVN r21088.
2009-04-28 14:30:52 +00:00
Shiqing Fan
d5f887c79b Add the copyright file for flex into the tarball too.
This commit was SVN r21084.
2009-04-28 10:40:34 +00:00
Shiqing Fan
7036e0e685 Cache the found path of CCP libraries.
This commit was SVN r21083.
2009-04-28 10:38:35 +00:00