1
1
Граф коммитов

17794 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
de486d3000 Silence compiler warnings
This commit was SVN r27589.
2012-11-12 02:51:05 +00:00
Ralph Castain
26f1cd0909 Fix compiler warnings
This commit was SVN r27588.
2012-11-12 02:50:45 +00:00
Ralph Castain
bd887f7f56 Add a new "test" component to the DFS that treats all files as remote in order to test the app-to-daemon interactions on a single machine. Set a global param to indicate we are using staged execution. Add a param to indicate it is okay for non-MPI processes to execute without finalizing. Cleanup file map load and fetch operations.
This commit was SVN r27587.
2012-11-10 14:09:12 +00:00
Ralph Castain
81d0b06842 Strip the domain info from the hostname if that option is specified, protecting IP address-based names
This commit was SVN r27586.
2012-11-10 14:05:27 +00:00
Ralph Castain
615cc66b44 Protect the HNP cleanup in cases where no session dirs are created
This commit was SVN r27585.
2012-11-10 14:03:07 +00:00
Ralph Castain
ddbbc0fb7c Add the ability to pack the contents of one buffer into another as a block, thus allowing the transfer of blocks of info as a unit without messing with load/unload
This commit was SVN r27584.
2012-11-10 14:02:23 +00:00
Ralph Castain
f9f07e9535 Add a function to test if a string is in the form of an IP address - doesn't test for validity of the address
This commit was SVN r27583.
2012-11-10 14:01:12 +00:00
Ralph Castain
2358228a1b Update platform files
This commit was SVN r27582.
2012-11-10 13:59:09 +00:00
Ralph Castain
fd632147df Per patch from Nathan, with a few fixes, cleanup the orte-info tool
This commit was SVN r27581.
2012-11-10 04:11:40 +00:00
Nathan Hjelm
e0f5137e46 add prototypes for lex destroy functions
This commit was SVN r27580.
2012-11-09 22:00:27 +00:00
Mike Dubman
9392f1894e Fixes MPI_Allgather when sendbuf is MPI_IN_PLACE. fixes 3342
This commit was SVN r27579.
2012-11-09 18:05:10 +00:00
Nathan Hjelm
a754674fd7 Per the specification for putenv (http://pubs.opengroup.org/onlinepubs/009604599/functions/putenv.html) the string given to putenv becomes part of the environment. The string must not be changed or freed.
cmr:v1.7

This commit was SVN r27578.
2012-11-09 16:33:14 +00:00
Nathan Hjelm
8658bbc902 instead of relying on yyterminate to clean up the lex context call the destroy functions directly (after closing the file)
This commit was SVN r27577.
2012-11-09 16:10:55 +00:00
Nathan Hjelm
842caae4c7 Fix a small leak in orte/util/name_fns.c
cmr:v1.7

This commit was SVN r27576.
2012-11-07 23:59:49 +00:00
Aleksey Senin
ae92f64842 Check that MXM runtime version match compiled.
Reviewed by Mike Dubman.

This commit was SVN r27575.
2012-11-07 14:44:33 +00:00
Ralph Castain
9b729794f2 A prior commit apparently broke the trunk when something was inadvertently left behind - so remove a reference to a no-longer-existing function
This commit was SVN r27574.
2012-11-07 11:11:05 +00:00
Nathan Hjelm
51bf75f8e9 fix typo
This commit was SVN r27573.
2012-11-06 21:25:19 +00:00
Nathan Hjelm
1400bab466 fix a couple of errors in r27569
This commit was SVN r27572.

The following SVN revision numbers were found above:
  r27569 --> open-mpi/ompi@f3ce12e71a
2012-11-06 20:06:54 +00:00
Nathan Hjelm
7fb5caea92 Remove the finish_parsing function from various .l files. The function is incomplete (doesn't clean up the lex state) and should be replaced by *_yylex_destroy which correctly cleans up the state.
Checked with the flex 2.5.35. Verified with valgrind that this fixes several "still reachable" leaks.

cmr:v1.7

This commit was SVN r27571.
2012-11-06 19:26:14 +00:00
Nathan Hjelm
bdedd8b0d3 Per RFC modify the behavior of mca_base_components_close to NOT close the output. Modify frameworks to always close their output and set to -1.
Reasoning: The old behavior was a little confusing. mca_base_components_open does not open an output stream so it is a little unexpected that mca_base_components_close does. To add to this several frameworks (that don't use mca_base_components_close) failed to close their output in the framework close function and others closed their output a second time. This change is an improvement to the symantics of mca_base_components_open/close as they are now symetric in their functionality.

This commit was SVN r27570.
2012-11-06 19:09:26 +00:00
Nathan Hjelm
f3ce12e71a Per RFC fix several leaks in opal and ompi. Details below.
pml/v:
  - If vprotocol is not being used vprotocol_include_list is leaked. Assume vprotocol never takes ownership (see below) and always free the string.

coll/ml:
  - (patch verified) calling mca_base_param_lookup_string after mca_base_param_reg_string is unnecessary. The call to mca_base_param_lookup_string causes the value returned by mca_base_param_reg_string to be leaked.
  - Need to free mca_coll_ml_component.config_file_name on component close.

btl/openib:
  - calling mca_base_param_lookup_string after mca_base_param_reg_string is unnecessary. The call to mca_base_param_lookup_string causes the value returned by mca_base_param_reg_string to be leaked.

vprotocol/base:
  - There was no way for pml/v to determine if vprotocol took ownership of vprotocol_include_list. Fix by always never ownership (use strdup).

mca/base:
  - param_lookup will result in storage->stringval to be a newly allocated string if the mca parameter has a string value. ensure this string is always freed.

cmr:v1.7

This commit was SVN r27569.
2012-11-06 18:57:46 +00:00
Jeff Squyres
d6e9a14b14 Fix minor issue: the argv_delete may change the top list pointer. So
be sure to save it.

This commit was SVN r27568.
2012-11-06 16:05:58 +00:00
Nathan Hjelm
906e29ed96 Fix leaks in the opal if posix code. Error paths were not calling OBJ_RELEASE on an opal_if_t created with OBJ_NEW.
This affects both trunk and 1.7 and might affect 1.6.

cmr:v1.7

This commit was SVN r27562.
2012-11-05 20:51:10 +00:00
Ralph Castain
27b41a7db4 If the nodename is an IP address, we need to retain the full name (even if keep_fqdn is false) so that the ssh tree spawn can proceed.
cmr:v1.7

This commit was SVN r27561.
2012-11-05 16:59:53 +00:00
Matthias Jurenz
280930e299 Do forward '--with-cuda[-libdir]' arguments to the VT configure
This commit was SVN r27560.
2012-11-05 14:53:00 +00:00
Mike Dubman
d47d550dfc performance optimization: process completions in the batch manner
This commit was SVN r27559.
2012-11-05 14:02:37 +00:00
Mike Dubman
ca308974e0 Switched FCA collectives component from dlopen to compile-time linking to libfca
This commit was SVN r27557.
2012-11-02 17:30:00 +00:00
Jeff Squyres
4569f77645 Remove redundant common_verbs.h include.
This commit was SVN r27556.
2012-11-02 14:16:55 +00:00
Jeff Squyres
3d05c5cca3 There's no point in having a separate opal_hotel_finalize() function
-- just move that functionality into the hotel destructor.

This commit was SVN r27555.
2012-11-02 14:00:54 +00:00
Jeff Squyres
92e197de73 * MPI_Reduce_scatter_block is now implemented, so remove it from the
"not yet implemented" list (it was already in the mpi_api_files
   macro, actually)
 * Add mistakenly missed palltoall file.  Thanks to Brian for
   noticing its absence.

This commit was SVN r27554.
2012-11-01 13:58:40 +00:00
Mike Dubman
5cdb3654d7 SRQ now supported in ConnectIB
This commit was SVN r27552.
2012-11-01 08:13:56 +00:00
Ralph Castain
3812979315 Remove stale file - we removed the size functions awhile back
This commit was SVN r27551.
2012-11-01 03:36:51 +00:00
Ralph Castain
bc54976f13 Silence warnings when threads are enabled
This commit was SVN r27550.
2012-11-01 03:34:51 +00:00
Brian Barrett
e61c00212d Add files found in svn but not tarball
This commit was SVN r27549.
2012-11-01 02:27:03 +00:00
Brian Barrett
25ccfe1bdd Add autogen.sh to list of files to be distributed, so autogen works off a
tarball

This commit was SVN r27548.
2012-11-01 02:26:28 +00:00
Brian Barrett
dd907e8d4c Remove unused macro
This commit was SVN r27547.
2012-11-01 02:04:57 +00:00
Vishwanath Venkatesan
0e6378bfc9 Modifying the static component accordingly for the modification of interfaces in io_ompio.c
This commit was SVN r27546.
2012-10-31 22:09:21 +00:00
Vishwanath Venkatesan
d1fc22883a Changing the dynamic component accordingly for the modified interfaces
This commit was SVN r27545.
2012-10-31 22:08:25 +00:00
Vishwanath Venkatesan
67463de96f Changing the two-phase component accordingly for the modified interfaces.
This commit was SVN r27544.
2012-10-31 22:07:02 +00:00
Vishwanath Venkatesan
2922fa28a6 Changes to the interface for extracting timing information,
to avoid accessing datastructures across frameworks.

This commit was SVN r27543.
2012-10-31 22:03:05 +00:00
Ralph Castain
ebe45b4b9c Cleanup warnings that may be messing up older compilers, remove unused variables
cmr:v1.7

This commit was SVN r27542.
2012-10-31 14:42:44 +00:00
Ralph Castain
a1c51dc1d6 Wow - fix an error that has been around for a long time. opal_path_access requires a NULL pointer, not an empty string, to correctly operate.
Thanks to Marco Atzeri for chasing this down!

cmr:v1.6,v1.7

This commit was SVN r27539.
2012-10-31 14:10:51 +00:00
Ralph Castain
6f9dd378eb Don't enable both shared and static as you wind up with static executables :-(
This commit was SVN r27530.
2012-10-30 20:36:21 +00:00
Edgar Gabriel
b519990eb8 add a line about the ompio component and the new frameworks into the NEWS file.
This commit was SVN r27529.
2012-10-30 20:23:23 +00:00
Nathan Hjelm
2acd0f83de Revert "Revert r27451 and r27456 - the cmd line parser is incorrectly marking the application as an MCA parameter".
It appears the problem was not with the command line parser but the rsh plm. I don't know why this problem was not occuring before the command line parser changes but it appears to be resolved now.

This commit was SVN r27527.

The following SVN revision numbers were found above:
  r27451 --> open-mpi/ompi@d59034e6ef
  r27456 --> open-mpi/ompi@ecdbf34937
2012-10-30 19:45:18 +00:00
Nathan Hjelm
df9bd0ed59 fix bug in plm/rsh that could add extraneous mca options to the orted argv
cmr:v1.7

This commit was SVN r27526.
2012-10-30 19:40:04 +00:00
Jeff Squyres
a980ee5ca1 Update to r27292: only use "valuelen-1" when the valuelen is
MPI_MAX_INFO_VAL (so that we don't go beyond the end of the string).
But otherwise, abibe by what says in MPI-2.2 / MPI-3.0 in the
description of MPI_INFO_GET (where "it" refers to the valuelen
argument):

    If it is less than the actual size of the value, the value is
    truncated. In C, valuelen should be one less than the amount of
    allocated space to allow for the null terminator.

cmr:v1.7

This commit was SVN r27522.

The following SVN revision numbers were found above:
  r27292 --> open-mpi/ompi@fb4af5e29c
2012-10-30 17:56:28 +00:00
Ralph Castain
7c3a2c3c92 Allow PMI support to find subdirs named lib64 instead of lib - thanks to Guillaume.Papaure of Bull for the patch
This commit was SVN r27519.
2012-10-30 17:38:57 +00:00
Ralph Castain
a080de188f Enable orterun to directly support staged execution, treating each app as a separate job. Support transfer of file maps when support exists.
This commit was SVN r27516.
2012-10-29 23:11:30 +00:00
Ralph Castain
e5e72c3137 Expand the dfs API to support retrieval, loading and purging of file maps.
This commit was SVN r27515.
2012-10-29 23:05:45 +00:00