1
1
Граф коммитов

2287 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
c27d1eeb80 Take out the destructor code I added to ompi_list - may be causing some problems in some pre-existing code, and we have more important issues to deal with for now!
This commit was SVN r2457.
2004-09-02 19:02:16 +00:00
Weikuan Yu
b025e66d7e -- Remove a bug on elan memory allocation
-- Still some bug to remove for Comp Queue support to elan_put

This commit was SVN r2456.
2004-09-02 19:01:11 +00:00
Jeff Squyres
c5d16f215f Also include addr.h
This commit was SVN r2455.
2004-09-02 18:55:22 +00:00
Ralph Castain
3cffa6b549 Inserts some debugging output for use by Tim to help debug a startup problem. Should be transparent to everyone else.
This commit was SVN r2453.
2004-09-02 18:39:42 +00:00
Ralph Castain
e5fab162ce Add tests to ensure that we are not destructing NULL-pointers.
This commit was SVN r2452.
2004-09-02 16:17:32 +00:00
Edgar Gabriel
5286e5205c remove a warning
This commit was SVN r2451.
2004-09-02 16:06:04 +00:00
Ralph Castain
4611c32661 Fix error that trashes stack....
This commit was SVN r2450.
2004-09-02 15:38:24 +00:00
Brian Barrett
7cebd36987 * Make sure we compile oob_tcp_addr
* add sys/types for OS X

This commit was SVN r2449.
2004-09-02 14:45:01 +00:00
Tim Woodall
238822faf3 oops - forgot to check these in
This commit was SVN r2448.
2004-09-02 14:20:13 +00:00
Ralph Castain
1da0e0512a Include <sys/types.h> to make it compile on the Mac OSX.
This commit was SVN r2447.
2004-09-02 13:26:39 +00:00
Jeff Squyres
ce740d191d Modify svn:ignore
This commit was SVN r2446.
2004-09-02 09:05:33 +00:00
Jeff Squyres
668287ef8f Need <unistd.h> for OS X.
This commit was SVN r2445.
2004-09-02 08:35:05 +00:00
Weikuan Yu
b70e1b05d0 -- To work more on debugging, e.g, mca_ptl_elan_putget_desc_construct ()
This commit was SVN r2444.
2004-09-02 06:25:52 +00:00
Jeff Squyres
15d6462bb1 Remove files from Makefile.am that are no longer in the repository
This commit was SVN r2443.
2004-09-02 06:02:09 +00:00
Weikuan Yu
7223524e27 -- Save the orginal test file as check.c
-- Simplify the basic ping-pong latency program.

This commit was SVN r2442.
2004-09-02 05:16:59 +00:00
Weikuan Yu
d03f3b4d36 -- Fix known bugs in data integrity.
This commit was SVN r2441.
2004-09-02 04:08:14 +00:00
Ralph Castain
03096df34a Correct a mis-match in a parameter passed to one of the functions. Somehow, the parameter list to construct a notify message got twisted to passing key values instead of tokens. This meant that the parameters sent to "get" were also incorrect, guaranteeing that the whole notify system would fail.
Not sure how the mistake got in there - possibly just some lack of coordination. Anyway, it has now been corrected.

This commit was SVN r2440.
2004-09-02 01:44:19 +00:00
Vishal Sahay
9dd541ef41 More f77 bindings
This commit was SVN r2439.
2004-09-02 01:17:31 +00:00
Sayantan Sur
662ba7c344 Basic MPI_Init() working now for 2 processes.
For any number of processes, dynamic connection
management needs to be in place.
This will be done using the OOB async interface.

The basic ptl_send and recv functions for short messages
are working, though in very preliminary form.

This commit was SVN r2438.
2004-09-02 00:39:30 +00:00
Tim Woodall
a5cf56c425 add support for packing/unpacking ompi_process_name_t as OMPI_NAME
This commit was SVN r2437.
2004-09-01 23:09:41 +00:00
Tim Woodall
9e31ea786c implemented timedwait for polling (non-pthread) case
This commit was SVN r2436.
2004-09-01 23:08:58 +00:00
Tim Woodall
083555cb2a dont try to allocate 0 bytes
This commit was SVN r2435.
2004-09-01 23:08:19 +00:00
Tim Woodall
f43f072673 - further integration w/ gpr - use synchro/subscribe to asynchronously populate
a cache of peer addresses as they come online
- resolved issues with the event library

This commit was SVN r2434.
2004-09-01 23:07:40 +00:00
Jeff Squyres
6548b50fce Arrrrgghhh!!! Fix typo.
This commit was SVN r2433.
2004-09-01 22:28:23 +00:00
Weikuan Yu
59df76dd7b -- Remove a bug in the threading code due to the truncation of hdr_type
-- Debugging is not fun :-((

This commit was SVN r2432.
2004-09-01 20:57:12 +00:00
Jeff Squyres
ad8b787341 Commit this fix so that the build tests can start building.
This commit was SVN r2431.
2004-09-01 19:26:28 +00:00
Weikuan Yu
7f1815f5b1 -- Add a ctrl QDMA to queue structure to avoid using buff-equipped desc.
This commit was SVN r2430.
2004-09-01 17:13:10 +00:00
Tim Woodall
e06857ad6c - removed debug
- fixed allocation errors

This commit was SVN r2429.
2004-09-01 16:42:43 +00:00
Ralph Castain
854ac21038 Provide a couple of new synchro options:
1. greater than or equal
2. less than or equal

Adjust ascending/descending mode to require transition through level. Change initial checks to only check levels.

This commit was SVN r2428.
2004-09-01 15:27:50 +00:00
Tim Woodall
ed7e103a96 return status
This commit was SVN r2427.
2004-09-01 15:25:27 +00:00
Jeff Squyres
b467fe225b Need <unistd.h> for OS X
This commit was SVN r2426.
2004-09-01 14:54:34 +00:00
Tim Woodall
fe92d2bf3c check for null pointers
This commit was SVN r2425.
2004-09-01 14:36:53 +00:00
Tim Woodall
6aa9388350 minor corrections
This commit was SVN r2424.
2004-09-01 14:22:06 +00:00
Jeff Squyres
330d8aa52d Make that message only print if we're in debugging mode (avoid
spurrious cron e-mails to the job owner :-) )

This commit was SVN r2423.
2004-09-01 08:59:13 +00:00
Vishal Sahay
d96c602465 More wrappers with MPI_Fint modifications
This commit was SVN r2422.
2004-09-01 00:57:55 +00:00
Ralph Castain
81a79e27f3 A change to the ompi_list system that *should* be transparent. It shows no affect on anything in the unit test, and works with everything I have used to test it.
The change is to the destructor for ompi_list_t - it now cleans releases all objects attached to the list. Before this change, you had to be careful to run the list of items yourself, releasing all of them, before calling OBJ_DESTRUCT on the head of the list - or else "leak" all the item memory. With this change, you don't have to do that yourself. The destructor takes care of it for you.

Please let me know if anyone observes undesirable behavior from the change.
Ralph

This commit was SVN r2421.
2004-08-31 20:52:18 +00:00
Ralph Castain
50fbf1d184 Clear up some silly compiler warnings.
This commit was SVN r2419.
2004-08-31 19:31:48 +00:00
Ralph Castain
0241d07866 Some upgrades to the registry system to help poor ol' Tim:
1. added a NOTIFY_PRE_EXIST flag that tells subscribe to check the registry for pre-existing entries that meet the specified conditions and return them

2. synchro now initializes its trigger count to the number of entries that meet conditions at the time synchro is called, instead of zero

3. notify messages include info on what caused the trigger event.

This commit was SVN r2418.
2004-08-31 18:59:16 +00:00
Tim Woodall
7b55e46f83 file needed for mca magic
This commit was SVN r2416.
2004-08-31 18:43:03 +00:00
Weikuan Yu
19ac676084 -- Just got basic Thread-based progression working
-- Will focus on more extensive testing

This commit was SVN r2415.
2004-08-31 18:37:06 +00:00
Laura Casswell
8b3b1c5414 add ompi_pack
This commit was SVN r2414.
2004-08-31 18:13:30 +00:00
Weikuan Yu
44febd0a86 -- Checkin more code to detect events being fired
-- To double check thread-triggered progress and join 

This commit was SVN r2413.
2004-08-31 17:45:33 +00:00
Jeff Squyres
a691dfbb4e Oops -- also remove the Makefile.am stuff for paramparse
This commit was SVN r2412.
2004-08-31 11:37:46 +00:00
Jeff Squyres
6abeab27ce Oops -- didn't mean to commit this. :-)
This commit was SVN r2411.
2004-08-31 11:37:20 +00:00
Jeff Squyres
7bb9f8d78d Oops -- ensure that we keep the "last tarball tested" file on a
per-host / per-config basis because some people (like me!) have a
common scratch root on NFS for multiple machines (e.g., share a
downloads directory).

This commit was SVN r2410.
2004-08-31 11:27:53 +00:00
Jeff Squyres
4faa486ee5 - New capability for MCA parameters to be read from files. Order of
resolution is now (effectively):
  - read from MPI keyval (if associated)
  - read from command line
  - read from environment
  - read from file
  More capabilities will be added shortly for developers to
  programatically set MCA parameter values
- Create [empty but commented] system-wide MCA param file that gets
  installed at $sysconf/openmpi-mca-params.conf (i.e.,
  $prefix/etc/openmpi-mca-params.conf)
- The following files are opened and read (in order):
  1. $sysconf/openmpi-mca-params.conf
  2. $HOME/.openmpi/mca-params.conf
  Specifically, the values in 2) will override the values in 1) (so
  users can override system-wide defaults.
- Update MCA string params to allow for "~/" in the middle to be
  expanded to the user's home directory.
- Added to default value of MCA parameter component_path to be:
    $pkglibdir:$HOME/.openmpi/components
  $pkglibdir is typically $prefix/lib/openmpi.  So now both of these
  directories will be searched for components at run time (in all Open
  MPI executables, including ompi_info).  Note that this is an MCA
  parameter, so it, too, can be changed at run-time.
- Updated all docs to match this behavior -- some is \internal, so it
  doesn't show unless you tell doxygen to generate internal docs
  (which is not the default)
- update ompi_info to handle new behavior; ompi_info --param now shows
  the *current* default value (i.e., it will look in the environment
  and/or files to find out what the default values are -- so if you
  change a value in $HOME/.openmpi/mca-params.conf, it should be
  reflected in "ompi_info --param all all")
- updated bunches of doxygen docs to match
- this diff is slightly artifically large -- some of the changes are
  converstions to 4 space tabs (I re-indented my own code that
  previously used 2 space tabs)

This commit was SVN r2408.
2004-08-31 09:49:56 +00:00
Jeff Squyres
1569ecdba3 Convert this flex scanner to prefix all of its symbols with
"mca_llm_base_yy" so that we can have multiple flex scanners in one
library.

This commit was SVN r2407.
2004-08-31 09:38:27 +00:00
Jeff Squyres
639882b063 Plug a minor memory leak
This commit was SVN r2406.
2004-08-31 09:37:28 +00:00
Jeff Squyres
2e31e2aaa6 Several improvements to the nightly build script:
- ensure to only keep a finite number of old build roots around (build
  roots are kept if you --leave-install or if there was a build error
  -- so you can go look and see what the error was, etc.)
- check to see if the last run already built the same tarball version
  that's currently on the nightly build website -- if we've already
  tested it, there's no point in compiling / testing it again.  This
  should cut down on the volume of [repetitive] testing e-mails to the
  mailing list. 
- add --force parameter to override ^^ -- *always* compile/test,
  regardless of whether we have tested that version before.
- fix logic removing old tarballs

This commit was SVN r2405.
2004-08-31 09:36:56 +00:00
Jeff Squyres
20b3898390 Remove a bunch of code and replace it with a single line:
ompi_argv_split().  Mmmm... the Goodness of a rich internal API...
tasty!

This commit was SVN r2404.
2004-08-31 09:34:12 +00:00