1
1
Граф коммитов

2463 Коммитов

Автор SHA1 Сообщение Дата
Sayantan Sur
4c122c6649 Added send descriptor management code
This commit was SVN r2639.
2004-09-13 21:56:28 +00:00
Jeff Squyres
388ac0dd4e Since everyone has to recompile almost everything because of changes
to the docs in src/utl/output.h from the last commit, might as well do
this one.  No real rush to update to this, but you'll need to
re-autogen.sh, etc.

Fixes an old, small bug that caused the static-components.h files to
always be regenerated, even if they didn't change.  This caused
several files to be unnecessarily recompiled if you ran the following:

./configure
make
./configure
make

Granted, this is a small error, but a) we have it Right in other
places, b) AC does the Right things for AC_CONFIG_HEADER, so we should
do the Right things too.

This commit was SVN r2638.
2004-09-13 20:29:21 +00:00
Jeff Squyres
d9ae96c259 Update output streams and associated docs.
IMPORTANT NOTE: when sendout ompi_output() strings to files, note that
the file goes into the session directory.  If the session directory
does not exist yet when you call ompi_output() and you are sending the
string to a file, THE STRING WILL BE DROPPED (in terms of the file --
it will still be sent to all other locations that were specified).
This should really only be a factor for a fairly small portion of
things during initialization of a process -- basically, until the
session directory exists, nothing will be written to a file.

ompi_output() will automatically take care of creating the file and
writing to it once the session directory has been created.  So it is
valid to:

id = ompi_output(...something with file set...)
ompi_output(id, "blah!");
/* ... process session directory is created ... */
ompi_output(id, "wazzlebort!");

In this case, the output file will be:

-----
[WARNING: 1 lines lost because the Open MPI process session directory did
 not exist when ompi_output() was invoked]
wazzlebort!
-----

This commit was SVN r2637.
2004-09-13 20:19:01 +00:00
Ralph Castain
6c8654d2b8 Update the utility to a kinder, gentler sort that leaves Jeff's ompi_output files alone. :-)
This commit was SVN r2636.
2004-09-13 19:02:43 +00:00
Vishal Sahay
af6fb8f347 This commit was SVN r2635. 2004-09-13 18:57:25 +00:00
Jeff Squyres
8e14d725ad Change all f2c() functions to use the safe
ompi_pointer_array_get_item() function instead of accessing the array
directly.  This is because accessing the array directly in
multi-threaded situations is unsafe (someone else may be writing to
the array, the array may get moved out from under you via realloc(),
etc.).

This commit was SVN r2634.
2004-09-13 18:14:03 +00:00
Ralph Castain
a14ee7eb48 Checkpoint the console and daemon.
Folks - there appears to be something unreliable about communication with the daemon at the moment. We are trying to track it down. Meantime, please be patient if experimenting with it.

This commit was SVN r2633.
2004-09-13 16:51:53 +00:00
Rich Graham
70231457ce change the code to allow for different fragment size on first, and then
on subsequent fragments.

This commit was SVN r2632.
2004-09-13 15:24:34 +00:00
Edgar Gabriel
d2d6379578 fixing a stupid bug in comm_nextcid: forgot to free the value in the pointer_array in case the others did not agree with my suggestion for the cid.
This commit was SVN r2631.
2004-09-13 15:14:09 +00:00
Edgar Gabriel
64e983a88a fixing logic for lowest_free and number_free
This commit was SVN r2630.
2004-09-13 15:13:13 +00:00
Ralph Castain
57ceb5225e Workaround the mca_oob_ping problem by doing rapid multiple checks - works just fine.
We now have the ability to generate and join a persistent universe. You can create one in two ways:

(a) issue the "openmpi" command. This will fork/exec a seed daemon on your local host. You can specify a universe name or else it will just use the default.

(b) issue the "ompid -seed" command. Starts the seed up directly. Takes all the same options as openmpi.

I will be adjusting mpirun2 and mpi_init to allow connection to existing persistent universes, but they don't do it right now. The ompiconsole program simply issues an exit command to the persistent universe, so you can use it to shut the universe down if you like (or a kill -9  - works too).

This commit was SVN r2629.
2004-09-13 14:14:00 +00:00
Rich Graham
63237b9b26 modifications to the shared memory setup to conform with current
ptl structure.

This commit was SVN r2628.
2004-09-13 13:52:43 +00:00
Jeff Squyres
3964ff873f Add svn:ignore
This commit was SVN r2627.
2004-09-13 13:25:48 +00:00
Tim Woodall
41fa392177 cleanup mess from macro change
This commit was SVN r2626.
2004-09-13 11:55:27 +00:00
Tim Woodall
c592e6af28 include recv request
This commit was SVN r2625.
2004-09-13 11:49:25 +00:00
Tim Woodall
573c45a3be correctly release recv resources in wait/test (e.g decrement communicator reference count)
This commit was SVN r2624.
2004-09-13 11:47:25 +00:00
Edgar Gabriel
0ca2a77b68 fixing some group function with respect to how to determine the rank of a process in the group.
This commit was SVN r2623.
2004-09-13 09:03:10 +00:00
Ralph Castain
e2a5ba5783 Just save this info for now - will use it later. File not included in the makefile.
This commit was SVN r2622.
2004-09-13 01:59:29 +00:00
Vishal Sahay
aae786ce09 Add FINT macros to bindings already implemented
This commit was SVN r2621.
2004-09-13 01:42:53 +00:00
Ralph Castain
55a2576f01 Update some basic functions, mostly with diagnostics.
This commit was SVN r2620.
2004-09-13 01:25:25 +00:00
Brian Barrett
7127f7f5c3 * move the unique_id code from the module to component level, since we don't
really want to fire up the modules early just to get the unique id and it
  will cause all kinds of fun along the way that we don't want.

  Update the pcm components to match (keeping in mind none of them can
  provide any extra info (although some day, I'm sure the ompid one
  could do so)

  Add a base function to access the unique id info.

This commit was SVN r2619.
2004-09-13 00:58:07 +00:00
Jeff Squyres
0ed4c8f663 A break from writing: fix the bug that Brian hacked around for me.
Ensure that OMPI_F77_* are always defined, even if a) a f77 compiler
is not found, or b) the user disables the f77 MPI bindings.

This commit was SVN r2618.
2004-09-11 19:13:43 +00:00
Brian Barrett
4fde8045b3 * F77 linking style #defines don't seem to be defined if f77 isn't enabled,
so remove all that logic in the no F77 bindings case

This commit was SVN r2617.
2004-09-11 18:04:27 +00:00
Jeff Squyres
04932c1666 Fix some f77 string bindings that were missed
This commit was SVN r2616.
2004-09-11 17:19:05 +00:00
Jeff Squyres
88e6d6ee1f Update the comment to match (duh)
This commit was SVN r2615.
2004-09-11 15:04:56 +00:00
Jeff Squyres
9835a2e12e Make coll basic's default priority be 10, not 0. This allows other
components to have a lower priority than the least common denominator.

This commit was SVN r2614.
2004-09-11 15:02:49 +00:00
Jeff Squyres
66bc0c0824 Change line item to "thread support" with values of "solaris",
"posix", or "no" rather than two line items (Solaris and POSIX
threads)

This commit was SVN r2613.
2004-09-11 13:40:36 +00:00
Jeff Squyres
1126e8cf56 Make ompi_info show which fortran symbols it supports
This commit was SVN r2612.
2004-09-11 13:26:51 +00:00
Ralph Castain
c6cbe33d50 Some of these didn't really change - I was just in/out of them for diagnostics while chasing a bug. Got caught by my good buddy Tim again :) on his parse_contact_info function, which requires that the space for the answer be allocated in advance. Sigh. Anyway, mpirun2 now works again. My apologies if you tried it in the last few hours and found it didn't.
Also removed the mpirun3 directory since we are basically dragging mpirun2 along with us - no need to create a new version after all.

Made a few changes to the universe info structure, eliminating the "webserver" and "socket" fields since we will do those contacts through the oob channel. Also changed the "silent_mode" field to "console" since silent mode is the default - the flag needs to tell you to turn the console on, not off.

Parse environ function now gets the ns and gpr replica contact info and loads it in the proper places to hand it off to the respective components, thus allowing me to check connection to them as part of determining if the named universe already exists. Changed the local_universe_exists function accordingly and gave it a new name (since the replicas may not be local). This name will shortly be changed to "ompi_rte_join_universe" as I complete the logic for doing that function.

Please let me know if you see any problems. I successfully ran some trivial multi-process functions in both mpirun2 and singleton modes, and ran the seed daemon as well, so I think it should all be okay.

This commit was SVN r2611.
2004-09-11 12:56:52 +00:00
Jeff Squyres
b72ecd12d5 Add src/tools/console/Makefile to configure.ac since it was just added
to src/tools/Makefile.am's SUBDIRS

This commit was SVN r2610.
2004-09-11 06:39:18 +00:00
Jeff Squyres
c91d66836e No need for DIST_SUBDIRS anymore
This commit was SVN r2609.
2004-09-11 06:05:13 +00:00
Ralph Castain
a60eeef223 Forgot to update the makefile to include the console subdirectory.
This commit was SVN r2608.
2004-09-11 02:53:45 +00:00
Ralph Castain
0071f032fe Update to the rte that ensures the defaults are properly set - in some cases, they weren't being set, leading to unexpecgted behavior when certain environmental variables weren't set. Added some more diagnostic messages to the registry.
Succeeded in contacting an existing persistent universe, and connecting!!! Thanks to Tim for the "ping" function.

First cut at a console - all it does right now is tell the universe to "die", but at least comm is being established.

This commit was SVN r2607.
2004-09-11 02:51:32 +00:00
Brian Barrett
3f11641c67 * don't have ns listed twice in DIST_SUBDIRS. Seems to cause failures for
make maintainer-clean

This commit was SVN r2606.
2004-09-11 01:49:26 +00:00
Weikuan Yu
af008ea157 -- Varios cleanup
This commit was SVN r2605.
2004-09-10 21:02:50 +00:00
Ralph Castain
b2b0a1b8f7 Okay, now got the name acceptable - and compiles too.
This commit was SVN r2604.
2004-09-10 18:18:19 +00:00
Tim Woodall
157082407b added debug flags
This commit was SVN r2603.
2004-09-10 18:16:18 +00:00
Tim Woodall
c042159461 framework for scheduling service - subscribes w/ registry to track
the list of available daemons - actual scheduling logic needs to be
filled in.

This commit was SVN r2602.
2004-09-10 18:15:40 +00:00
Ralph Castain
d816c2620f one more time...
This commit was SVN r2601.
2004-09-10 18:09:11 +00:00
Ralph Castain
f9da579582 Need to do an early commit for a name change.
This commit was SVN r2600.
2004-09-10 18:06:55 +00:00
Brian Barrett
dde3dfee76 * fix stupid typo
* make priority for rms a bit higher
* make rms use NS by default

This commit was SVN r2599.
2004-09-10 17:00:50 +00:00
Brian Barrett
75b3d0691a * reorg the RSH pcm to not have any static data structure (keeping
everything in the "this" structure) so that multiple RSH pcm
  modules can be loaded at once

This commit was SVN r2598.
2004-09-10 16:41:25 +00:00
Brian Barrett
4322c77874 * remove declaration of variable that no longer exists.
This commit was SVN r2597.
2004-09-10 16:38:54 +00:00
Brian Barrett
7d779761ce * fix typo that somehow compiled on OS X... *shrug*
This commit was SVN r2596.
2004-09-10 16:35:27 +00:00
Brian Barrett
aff5acde22 * add some to-dos for the component
This commit was SVN r2595.
2004-09-10 16:01:21 +00:00
Brian Barrett
7557ef7484 * Make the RMS pcm thread safe and enable user threads for the component
- move the list management code into one set of functions that can be
    locked / unlocked as needed
  - clean up some stupidity in code

This commit was SVN r2594.
2004-09-10 15:57:53 +00:00
Brian Barrett
c2d7fab259 * Activate the RMS PCM. It still isn't thread safe, but marks itself as
such, so no reason not to turn it on.
* Add some more comments to the pcm interface about recent changes to
  the interface
* Add some comments to the RMS PCM and move the global list destruct
  from finalize to close, since finalize may be called many times.

This commit was SVN r2593.
2004-09-10 13:48:06 +00:00
Jeff Squyres
ac0043cf6d Remove a lot of kruft that wasn't necessary (looks like a resuly of a
previous cut-n-paste, perhaps from ompi_info's Makefile.am).

This commit was SVN r2592.
2004-09-10 08:46:59 +00:00
Jeff Squyres
7d8e03b34a Odd -- this seems to compile without "util/output.h" manually, but
several test builds failed tonight without it.  [shrug]

This commit was SVN r2591.
2004-09-10 08:46:26 +00:00
Brian Barrett
c8b03b0897 * change the pcm slection to allow for both multiple components to be loaded
at the same time and multiple modules of the same component to be loaded
  at the same time (but not launching procs in the same job).
  - add a "this" pointer to all the PCM functions
  - make base select() function return a list of selected pcms, based on
    given criteria bitmask
  - update all the pcms to match
* Add a insert before position function to the ompi_list code

This commit was SVN r2590.
2004-09-10 04:54:17 +00:00