mquinson [Thu, 1 Jun 2006 23:20:38 +0000 (23:20 +0000)]
cosmetics, and make two different error messages look different
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2345
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 23:20:01 +0000 (23:20 +0000)]
New function to get the element count of a set
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2344
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 23:17:49 +0000 (23:17 +0000)]
Exchange master socket port number on client socket creation: gras_socket_peer_port on an accepted socket will point to the port on which the peer is listening, not the one on which he is speaking since the latter is assigned by the system and meanless to human beings
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2343
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 23:07:50 +0000 (23:07 +0000)]
GRAS is not the only one using libdata. AMOK does too, and other could. So, it is possible that some fabrics get registered after the call to gras_procdata_init. In that case, the corresponding libdata won't get created on time, and we should create them afterward on need. Plus some cosmetics and bunch of paranoid self-checking
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2342
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 21:49:04 +0000 (21:49 +0000)]
virtu needs this
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2341
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 21:48:46 +0000 (21:48 +0000)]
fixme--: found a better place
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2340
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 20:31:17 +0000 (20:31 +0000)]
allow to request remote measurement with a min_duration (that was easy, apparently)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2339
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 20:28:38 +0000 (20:28 +0000)]
Port to the new amok_bw API (with min_duration)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2338
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 20:21:03 +0000 (20:21 +0000)]
Well. RPC answer need some extra love.
(the answer type is not the request type, stupid me)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2337
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 20:11:22 +0000 (20:11 +0000)]
And now, do it right. Change the experience total size instead of the individual message one; do not tend to the right value, but try to eventually pass it since it's our ending condition. I swear I did test previous version (but poorly, apparently)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2336
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 19:59:23 +0000 (19:59 +0000)]
Plug a bunch of memleaks, and play with the new min_duration argument
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2335
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 19:58:37 +0000 (19:58 +0000)]
In amok_bw_test, it is now possible to ask that the measurement lasts at least N seconds (N being a double). If it is not the case with the provided msg_size, it will get adjusted and a new test is made (using the same measurement sockets so you won't pay the handshaking twice). This mecanism is still to be offered in amok_bw_request(), but I'm on it.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2334
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 19:55:10 +0000 (19:55 +0000)]
yet some more debug
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2333
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 19:54:19 +0000 (19:54 +0000)]
ctor/dtor of msg_cb_t + an new function to wait for any of a set of message types (+logging cleanups)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2332
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 19:52:51 +0000 (19:52 +0000)]
ctor/dtor of msg_cb_t + an new function to wait for any of a set of message types
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2331
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 19:50:53 +0000 (19:50 +0000)]
New functions to search in dynars (plus some cosmetics)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2330
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 1 Jun 2006 19:50:08 +0000 (19:50 +0000)]
New functions to search in dynars
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2329
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 30 May 2006 08:32:18 +0000 (08:32 +0000)]
Next version will be 3.1
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2328
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 21:57:03 +0000 (21:57 +0000)]
Now that MSG uses sain units, we don't have to convert anymore
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2327
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 21:51:17 +0000 (21:51 +0000)]
Now that MSG uses sain units, we don't have to convert anymore
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2326
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 21:45:23 +0000 (21:45 +0000)]
program expects name:port on each maestro argument, not one argument for each
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2325
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 29 May 2006 21:39:26 +0000 (21:39 +0000)]
Providing up-to-date xml files... Version 1 with units in Bytes, Flops and Seconds.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2324
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 29 May 2006 21:34:55 +0000 (21:34 +0000)]
Ask for version 1. Default is version 0.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2323
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 21:13:02 +0000 (21:13 +0000)]
don't weep you little user, we are no bastards, we only look like
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2322
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 29 May 2006 20:27:55 +0000 (20:27 +0000)]
Keeping up with last surfxml modifications.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2321
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 29 May 2006 20:01:09 +0000 (20:01 +0000)]
Update changelog. Move the most important modifications to the beginning.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2320
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 29 May 2006 19:53:17 +0000 (19:53 +0000)]
Adding a "version" tag to platform_description. This tag is used to issue a BIG FAT WARNING when people use old platforms with wrong units.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2319
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 18:59:57 +0000 (18:59 +0000)]
Plug a memleak
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2318
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 29 May 2006 14:43:48 +0000 (14:43 +0000)]
retrive -> retrieve
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2317
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 13:31:38 +0000 (13:31 +0000)]
Make a log message more explicit
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2316
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 12:33:06 +0000 (12:33 +0000)]
Reduce default verbosity, shorten a message
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2315
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 12:29:43 +0000 (12:29 +0000)]
Yeah, right. I'm plainly stupid and cannot come up with the right boolean
expression in less than 325325 iterations. So what?
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2314
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 12:26:39 +0000 (12:26 +0000)]
Take arguments as <host>:<port> (=cleanups); reduce default verbosity
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2313
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 12:16:34 +0000 (12:16 +0000)]
Do obey messages asking to stop the saturation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2312
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 11:55:51 +0000 (11:55 +0000)]
Yet some more logging
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2311
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 10:00:47 +0000 (10:00 +0000)]
No timeout for saturations in full_fledged
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2310
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:59:55 +0000 (09:59 +0000)]
Should compile before committing (sorry for the spam)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2309
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:58:00 +0000 (09:58 +0000)]
Allow endless saturation experiment (by passing duration=0)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2308
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:55:40 +0000 (09:55 +0000)]
Some more debugging informations
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2307
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:44:51 +0000 (09:44 +0000)]
- Some more debugging info
- Only use kasher debugging info source to avoid the segfault
- Add a FIXME comment since the proper solution is not this hack but something
in GRAS itself
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2306
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:32:00 +0000 (09:32 +0000)]
Tell who sent the discarded messages
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2305
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:28:50 +0000 (09:28 +0000)]
Better choice of what is debug and what is verbose
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2304
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:25:17 +0000 (09:25 +0000)]
kill old cruft moved since a looooong time to a better place
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2303
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:23:58 +0000 (09:23 +0000)]
found one source of issues in RL
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2302
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:20:32 +0000 (09:20 +0000)]
Declare the payload we expect (stupid me)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2301
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 29 May 2006 09:04:10 +0000 (09:04 +0000)]
Let's shild the user against himself: check that he provided a payload variable when he is supposed to (and only when he's supposed to)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2300
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 29 May 2006 07:54:13 +0000 (07:54 +0000)]
An additionnal example that causes a weird behavior when doing a full_fledge_saturation.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2299
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 28 May 2006 21:34:46 +0000 (21:34 +0000)]
Let's be really verbose about what gets copied on the socket
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2298
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 28 May 2006 20:42:32 +0000 (20:42 +0000)]
Unregister failed sockets from the big socket dynar
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2297
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Fri, 26 May 2006 19:53:02 +0000 (19:53 +0000)]
Moving platform_generation to contrib/
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2296
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Fri, 26 May 2006 15:40:39 +0000 (15:40 +0000)]
Add new convenient functions.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2295
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 25 May 2006 20:20:30 +0000 (20:20 +0000)]
we need libtool too to compile the CVS (but who don't have libtool on his box anyway)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2294
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 24 May 2006 12:22:21 +0000 (12:22 +0000)]
Stupid me, use an existing channel as external default
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2293
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 24 May 2006 12:13:12 +0000 (12:13 +0000)]
Sanitize the log channels naming scheme in GRAS too
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2292
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 24 May 2006 10:06:35 +0000 (10:06 +0000)]
Search typedefs of pointer to function where they now live
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2291
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 24 May 2006 10:00:17 +0000 (10:00 +0000)]
Search typedefs of pointer to function where they now live
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2290
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 24 May 2006 09:56:55 +0000 (09:56 +0000)]
A new datacontainer: the matrix
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2289
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 24 May 2006 09:39:56 +0000 (09:39 +0000)]
Move typedefs of pointer to function to their own file so that I find them when I want to
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2288
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 22:37:47 +0000 (22:37 +0000)]
Shut automake up
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2287
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 21:54:02 +0000 (21:54 +0000)]
Found the bug: the 'from' variable was used to receive a message from the master at the beginning of step, and then to exchange data between peers. So at the end of the step, when the slave wanted to answer to the master using it, it pointed to somewhere else (plus a whole bunch of cosmetics)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2286
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 21:39:12 +0000 (21:39 +0000)]
Cosmetics: rename the initial message to 'assignment' since it's kinda what it is, and since it sounds better
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2285
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 21:18:43 +0000 (21:18 +0000)]
Pass the neighbor in the enroll message, not on the command line
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2284
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 20:37:35 +0000 (20:37 +0000)]
Allow to exchange fixed-sized arrays which size is \#defined
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2283
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 19:21:09 +0000 (19:21 +0000)]
Learn how to exchange xbt_host_t data
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2282
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 18:47:04 +0000 (18:47 +0000)]
New function: xbt_host_copy (even if it's trivial)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2281
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 22 May 2006 16:39:41 +0000 (16:39 +0000)]
Sorry for the noise. This script should not stay in this place.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2280
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 22 May 2006 16:35:23 +0000 (16:35 +0000)]
Putting it here in the meantime.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2279
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 22 May 2006 13:54:42 +0000 (13:54 +0000)]
A few more tiers file and a convenient perl script for the alnem project. I
put it here because it uses these .pm files.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2278
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 22 May 2006 13:52:40 +0000 (13:52 +0000)]
More careful inclusion of files in the distribution.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2277
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 22 May 2006 13:51:59 +0000 (13:51 +0000)]
Can't stand Mt modified this file after me! ;)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2276
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 09:11:46 +0000 (09:11 +0000)]
stupid me. col means row in french, so use line/row and not row/col, which means nothing
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2275
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 22 May 2006 09:02:31 +0000 (09:02 +0000)]
stupid me. col means row in french, so use line/row and not row/col, which means nothing
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2274
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 21:23:09 +0000 (21:23 +0000)]
Compute the slave position in the grid on the fly instead of passing it on command line (plus cosmetics such as typos hunting)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2273
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 21:07:57 +0000 (21:07 +0000)]
first step to separate the host matrix size from the data matrix sizes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2272
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 21:04:12 +0000 (21:04 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2271
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 20:42:34 +0000 (20:42 +0000)]
reindent
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2270
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 20:40:54 +0000 (20:40 +0000)]
add the parallel matrix multiplication example (plus cosmetics)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2269
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 20:40:14 +0000 (20:40 +0000)]
add the parallel matrix multiplication example
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2268
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 20:38:39 +0000 (20:38 +0000)]
integrate to the distribution
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2267
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 21 May 2006 20:35:11 +0000 (20:35 +0000)]
Rename maestro/sensors to master/slaves
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2266
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
amad206 [Sun, 21 May 2006 19:23:38 +0000 (19:23 +0000)]
correct pmm_deployment.xml
correct pmm.c
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2265
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
amad206 [Sat, 20 May 2006 11:57:32 +0000 (11:57 +0000)]
add comments to pmm example & modify pmm_deployment.xml ( Function : Maestro -> maestro)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2264
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 20 May 2006 10:03:59 +0000 (10:03 +0000)]
I cannot stand that Arnaud commited more recently than me on this file
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2263
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 20 May 2006 09:09:56 +0000 (09:09 +0000)]
Update slides' reference.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2262
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
amad206 [Sat, 20 May 2006 08:42:39 +0000 (08:42 +0000)]
add the "test_sg" file : to use "examples/msg/msg_platform.xml"
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2261
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 20 May 2006 08:18:27 +0000 (08:18 +0000)]
Flavien's platform generator. It can produce lines, rings, tores, meshes, hypercubes and such
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2260
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
amad206 [Sat, 20 May 2006 08:06:44 +0000 (08:06 +0000)]
example of parallel matrix multiplication : I modified the pmm.c by adding messages of Ack (from sensor to maestro) to indicate the
end of each step
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2259
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 20 May 2006 07:11:13 +0000 (07:11 +0000)]
Adding a FAQ entry on the CVS compiling.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2258
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
amad206 [Fri, 19 May 2006 20:23:00 +0000 (20:23 +0000)]
*** empty log message ***
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2257
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 15:46:33 +0000 (15:46 +0000)]
Add a big fat warning to users seeming to use the CVS without --enable-maintainer-mode since it's a probable source of trouble
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2256
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
amad206 [Fri, 19 May 2006 13:40:00 +0000 (13:40 +0000)]
*** empty log message ***
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2255
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 10:36:26 +0000 (10:36 +0000)]
Modify the experimentation suite so that it can test the broadcast stuff
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2254
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 10:21:08 +0000 (10:21 +0000)]
Roots of the experimentation infrastructure for Ahmed
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2253
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 10:20:26 +0000 (10:20 +0000)]
update to pass the arguments the processes expect
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2252
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 10:19:56 +0000 (10:19 +0000)]
Accept one peer (described as hostname:port) per argument
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2251
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 09:44:14 +0000 (09:44 +0000)]
And now, this damn 5-lines function without trivial bug. :( Sorry for the spam
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2250
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 09:35:23 +0000 (09:35 +0000)]
New function: xbt_host_from_string() creating an xbt_host_t from a 'toto:42042' string
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2249
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 09:33:52 +0000 (09:33 +0000)]
Enable timestamping of messages emitted out of any user process (such as the terminaison message in MSG)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2248
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 09:21:55 +0000 (09:21 +0000)]
Various cleanups; dupplicate the processes on each machine since MSG (and thus GRAS/SG) have a 1-port model. A given process cannot send and receive at the same time. this leaded to a whole bunch of deadlocks when trying to do an uncoordoned all2all exchange. And synchronisation delays would have falsified the timing
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2247
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 19 May 2006 07:48:11 +0000 (07:48 +0000)]
cosmetics: rename alltoall to all2all
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2246
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
