mquinson [Mon, 18 May 2009 08:32:55 +0000 (08:32 +0000)]
Be more verbose when propagating local exceptions
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6287
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 18 May 2009 08:32:48 +0000 (08:32 +0000)]
Add regression tests for the lastly fixed bugs in GRAS dadadesc
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6286
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 18 May 2009 08:32:41 +0000 (08:32 +0000)]
Allow the exchange of 0-long dynamic vectors
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6285
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 18 May 2009 08:32:35 +0000 (08:32 +0000)]
Less debugging info
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6284
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 18 May 2009 08:32:29 +0000 (08:32 +0000)]
Reindent some code, no real change (should do it for all my code once for good)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6283
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 18 May 2009 08:32:22 +0000 (08:32 +0000)]
free the exception we display + suporious update of the output
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6282
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 18 May 2009 08:32:17 +0000 (08:32 +0000)]
xbt_ex_display(): do not free the exception after displaying
This allows to do more with the given exception afterward.
Users should call xbt_ex_free() themselves.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6281
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 18 May 2009 08:32:10 +0000 (08:32 +0000)]
Reindent, no real change
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6280
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 11 May 2009 22:37:59 +0000 (22:37 +0000)]
Fix a bug in task exchange which broke MSG_task_get_sender()
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6277
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 11 May 2009 22:37:40 +0000 (22:37 +0000)]
more informative debug messages
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6276
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 7 May 2009 12:16:23 +0000 (12:16 +0000)]
Add an example of process migration in examples/msg/migration
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6275
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 7 May 2009 12:16:00 +0000 (12:16 +0000)]
Document last changes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6274
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 7 May 2009 12:15:58 +0000 (12:15 +0000)]
Plug a bunch of memleaks
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6273
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 5 May 2009 16:12:23 +0000 (16:12 +0000)]
ooops, typo
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6272
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 5 May 2009 16:10:10 +0000 (16:10 +0000)]
draft implementation of the 4 basic actions
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6271
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 5 May 2009 15:59:49 +0000 (15:59 +0000)]
first draft of the action runner
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6270
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:15:32 +0000 (10:15 +0000)]
Initiate the 3.4 development cycle
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6267
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:10:55 +0000 (10:10 +0000)]
Use the aligned size for structures instead of raw size, so that GRAS can exchange arrays of heterogeneous structures
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6266
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:10:53 +0000 (10:10 +0000)]
Some more debug
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6265
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:10:30 +0000 (10:10 +0000)]
More explicit error reporting around the trace connection mechanism (yeah, just cosmetics)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6264
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:10:28 +0000 (10:10 +0000)]
little improvement to the documentation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6263
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:10:05 +0000 (10:10 +0000)]
Add SIMIX_process_set_name()
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6262
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:10:03 +0000 (10:10 +0000)]
Change form of error message so that when host name is empty, you get it right at first glance without thinking that the error message forgot to tell you what the borken name was
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6261
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:09:40 +0000 (10:09 +0000)]
Also include strbuff from xbt.h
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6260
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 28 Apr 2009 10:09:38 +0000 (10:09 +0000)]
RELEASING VERSION 3.3
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6259
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 12 Apr 2009 02:27:20 +0000 (02:27 +0000)]
plug some memleaks
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6245
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 12 Apr 2009 02:27:15 +0000 (02:27 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6244
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 12 Apr 2009 01:33:43 +0000 (01:33 +0000)]
do not brutally kill the listener but ask him politely to die (extending the awaking mecanism). If not, it fails to release the locks it holds on darwin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6243
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 12 Apr 2009 01:04:45 +0000 (01:04 +0000)]
do not close sockets from main thread, but from listener (queue added for that) because the listener is selecting on that sock, and it makes darwin angry to close selected sockets
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6242
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 12 Apr 2009 01:04:40 +0000 (01:04 +0000)]
reduce the verbosity of *all* setsockopt failures to set buffer sizes to keep darwin's mouth shut
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6241
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 23:41:02 +0000 (23:41 +0000)]
Also add config.log to checkall.log to get configure option used for the test, at least
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6240
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 23:28:40 +0000 (23:28 +0000)]
add the svn version to the checkall log. I'm getting lost out there
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6239
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 23:19:46 +0000 (23:19 +0000)]
Erm. We moved from CVS to SVN a looooong while ago
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6238
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 23:12:30 +0000 (23:12 +0000)]
lower the gravity of the message indicating that setsockopt failed to set the buffer sizes. It do fail on darwin, but I doubt nowadays that this setting does any good. OSes know better than me what the good buffer size should be. Maybe we should kill that completely
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6237
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 23:12:05 +0000 (23:12 +0000)]
do not use -ffast-math, it sometimes breaks lagrangian computations on 64bits
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6236
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 23:12:01 +0000 (23:12 +0000)]
fixup the timings after the lastest updates
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6235
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 17:08:20 +0000 (17:08 +0000)]
typo--
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6234
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:40:54 +0000 (16:40 +0000)]
tiny documentation improvement
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6233
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:40:49 +0000 (16:40 +0000)]
Also include the str.h header from main xbt.h
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6232
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:40:44 +0000 (16:40 +0000)]
allow to define a default value to registered functions. This allows to use generic functions deciding what to do from the name used, and thus deal with unknown function names in deployment files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6231
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:40:39 +0000 (16:40 +0000)]
fix previous commit on gras_userdata_new() by using the result of gras_userdata_set() instead of calling gras_userdata_get right after
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6230
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:40:13 +0000 (16:40 +0000)]
make sure that the gras_userdata_new() macro returns what just got malloced
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6229
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:40:08 +0000 (16:40 +0000)]
a new disabiguisation on the GRAS communication model
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6228
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:40:02 +0000 (16:40 +0000)]
gras_userdata_new expects the pointed type, not the pointer type. Fix tutorial, and add a bunch of warnings in the doc
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6227
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:39:37 +0000 (16:39 +0000)]
Add a link from the documentation down to the relevant tutorial section
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6226
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:39:32 +0000 (16:39 +0000)]
ignore generated files from our tutorial file generation dept
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6225
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 11 Apr 2009 16:39:27 +0000 (16:39 +0000)]
Regenerate the output captures of the gras tutorial after the numerous timing changes we had in the kernel
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6224
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 8 Apr 2009 10:32:08 +0000 (10:32 +0000)]
tiny improvement of the gras_comm figure
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6223
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 8 Apr 2009 10:22:13 +0000 (10:22 +0000)]
Put associate the INRIA keyword to contributers where needed
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6222
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 8 Apr 2009 10:22:10 +0000 (10:22 +0000)]
Add some more explaination about the network model used in GRAS, featuring a 'neat' figure
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6221
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 8 Apr 2009 10:22:07 +0000 (10:22 +0000)]
Little reword in the introduction (switch to current slide location), new section on dev snapshots found on my web page
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6220
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 6 Apr 2009 08:45:19 +0000 (08:45 +0000)]
also ignore java cruft in git
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6193
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 6 Apr 2009 08:45:12 +0000 (08:45 +0000)]
Also work with older test binary which don't understand the -a flag
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6192
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 6 Apr 2009 08:45:05 +0000 (08:45 +0000)]
Allow autoconf stuf to be run from another directory
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6191
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 6 Apr 2009 08:44:57 +0000 (08:44 +0000)]
-Wall by default
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6190
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 6 Apr 2009 08:44:47 +0000 (08:44 +0000)]
Give tesh the ability to get the CWD to use from the --cd command line argument
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6189
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Fri, 27 Mar 2009 13:22:57 +0000 (13:22 +0000)]
Cope with model best gtnets fit, that is latency_factor = 10.4, bandwidth_factor = 0.92 and weight_S_parameter = 8775.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6187
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Fri, 27 Mar 2009 13:22:55 +0000 (13:22 +0000)]
Cope with model best gtnets fit, that is latency_factor = 10.4, bandwidth_factor = 0.92 and weight_S_parameter = 8775.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6186
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Fri, 27 Mar 2009 13:22:53 +0000 (13:22 +0000)]
Cople with new gtnets.c and deployment files.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6185
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Fri, 27 Mar 2009 13:22:51 +0000 (13:22 +0000)]
Reduced message size in dogbone topollogy to speed check.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6184
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Fri, 27 Mar 2009 13:22:49 +0000 (13:22 +0000)]
Added a parameter for each process, given flow id, cople with gtnets.c .
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6183
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 26 Mar 2009 15:32:15 +0000 (15:32 +0000)]
also awake the listener after opening a file socket
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6174
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 26 Mar 2009 13:37:06 +0000 (13:37 +0000)]
Revalidate some forgotten 32bit outputs after gras listener improvement
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6173
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 26 Mar 2009 13:37:04 +0000 (13:37 +0000)]
Deal with negative indices the same way than too big indices
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6172
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 26 Mar 2009 13:37:02 +0000 (13:37 +0000)]
revalidate the 64bits test outputs after the listener modification
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6171
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 26 Mar 2009 08:08:31 +0000 (08:08 +0000)]
dynars: use inline functions instead of macros + release the lock on bound error
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6170
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 16 Mar 2009 11:34:55 +0000 (11:34 +0000)]
Removing these trailing spaces seems to help doxygen. I can't believe it! And it took me a while before figuring out where the problem came from.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6156
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 16 Mar 2009 11:34:30 +0000 (11:34 +0000)]
Add publication (Thanks Derrick!) and fix count in the table.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6155
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:39 +0000 (15:02 +0000)]
ignore more ignorable
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6154
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:37 +0000 (15:02 +0000)]
kill generated files out of the svn, damnit
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6153
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:36 +0000 (15:02 +0000)]
Revalidate the outputs I can test on my machine (build daemons are currently dead)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6152
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:34 +0000 (15:02 +0000)]
Gras listener thread of each process do select(-1) instead of while(1) {select(0.5)}
So, it won't get active before a message actually arrived.
The previous design was because when the main thread opens a server
socket, it has to be added to the set of sockets select()ed by the
listener. My solution is to add a new socket between main and listener
threads. Each time the main opens a socket, it writes something on
this socket, unlocking the listener. Who detects the fact that it's
not a real message to handle, and get back to select(-1)
(automagically adding the new socket to the set).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6151
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:32 +0000 (15:02 +0000)]
be more explicit when we die because we got an unknown message type (plus eclipse reindenting, sorry)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6150
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:30 +0000 (15:02 +0000)]
no need to sleep(1), it slows down my tests
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6149
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:28 +0000 (15:02 +0000)]
don't open master socket before we declared the messages, or remote may be faster than us leading to an exception 'unknown message type' in RL (plus eclipse reindenting, sorry)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6148
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 10 Mar 2009 15:02:26 +0000 (15:02 +0000)]
Fix a small bug found by eclipse, and change some details to remove false positives from eclipse
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6147
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 10 Mar 2009 12:02:57 +0000 (12:02 +0000)]
Add references from myself, Loris and Kurt.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6146
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Mar 2009 22:46:16 +0000 (22:46 +0000)]
fix up synchronization of documentation on web site
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6145
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Mar 2009 15:38:57 +0000 (15:38 +0000)]
typo in unit test name
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6144
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Mar 2009 15:38:51 +0000 (15:38 +0000)]
Readd the example with pointed data in the documentation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6143
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 24 Feb 2009 08:44:05 +0000 (08:44 +0000)]
Fix the bug on MSG_task_put_bounded raised by Fredo, Henri and Matthieu.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6142
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 24 Feb 2009 08:44:02 +0000 (08:44 +0000)]
Avoid C++ comments.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6141
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
casanova [Sun, 15 Feb 2009 18:46:07 +0000 (18:46 +0000)]
Fixed a race condition in msg for communication between two processes. The
problem was with task->simdata->comm being used and then freed by the
receiver, and being used by the sender. The problem was that the sender
could be awakened after the receiver, and thus try to read de-allocated
memory.
The bug was fixed by adding a refcount field to task->simdata->comm, to ensure
that the freeing can be done by either the received and the sender, and always
by the last one to be awakened. The refcount is initialized to 2, each process
reduces it by one, and the last one does the free. Setting the refcount to 2 is
a bit of a hack, but in the end seemed cleaner than sprinkling refcount++ and refcount--
all over the code, especially because this is already done for the refcount field
in task->simdata. Perhaps this refcount can be used for this purpose, but I
don't know enough about the innards of msg/simix to be 100% sure. This will
likely due for now, at least until the next rewrite ;)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6139
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
casanova [Sat, 14 Feb 2009 08:45:22 +0000 (08:45 +0000)]
Added a reference counter to the s_smx_action data structure
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6138
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Feb 2009 11:37:13 +0000 (11:37 +0000)]
I shouldn't do dcommit that fast
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6135
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Feb 2009 11:35:22 +0000 (11:35 +0000)]
actually, I have some more gitignore files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6134
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Feb 2009 11:34:35 +0000 (11:34 +0000)]
yeah, right all of us use git-svn these days. No need to hide it
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6133
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 29 Jan 2009 09:38:21 +0000 (09:38 +0000)]
memleak founds by cppcheck
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6132
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 12 Dec 2008 00:43:24 +0000 (00:43 +0000)]
Introduce log appenders
They allow redirecting the output of the specified categories to a file.
Please check the doc on more info about how to use them, but
--log=xbt.app:file:xbt.log --log=surf.app:file:surf.log
does what you may hope.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6117
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 12 Dec 2008 00:43:20 +0000 (00:43 +0000)]
do not display backtrace capturing stuff in the displayed backtrace (plus, be eclipse friendly)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6116
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 12 Dec 2008 00:43:15 +0000 (00:43 +0000)]
kill an unused variable, fix a segfault after resizing the buffer while varsubsting, and also copy the terminating \0 while resizing the buffer in similar conditions
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6115
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 12 Dec 2008 00:43:11 +0000 (00:43 +0000)]
Add --surf-path so that testing remains possible with vpath
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6114
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 11 Dec 2008 20:12:24 +0000 (20:12 +0000)]
rename build directory into buildtool to avoid name clashes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6113
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 11 Dec 2008 15:00:54 +0000 (15:00 +0000)]
keep documentation in sync with implementation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6112
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 26 Nov 2008 14:03:06 +0000 (14:03 +0000)]
Add useful gnets headers to the dist.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6109
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 26 Nov 2008 14:02:53 +0000 (14:02 +0000)]
Add debug on topology.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6108
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Tue, 25 Nov 2008 16:46:18 +0000 (16:46 +0000)]
Removing libtool file from repository.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6107
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Tue, 25 Nov 2008 16:18:47 +0000 (16:18 +0000)]
Solving libtool conflict.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6106
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Tue, 25 Nov 2008 16:18:44 +0000 (16:18 +0000)]
Removed bogus +1 while printing flow id coping with all experiments done before, don't modify it, work in a new copy.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6105
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
velho [Tue, 25 Nov 2008 16:18:41 +0000 (16:18 +0000)]
Added the weight model improvements to lagrange models.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@6104
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
