cherierm [Mon, 7 Apr 2008 13:37:05 +0000 (13:37 +0000)]
test files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5328
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Mon, 7 Apr 2008 13:34:31 +0000 (13:34 +0000)]
add new files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5327
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Mon, 7 Apr 2008 13:26:23 +0000 (13:26 +0000)]
last version of tesh
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5326
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 7 Apr 2008 09:42:19 +0000 (09:42 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5325
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 7 Apr 2008 09:40:08 +0000 (09:40 +0000)]
Look for the GRAS_ROOT in /usr/local if nothing specified
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5324
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 2 Apr 2008 16:32:16 +0000 (16:32 +0000)]
Bug Fix. This function used to return the last task of the list when no matching task was found.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5323
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 19 Mar 2008 03:02:40 +0000 (03:02 +0000)]
Update Makefile.am so that all tesh files are distributed as well as other xml/trace files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5321
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Mon, 17 Mar 2008 16:25:04 +0000 (16:25 +0000)]
add the source file network_constant.c in the visual project of simgrid dll.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5311
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Mon, 17 Mar 2008 16:23:39 +0000 (16:23 +0000)]
rand_r() function is not implemented on Windows
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5310
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 15 Mar 2008 22:41:50 +0000 (22:41 +0000)]
The dirty commit Arnaud refuses to do
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5309
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:42:57 +0000 (20:42 +0000)]
Use a random generator that updates its seed so that our simulation is deterministic.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5308
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:42:38 +0000 (20:42 +0000)]
Handle action whose latency is 0.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5307
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:42:24 +0000 (20:42 +0000)]
Use a random number for latency instead of a constant value.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5306
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:42:10 +0000 (20:42 +0000)]
Getting rid of rates.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5305
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:41:56 +0000 (20:41 +0000)]
Getting rid of lmm_variables and lmm_systems.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5304
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:41:42 +0000 (20:41 +0000)]
Getting rid of network_links.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5303
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:41:28 +0000 (20:41 +0000)]
Getting rid of the routing table.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5302
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:41:11 +0000 (20:41 +0000)]
Don't call lmm_solve and correctly use latency in update_actions_state
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5301
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:40:56 +0000 (20:40 +0000)]
update_actions_state
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5300
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:40:42 +0000 (20:40 +0000)]
Clean the communicate function.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5299
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:40:25 +0000 (20:40 +0000)]
Starting to code a "constant" network model.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5298
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:37:26 +0000 (20:37 +0000)]
Add seed management to have independant streams.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5297
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:37:09 +0000 (20:37 +0000)]
Use doubles for random generators instead of ints and floats.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5296
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:34:40 +0000 (20:34 +0000)]
Move a definition around so that other network models don't need to see CM02 internals...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5295
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:34:24 +0000 (20:34 +0000)]
The interaction between the parser and the surf modules makes more sense now.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5294
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:34:07 +0000 (20:34 +0000)]
cosmetic
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5293
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:33:45 +0000 (20:33 +0000)]
Cleanup in includes.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5292
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sat, 15 Mar 2008 20:33:18 +0000 (20:33 +0000)]
Bug fix. I wonder how this function may have ever worked...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5291
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Thu, 13 Mar 2008 13:46:37 +0000 (13:46 +0000)]
tesh version 2
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5289
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 12 Mar 2008 13:33:47 +0000 (13:33 +0000)]
welcome to modernity
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5288
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 10 Mar 2008 09:18:15 +0000 (09:18 +0000)]
Revalidate output (surf model improvement: latency is now taken into
account in the bandwidth bound).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5287
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 10 Mar 2008 09:17:39 +0000 (09:17 +0000)]
Revalidate output due to:
1) recent parsings modification and maybe xbt_dict modifications:
the first two hosts of the list are not the same ones as they
used to.
2) improvements in the surf model (latency is taken into account in
the bandwidth bound)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5286
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sun, 9 Mar 2008 21:47:18 +0000 (21:47 +0000)]
revalidate output.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5285
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Sun, 9 Mar 2008 21:47:12 +0000 (21:47 +0000)]
Bugfix.
1) Use ptask_L07 as other models do not support parallel tasks anymore.
2) Do not free arguments defining parallel tasks. They are automatically desalocated when the action is destroyed.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5284
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Mar 2008 21:32:52 +0000 (21:32 +0000)]
Fix a god damn race condition: make sure nobody changes the dynar->used value before we get the lock by actually locking before reading it (inline xbt_dynar_insert_at in xbt_dynar_push)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5283
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Mar 2008 21:10:41 +0000 (21:10 +0000)]
Right. time_t is not long int on AIX. ahah. Very funny. Ok, I've found worse than windows to play with
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5282
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Mar 2008 14:31:52 +0000 (14:31 +0000)]
publish a symbol used by a test, for the poor souls using windows
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5281
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Mar 2008 13:52:17 +0000 (13:52 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5280
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Mar 2008 13:42:10 +0000 (13:42 +0000)]
Revalidate the output
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5279
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Mar 2008 13:37:54 +0000 (13:37 +0000)]
avoid redeclaring standard headers to avoid mismatches
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5278
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Mar 2008 13:25:31 +0000 (13:25 +0000)]
Ignore cruft
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5277
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
markls [Fri, 7 Mar 2008 08:32:09 +0000 (08:32 +0000)]
moved XBT_LOG_CONNECT calls into smpi_global so that they will only be called
once per program invocation rather than once per process. Also surrounded by
#ifdef so that smpi can still compile against older versions of simgrid.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5276
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
markls [Thu, 6 Mar 2008 23:28:11 +0000 (23:28 +0000)]
fixed nasty border condition bug.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5275
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 14:09:45 +0000 (14:09 +0000)]
Document last changes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5274
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 13:24:18 +0000 (13:24 +0000)]
Don't be paranoid, people won't connect stuff badly, will they?
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5273
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 13:23:40 +0000 (13:23 +0000)]
Connect correctly the log channels manually for windows. I didn't try to compile last commit, sorry for the noise
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5272
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 12:01:33 +0000 (12:01 +0000)]
Fix detection of old versions
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5271
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 10:26:18 +0000 (10:26 +0000)]
Do connect all log channel manually to parent using XBT_LOG_CONNECT() too, so that it works properly under windows
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5270
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 10:24:56 +0000 (10:24 +0000)]
Make sure that all the categories are connected with XBT_LOG_CONNECT() too, so that it works properly under windows
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5269
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 09:36:13 +0000 (09:36 +0000)]
Cosmetics on the way the documentation of the tree of log channels is generated
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5268
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 09:05:37 +0000 (09:05 +0000)]
Stupid microsoft compiler not understanding the basic bases of C99
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5267
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 08:59:50 +0000 (08:59 +0000)]
Strict ANSI compilers (such as microsoft ones) do not allow non-constant initializers. So, work it arround by adding a XBT_LOG_CONNECT macro to connect the child to the parent afterward
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5266
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 08:51:01 +0000 (08:51 +0000)]
autogenerated
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5265
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 08:04:44 +0000 (08:04 +0000)]
Regenerate all flexml-generated files to fix a FTBFS on buildbots
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5264
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 08:02:38 +0000 (08:02 +0000)]
regenerate all makefiles.in with the lastest autotools
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5263
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Mar 2008 08:00:56 +0000 (08:00 +0000)]
Get the flexml-generated file regenerated on need
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5262
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:49 +0000 (01:51 +0000)]
Revalidate output. There is no task root in this exemple... probably an old cleanup of basic6.c and somebody forgot to update the tesh file accordingly.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5261
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:45 +0000 (01:51 +0000)]
Bug Fix: The symptom is hard to describe. Sometimes some tasks were not scheduled whereas they should have...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5260
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:39 +0000 (01:51 +0000)]
Revalidate output. It did not take into account last improvements to ptask_07 (a better management of 0s and the RTT bound).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5259
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:30 +0000 (01:51 +0000)]
Cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5258
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:23 +0000 (01:51 +0000)]
Revalidate output. It did not take into account last improvements to
ptask_07 (a better management of 0s and the RTT bound).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5257
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:18 +0000 (01:51 +0000)]
Bug fix in the trace management: the power value was not scaled. The
power of a host was between 0 and 1 when using traces instead of being
in Mflop/s.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5256
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:12 +0000 (01:51 +0000)]
Remove useless \n in DEBUG messages.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5255
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:07 +0000 (01:51 +0000)]
Help running properly after last XML parsing/DTD changes.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5254
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:51:01 +0000 (01:51 +0000)]
Remove useless \n in DEBUG messages.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5253
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:50:51 +0000 (01:50 +0000)]
Help running properly after last XML parsing/DTD changes.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5252
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:50:44 +0000 (01:50 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5251
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 5 Mar 2008 01:40:41 +0000 (01:40 +0000)]
Help compiling after last xml parsing/dtd modifications.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5250
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
markls [Thu, 28 Feb 2008 08:18:17 +0000 (08:18 +0000)]
added simple usage message to smpirun and removed nonworking sample
programs.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5249
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 3 Feb 2008 14:16:15 +0000 (14:16 +0000)]
dead link--
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5244
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 3 Feb 2008 14:15:38 +0000 (14:15 +0000)]
Fix the DTD wrt trace:connect. I like it better this way ;)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5243
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 3 Feb 2008 10:54:13 +0000 (10:54 +0000)]
kill properly the surfxml.h file since it got renamed to simgrid_dtd.h a while ago
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5242
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 3 Feb 2008 10:52:34 +0000 (10:52 +0000)]
Reintroduce a more robust hack here
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5241
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 3 Feb 2008 10:40:57 +0000 (10:40 +0000)]
Remove a useless and now harmful rapid hack
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5240
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Fri, 25 Jan 2008 12:57:34 +0000 (12:57 +0000)]
Add comments on how to hack surf to seamlessly handle large
hierarchical platforms.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5234
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 24 Jan 2008 18:47:00 +0000 (18:47 +0000)]
Use unsigned integers for malloc arguments to allow mallocation of more than 2Gb at once
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5233
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 16:37:24 +0000 (16:37 +0000)]
Oops. Forgot a stub
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5230
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 16:18:50 +0000 (16:18 +0000)]
Do not prepare backtraces for display in backtrace_current since this is a very long operation. Only do so when we actually have to display them
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5229
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 15:58:45 +0000 (15:58 +0000)]
And now, fix the linux layer for backtrace extraction by not using values of backtrace (ie, e->bt & e->used) before actually calling the backtrace function
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5228
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 15:46:03 +0000 (15:46 +0000)]
Regenerating autogenerated files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5227
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 15:27:13 +0000 (15:27 +0000)]
avoid naming conflicts
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5226
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 15:09:44 +0000 (15:09 +0000)]
do not call backtrace directly when raising an exception, but our nice portability layer (xbt_backtrace_current)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5225
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 14:49:23 +0000 (14:49 +0000)]
Damnit. Cross-compiling adds the pickyness of gcc to the brutality of visual-C. I'm getting tired
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5224
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 14:43:31 +0000 (14:43 +0000)]
Allow gcc cross-compiling targetting windows, which does not like exported symbols of type []
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5223
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 14:29:01 +0000 (14:29 +0000)]
Damit, leftover of the gcc syntax for inline assembly
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5222
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 14:26:13 +0000 (14:26 +0000)]
give up on cross-compiling backtrace code from gcc: inline assembly code cannot be written in a way that both compilers like (so stick to MSVC syntax, and disable the feature when using gcc either in cross or natively with mingw32)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5221
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 13:56:28 +0000 (13:56 +0000)]
Allow gcc cross-compilation targetting windows, and hopefully kill the dbghelp singleton
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5220
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 13:53:36 +0000 (13:53 +0000)]
Allow gcc cross-compilation targetting windows
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5219
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 13:52:20 +0000 (13:52 +0000)]
Various cleanups+allow cross-compilation with gcc
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5218
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 09:44:42 +0000 (09:44 +0000)]
Other try at killing that singleton
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5217
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 23 Jan 2008 09:38:35 +0000 (09:38 +0000)]
Other try at killing that singleton
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5216
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 16:39:40 +0000 (16:39 +0000)]
try to kill the singleton design pattern
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5215
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 16:16:57 +0000 (16:16 +0000)]
End of last commit about cleaning up windows backtraces (sorry, codding through the build bots)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5214
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 16:13:34 +0000 (16:13 +0000)]
Separate the dummy implementation of the backtraces into its own file (backtrace_dummy), better init/exit mecanism for the backtraces since windows actually has something to do there
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5213
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 15:54:13 +0000 (15:54 +0000)]
Cleanups: kill code dupplication
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5212
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 11:01:21 +0000 (11:01 +0000)]
xbt_new raises an exception on error, no need to check for ENOMEM
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5211
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 10:45:12 +0000 (10:45 +0000)]
Erm. Reindenting did not make the trick on assembly code
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5210
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 10:43:30 +0000 (10:43 +0000)]
Fix another small signess issue
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5209
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 10:41:32 +0000 (10:41 +0000)]
Fix a small signess issue
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5208
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 22 Jan 2008 10:35:23 +0000 (10:35 +0000)]
Reindent (no real change)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5207
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
