mquinson [Wed, 4 Jun 2008 16:15:44 +0000 (16:15 +0000)]
kill dead code
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5531
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Jun 2008 16:11:13 +0000 (16:11 +0000)]
plug a memleak on thread cancelation (gras listener are always canceled)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5530
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Jun 2008 16:09:54 +0000 (16:09 +0000)]
fix expected output
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5529
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Jun 2008 16:05:55 +0000 (16:05 +0000)]
Solve the empty_main bug: simply yield main gras thread right after the listener creation so that it gets a chance to finish its initialization before we cancel it right away if the main is empty
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5528
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Jun 2008 16:04:24 +0000 (16:04 +0000)]
Some more debug output to understand the gras/empty_main bug
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5527
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 4 Jun 2008 15:16:26 +0000 (15:16 +0000)]
Make sure all the source files have an reference of the copyright and of the licence
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5526
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 3 Jun 2008 20:56:06 +0000 (20:56 +0000)]
Inform valgrind about ucontextes [Sékou Diakite]
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5512
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 3 Jun 2008 20:53:20 +0000 (20:53 +0000)]
Well. Some of the tests do not fail
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5511
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 3 Jun 2008 20:36:52 +0000 (20:36 +0000)]
Ok, amok tests do fail. I know that, but I'm sick of delaying the release for that piece of code that nobody uses (beside Lionel of course)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5510
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 3 Jun 2008 16:04:43 +0000 (16:04 +0000)]
Well, actually, yes, top_srcdir is supposed to be a relative path. But since it's not in eclipse, let's go for abs_top_srcdir instead of crude hacks around the directory when calling a command from a subdir
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5509
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 3 Jun 2008 15:56:24 +0000 (15:56 +0000)]
top_srcdir is not supposed to be a relative path
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5508
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 3 Jun 2008 15:53:43 +0000 (15:53 +0000)]
top_srcdir is not supposed to be a relative path
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5507
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 May 2008 10:13:00 +0000 (10:13 +0000)]
dos2unix + indent
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5501
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Thu, 22 May 2008 12:06:20 +0000 (12:06 +0000)]
the makefile used on bob
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5496
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Thu, 22 May 2008 12:04:26 +0000 (12:04 +0000)]
not needed
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5495
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 18:28:29 +0000 (18:28 +0000)]
Display any possible format sequence when the user provided an invalid one
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5493
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 15:48:07 +0000 (15:48 +0000)]
These files are missing from the archive sometimes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5492
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 15:35:48 +0000 (15:35 +0000)]
Because of numerical stability issues (once again), the timings are not exactly the same on 32bits and 64bits for the small_sleep test case
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5491
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 15:29:47 +0000 (15:29 +0000)]
Damn. The order of events changed again in surf. Revalidate once again
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5490
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 15:28:38 +0000 (15:28 +0000)]
Damn. The order of events changed again in surf. Revalidate once again
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5489
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 21 May 2008 15:11:47 +0000 (15:11 +0000)]
Refactoring of code and documentation.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5488
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 21 May 2008 15:11:19 +0000 (15:11 +0000)]
Add a class of exception for file not found and Msg files containing the declarations and implementation of the function connected to MSG.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5487
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 15:00:43 +0000 (15:00 +0000)]
also capture the output of the tests (which all write to stderr)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5486
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 14:58:52 +0000 (14:58 +0000)]
kill all free_string functions declared locally to cleanup dynars afterward, and use xbt_free_ref consitantly for that (plus plug two memleaks on the way in surf parser)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5485
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 May 2008 13:48:56 +0000 (13:48 +0000)]
Speed up update_action_state AND generic_maxmin_share for sleep actions with no maximum duration.
These sleep actions are not put in the running_action swag but in
another private one. Thus, they're somehow hiden from the surf kernel
but it does not need to see them anyway. The cpu module takes care of
them as the only thing that might happen to them (from the surf kernel
perspective) would be to fail, which is now done through the
update_resource function.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5484
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 May 2008 09:53:01 +0000 (09:53 +0000)]
Well the output of printf("%p",NULL) might depend on the system. So let's use a more stable convetion and revalidate the output.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5483
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 09:46:39 +0000 (09:46 +0000)]
Do not use NULL as a name since AIX and linux do not display them the same way (AIX->"";Linux->"(null)")
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5482
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 21 May 2008 09:45:35 +0000 (09:45 +0000)]
complete the auto.tesh file
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5481
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 May 2008 09:44:55 +0000 (09:44 +0000)]
Do not print pointers (%p) when using INFO. This kind of output is system dependant and it breaks the testsuites..
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5480
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 21 May 2008 09:41:15 +0000 (09:41 +0000)]
add catch-wrong-output.tesh to auto.tesh
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5479
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 09:36:45 +0000 (09:36 +0000)]
revalidate output
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5478
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 May 2008 09:32:47 +0000 (09:32 +0000)]
This is the right condition that prevents deadlocks! :)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5477
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 21 May 2008 09:04:48 +0000 (09:04 +0000)]
change the free function by a free_error function to clean all the tesh errors from the dictionary.
change some distraction in tesh files.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5476
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 09:04:36 +0000 (09:04 +0000)]
add one more testing script to the archive
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5475
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 May 2008 09:04:35 +0000 (09:04 +0000)]
Change output of an example and fix the corresponding tesh file.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5474
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 May 2008 08:45:31 +0000 (08:45 +0000)]
Revalidate output!
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5473
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 May 2008 08:45:24 +0000 (08:45 +0000)]
Handle failed tasks in update_resource_state function instead of the update_action_state function.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5472
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 08:42:02 +0000 (08:42 +0000)]
do not try to malloc 0-sized arrays
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5471
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 08:18:31 +0000 (08:18 +0000)]
add every platform example that we have. Moreover, one of them is mandatory for a test
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5470
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 08:08:17 +0000 (08:08 +0000)]
Add the test runner to the archive
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5469
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 May 2008 07:26:52 +0000 (07:26 +0000)]
fix the test integration (they were meant to be run from teshsuite/gras/empty_main instead of teshsuite/)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5468
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 23:22:03 +0000 (23:22 +0000)]
Mmm. Time to bed
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5467
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 23:20:37 +0000 (23:20 +0000)]
Extra care for environ shouldn't be mandatory here, since that's not a dynamic library but a binary. But the linker complains anyway
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5466
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 23:09:16 +0000 (23:09 +0000)]
I changed the way to connect traces afterward about 5 months ago. good time to notice that this example didn't got updated
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5465
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 23:07:40 +0000 (23:07 +0000)]
Second try to let the darwin linker find the environment variables
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5464
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 23:02:52 +0000 (23:02 +0000)]
revalidate the output; more details get now displayed in about the platform, somehow. Not sure of what happened to these outputs. That's what happen when a test remains broken for 5 months, stupid me
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5463
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 22:59:00 +0000 (22:59 +0000)]
Ops, I was saving the data for further use, but I inverted key and value in the dict. This bug is at least 5 months old, and is yet another proof than nobody uses traces (even if you can use them without trigering the bug, with the old interface)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5462
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 22:26:51 +0000 (22:26 +0000)]
Some extra love is needed to use the environ variable in a dynamic library on darwin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5461
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 21:38:42 +0000 (21:38 +0000)]
Revalidate output after last change in surf changing the event order
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5460
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 21:02:01 +0000 (21:02 +0000)]
Revalidate output after last change in surf changing the event order
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5459
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 20:59:19 +0000 (20:59 +0000)]
Revalidate the output after last change in surf messing with event order, and fix a typo
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5458
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 20:54:37 +0000 (20:54 +0000)]
Explain the copious warnings of automake in this directory in the comment
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5457
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 20:50:37 +0000 (20:50 +0000)]
Do append the right buffer to the message when we switched to dynamic logs; be more paranoid on static buffer overflow
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5456
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 20:49:33 +0000 (20:49 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5455
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 20:08:50 +0000 (20:08 +0000)]
Fix make dist
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5454
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 17:31:10 +0000 (17:31 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5453
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 17:29:46 +0000 (17:29 +0000)]
Rework all examples and test cases of the properties
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5452
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 16:09:22 +0000 (16:09 +0000)]
Implement properties on hosts and processes in RL (host is inited to empty, and process is inited to system process environment)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5451
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 15:51:09 +0000 (15:51 +0000)]
add the ability to tesh to mess with the processes' environment
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5450
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 15:46:28 +0000 (15:46 +0000)]
plug a memleak
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5449
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 13:05:11 +0000 (13:05 +0000)]
cleanup around the free-like functions used as dealloc callbacks in dynar and dicts. One function was publicj with its code in xbt_os_time, and the other one was private to xbt_str.c. Now, they are both public and have their code in xbt_main (don't want to do a new .c file for them)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5448
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 12:52:06 +0000 (12:52 +0000)]
Get ride of a supurious compiler warning
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5447
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 20 May 2008 09:59:00 +0000 (09:59 +0000)]
Speed up update_action_state for sleep actions with no maximum duration.
I use the following "now classical" hack. The swag has two parts:
actions that really need being updated are in the head part and the
ones for which no update is needed are in the tail part.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5446
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 20 May 2008 09:58:11 +0000 (09:58 +0000)]
Change the prototype for update_resource_state.
Now that update_resource_state also takes care of failed actions, it
is mendatory to get correct informations about action's finish time.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5445
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 20 May 2008 09:57:29 +0000 (09:57 +0000)]
Try to make actions as failed only when needed.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5444
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 20 May 2008 09:56:47 +0000 (09:56 +0000)]
Incredible bug fix in function lmm_get_var_from_cnst with bad consequences on trace handling...
This function could not work so I had to change the prototype and update parts of the code that relied on it.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5443
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 09:01:08 +0000 (09:01 +0000)]
Rewrite the timeout computation to reduce numerical instabilities issues
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5442
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 20 May 2008 08:05:24 +0000 (08:05 +0000)]
Stupid us (as in Arnaud and me): we should not set sleep(-1) to sleep(MAXMIN_PRECISION). the fix targets durations close to 0, not negative ones. This change broke almost all simulated examples.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5441
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 19 May 2008 22:16:37 +0000 (22:16 +0000)]
Ignore autogenerated cruft
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5440
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 19 May 2008 13:20:05 +0000 (13:20 +0000)]
Test case trying to handle messages for a very short period
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5439
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 19 May 2008 12:42:01 +0000 (12:42 +0000)]
do not wait for too small durations to avoid numerical instabilities
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5438
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 19 May 2008 11:33:35 +0000 (11:33 +0000)]
Also release the mutex when I didn't got what I wanted
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5437
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 16 May 2008 16:08:54 +0000 (16:08 +0000)]
Sometimes, the time to wait is soo small that the increment does not change the global clock (for example, 100000 + 10E-15=100000). Add an epsilon to the wait to avoid the issue. I'm still not completely sure that the fix is 100% effective (because 10E300 + 1 = 10E300), but I don't have any better idea
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5436
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Fri, 16 May 2008 15:06:13 +0000 (15:06 +0000)]
Some new files of the cxx version of the MSG API
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5435
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Fri, 16 May 2008 15:04:54 +0000 (15:04 +0000)]
Code refactoring
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5434
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Fri, 16 May 2008 15:04:21 +0000 (15:04 +0000)]
not needed now.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5433
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Fri, 16 May 2008 09:52:21 +0000 (09:52 +0000)]
Add comment and exception mechanism to the Host class.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5432
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 16 May 2008 09:27:32 +0000 (09:27 +0000)]
documentation improvement
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5431
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Thu, 15 May 2008 16:13:59 +0000 (16:13 +0000)]
Some modifications and some files added.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5430
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 15 May 2008 14:24:16 +0000 (14:24 +0000)]
Remove a crude hack where first call to surf_solve() finalize the initialization, and introduce a surf_presolve() function to do so. That a tiny little cleanup, but well, I had to
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5429
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 14 May 2008 21:35:10 +0000 (21:35 +0000)]
autogen commit fest (some of them being tests related and mandatory for SHA addition)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5428
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 14 May 2008 21:21:10 +0000 (21:21 +0000)]
An implementation of the SHA1 hashing algorithm
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5427
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 14 May 2008 21:15:36 +0000 (21:15 +0000)]
Make an error message a bit more explicit about the most probable error cause
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5426
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 14 May 2008 15:57:57 +0000 (15:57 +0000)]
started of C++ wrappers for Msg
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5425
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 14 May 2008 13:31:55 +0000 (13:31 +0000)]
do not declare these functions both private and public, since it mess this poor windows linker
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5424
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 14 May 2008 13:27:40 +0000 (13:27 +0000)]
If the queue is empty after the timeout, do raise an exception, don't try to get an element anyway
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5423
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 May 2008 15:10:30 +0000 (15:10 +0000)]
do not use static buffer of log event for headers when switching to dynamic logs: when user plays hardcore on seeking scalability limit, this buffer is completely empty
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5420
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 7 May 2008 20:51:00 +0000 (20:51 +0000)]
Export some mailbox related functions
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5413
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 13:27:29 +0000 (13:27 +0000)]
Change of Visual C++ project of SimGrid DLL consecutive to the code refactoring of Martin.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5411
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 12:50:12 +0000 (12:50 +0000)]
correct a thoughtlessness (access violation)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5410
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 12:04:35 +0000 (12:04 +0000)]
add the tesh suites relative with gras, simdag and xbt
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5409
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 12:02:59 +0000 (12:02 +0000)]
some changes for Windows (shell) compatibility and comment the sync call to test (not implemented on Windows)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5408
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 12:00:48 +0000 (12:00 +0000)]
modify the name of the cat windows version (cat instead Cat)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5407
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 11:59:38 +0000 (11:59 +0000)]
replace the long option semantic by read-me
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5406
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 07:54:24 +0000 (07:54 +0000)]
not used files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5405
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 07:52:56 +0000 (07:52 +0000)]
Add the new integrated files version (use xbt data structures instead my own data structures)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5404
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
cherierm [Wed, 7 May 2008 07:49:24 +0000 (07:49 +0000)]
delete all the files to add the new integrated files version
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5403
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 30 Apr 2008 09:54:48 +0000 (09:54 +0000)]
Propagate file renaming to windows
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5362
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 30 Apr 2008 09:47:58 +0000 (09:47 +0000)]
do not to uniformize the type naming scheme today
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5361
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 30 Apr 2008 09:37:25 +0000 (09:37 +0000)]
New test making sure that it's ok to have empty processes (ending before doing anyting) in simix. For now, it's not, but anyway
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@5360
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
