mquinson [Mon, 10 Jul 2006 00:11:06 +0000 (00:11 +0000)]
Still more to do
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2515
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 10 Jul 2006 00:02:50 +0000 (00:02 +0000)]
cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2514
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 9 Jul 2006 23:59:56 +0000 (23:59 +0000)]
No need to scary the users for this
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2513
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 9 Jul 2006 23:39:23 +0000 (23:39 +0000)]
Canceled RPC are a pain right now (until we move to a N/1-port model, they are likely to create a dedlock), so give a lot of information about what went wrong
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2512
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 9 Jul 2006 23:37:42 +0000 (23:37 +0000)]
Shorter delays, more logging
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2511
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 9 Jul 2006 23:35:59 +0000 (23:35 +0000)]
shorter delays, more informative logs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2510
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 9 Jul 2006 23:34:37 +0000 (23:34 +0000)]
Plug a memleak
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2509
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 9 Jul 2006 23:28:08 +0000 (23:28 +0000)]
- Do not close sockets closed on other side: user may keep references to them.
Instead, mark them as invalid and ignore them.
- Do raise a timeout when a poll fails (timeout=0, but no message here)
- New function: gras_trp_socketset_dump to display the content of the socket
set.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2508
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 9 Jul 2006 09:11:48 +0000 (09:11 +0000)]
Deal properly with assymetric RPCs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2507
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
dimitrov [Fri, 7 Jul 2006 15:08:18 +0000 (15:08 +0000)]
doing bandwith measurments with MSG
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2506
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Jul 2006 10:10:48 +0000 (10:10 +0000)]
Sends in SG now have a 60s timeout. The right way to go is to have an adaptative timeout for both SG and RL, and also for receptions, but it won't be possible before 3.1. Moreover, the previous situation wasn't satisfactory either since we also had a semantic glitch between RL and SG (thanks to the channel semantic)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2505
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 7 Jul 2006 09:42:32 +0000 (09:42 +0000)]
Take back the cvs ownership of this file
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2504
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 7 Jul 2006 09:35:32 +0000 (09:35 +0000)]
Free the memory
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2503
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 7 Jul 2006 09:35:02 +0000 (09:35 +0000)]
Forgot to initialise a pointer
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2502
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Fri, 7 Jul 2006 09:00:02 +0000 (09:00 +0000)]
Update the "authors and contributors" section.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2501
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 7 Jul 2006 07:55:40 +0000 (07:55 +0000)]
Use action_free to free the actions
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2500
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 21:38:23 +0000 (21:38 +0000)]
also save the cbps in SG, the same way I do in RL since a long time. I love calltree
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2499
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 6 Jul 2006 21:08:48 +0000 (21:08 +0000)]
Remove a huge memory leak
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2498
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 19:17:35 +0000 (19:17 +0000)]
Works both in RL and SG. Those processes are as stupid as lemmings. They we blocking trying to speak to themselves when all deployed on the same host (the case in RL)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2497
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 6 Jul 2006 18:59:32 +0000 (18:59 +0000)]
Add missing SimDag directories
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2496
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 17:40:22 +0000 (17:40 +0000)]
Let's be good boys again and document our changes when they occure
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2495
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 17:38:30 +0000 (17:38 +0000)]
Also display the backtrace on Ctrl-C (use Ctrl-C Ctrl-C to quit)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2494
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 6 Jul 2006 12:43:25 +0000 (12:43 +0000)]
Use _XBT_INLINE instead of inline
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2493
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 09:32:26 +0000 (09:32 +0000)]
also display CPPFLAGS
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2492
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 02:01:26 +0000 (02:01 +0000)]
SimGrid now works perfectly on gcc 4.x
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2491
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 02:00:11 +0000 (02:00 +0000)]
Document my last changes (still missing a lot of things in there)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2490
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 01:48:07 +0000 (01:48 +0000)]
Fix a stack corruption when xbt_dynar_remove_at was used with object larger than 8bytes. It was the case in gras_procdata_exit: we had a by 4 overflow. This leaded to segfault when using optimization on gcc 4.x since %ebx is stored right after our data in that case. YEAH, it works on gcc 4.x now. I would never have found this without Vince, thanks so much
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2489
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 00:46:10 +0000 (00:46 +0000)]
Ah! Some vpath build issues that vince didn't saw (he didn't install flexml ;)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2488
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 00:26:14 +0000 (00:26 +0000)]
Another small fix for Vince, the only guy I know who tries a vpath build on a fresh cvs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2487
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 6 Jul 2006 00:21:58 +0000 (00:21 +0000)]
Little fix for vpath building
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2486
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Wed, 5 Jul 2006 13:55:07 +0000 (13:55 +0000)]
Change some function names
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2485
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Wed, 5 Jul 2006 13:48:42 +0000 (13:48 +0000)]
Add a SimDag example (mixtesim) created from an old SG 2.18 example
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2484
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 5 Jul 2006 07:47:00 +0000 (07:47 +0000)]
Small typo (thanks gcc-4.1 for noticing)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2483
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 5 Jul 2006 07:46:08 +0000 (07:46 +0000)]
Do use the result value of this test (thanks gcc-4.1 for noticing)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2482
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 5 Jul 2006 07:09:45 +0000 (07:09 +0000)]
Investiguated a bit the problems we have with gcc-4.0. Let's document what I've found, as well as the --disable-compiler-optimizations workaround
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2481
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 5 Jul 2006 06:21:55 +0000 (06:21 +0000)]
Make sure some paranoid compilers don't choke on our example (unused variable globalcontext)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2480
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Tue, 4 Jul 2006 16:42:25 +0000 (16:42 +0000)]
Add functions SD_workstation_get_computation_time, SD_workstation_route_get_communication_time, SD_workstation_route_get_latency, SD_workstation_route_get_bandwidth and SD_task_get_execution_time
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2479
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 4 Jul 2006 14:37:00 +0000 (14:37 +0000)]
Adding a MSG_task_put_with_timeout function and giving better return values
for __MSG_task_get_with_time_out_from_host.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2478
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 14:00:25 +0000 (14:00 +0000)]
Automatic update
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2477
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:44:31 +0000 (13:44 +0000)]
Display the name of the canceled RPC
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2476
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:42:42 +0000 (13:42 +0000)]
More debugging (and smaller matrices during the debugging on RL phase)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2475
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:41:45 +0000 (13:41 +0000)]
More debugging and shorter timeouts
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2474
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:40:42 +0000 (13:40 +0000)]
Not testing in RL would be like hiding problems. We wont do it
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2473
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:38:57 +0000 (13:38 +0000)]
Some more debugging messages, and shorter delays 'cause I hate waiting in RL
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2472
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:37:40 +0000 (13:37 +0000)]
At least I found that damn borken char
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2471
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:35:50 +0000 (13:35 +0000)]
When a GRAS process receives the SIGUSR1 signal, it displays its current backtrace. Another neat stuff for debugging
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2470
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:34:17 +0000 (13:34 +0000)]
Signal-related changes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2469
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 4 Jul 2006 13:33:32 +0000 (13:33 +0000)]
Check for signal mechanism
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2468
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Mon, 3 Jul 2006 14:23:54 +0000 (14:23 +0000)]
Add functions SD_link_get_list() and SD_link_get_number() + optimize SD_workstation_get_list()
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2467
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Mon, 3 Jul 2006 08:42:24 +0000 (08:42 +0000)]
Allow how_long to be negative
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2466
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Mon, 3 Jul 2006 07:00:46 +0000 (07:00 +0000)]
Use macros DEBUG and INFO instead of printf
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2465
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 30 Jun 2006 13:53:15 +0000 (13:53 +0000)]
Amount is finally back
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2464
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 30 Jun 2006 11:34:28 +0000 (11:34 +0000)]
Remove attribute amount from structure task
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2463
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 29 Jun 2006 12:15:17 +0000 (12:15 +0000)]
Forgot to remove some debug information
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2462
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 29 Jun 2006 12:12:10 +0000 (12:12 +0000)]
Optimize SD_simulate and fix bugs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2461
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Wed, 28 Jun 2006 13:26:42 +0000 (13:26 +0000)]
Finish SimDag documentation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2460
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Wed, 28 Jun 2006 09:05:15 +0000 (09:05 +0000)]
Add documentation for workstations and fix a few bugs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2459
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 21:40:27 +0000 (21:40 +0000)]
How can I be so bad at boolean expression?
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2458
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 21:25:34 +0000 (21:25 +0000)]
React sainly in gras_msg_wait() when the sender dies in RL between sending the header and sending the body: ignore the message (and hope for a better one afterward)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2457
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 21:22:22 +0000 (21:22 +0000)]
Fix the gras_socket_server_range function: only return the tested socket when it works (in SG, it worked every time, but bad things sometimes happen on RL)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2456
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 21:20:11 +0000 (21:20 +0000)]
The fact that peer closed the socket should be detected right after the select(). If we see it here, there was a problem. So, raise the exception in any case to allow messaging layer to react sainly to it (right now, the receiver would go into an infinite loop when the sender dies between the header send and the body send)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2455
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 19:47:25 +0000 (19:47 +0000)]
Do not go for an endless block wait in handleall (ie, make sure the waiting
time is positive)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2454
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Tue, 27 Jun 2006 14:42:50 +0000 (14:42 +0000)]
Missing cases
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2453
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Tue, 27 Jun 2006 14:31:11 +0000 (14:31 +0000)]
Add a state SD_READY to the tasks to optimize SD_simulate
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2452
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Tue, 27 Jun 2006 14:29:48 +0000 (14:29 +0000)]
Add references
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2451
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 13:24:51 +0000 (13:24 +0000)]
Make sure that doxygen don't complain and do the right thing at the same time (tricky to achieve)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2450
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 13:11:47 +0000 (13:11 +0000)]
Documentation fixups
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2449
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Tue, 27 Jun 2006 12:33:59 +0000 (12:33 +0000)]
Update SimDag documentation: main page and datatypes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2448
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 12:17:18 +0000 (12:17 +0000)]
Calm down, doxygen
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2447
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 12:01:50 +0000 (12:01 +0000)]
Add SimDag to the picture (since my mouse is working right now)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2446
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Tue, 27 Jun 2006 08:23:28 +0000 (08:23 +0000)]
Working on SimDag documentation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2445
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 07:12:12 +0000 (07:12 +0000)]
Add SD to the doc hierarchie
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2444
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 07:11:25 +0000 (07:11 +0000)]
Add SimDag to the documentation hierarchie
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2443
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Tue, 27 Jun 2006 07:10:05 +0000 (07:10 +0000)]
Create a proper channel hierarchie for SD
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2442
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 14:49:17 +0000 (14:49 +0000)]
This is how this test should be started in RL when it will be
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2441
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 14:48:41 +0000 (14:48 +0000)]
Take it easy : don't run this test in RL since it's not ready yet
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2440
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 14:47:59 +0000 (14:47 +0000)]
Register the messages in the master too
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2439
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 14:36:03 +0000 (14:36 +0000)]
Reduce the PROC_MATRIX_SIZE so that the deployment file works; do not display the matrix when it's too large
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2438
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 14:32:06 +0000 (14:32 +0000)]
Do test this in RL too
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2437
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 14:08:25 +0000 (14:08 +0000)]
Adapt the deployment file to the lastest changes in hte source (msgsize added)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2436
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 14:07:55 +0000 (14:07 +0000)]
Useless cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2435
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Mon, 26 Jun 2006 14:04:11 +0000 (14:04 +0000)]
Fix SD_simulate return value
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2434
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 26 Jun 2006 13:57:20 +0000 (13:57 +0000)]
Detect socket closed by peer in the transport layer, so that such things don't polute the messaging layer. This is done by recv() on the socket, which is not allowed on 'file sockets'. So, add a property to the sockets indicating whether it is ok to recv() on the socket or not, and then use it right after the select to detect dead horses
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2433
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Mon, 26 Jun 2006 13:28:33 +0000 (13:28 +0000)]
Fix SimDag compilation warnings.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2432
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 26 Jun 2006 12:08:48 +0000 (12:08 +0000)]
Cosmetics + adding simdag to the main compilation branch.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2431
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 26 Jun 2006 11:47:57 +0000 (11:47 +0000)]
Moving sd_test.c to examples
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2430
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Mon, 26 Jun 2006 08:33:54 +0000 (08:33 +0000)]
Update function SD_simulate.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2429
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Fri, 23 Jun 2006 20:47:47 +0000 (20:47 +0000)]
Do not forget to insert task in the correct swag after having updated
task->state_set.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2428
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 23 Jun 2006 14:59:53 +0000 (14:59 +0000)]
Detect when a task reaches a watch point.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2427
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
amad206 [Fri, 23 Jun 2006 12:58:52 +0000 (12:58 +0000)]
Modify the all2all to support all data Message Size
and make_deployment to add a massage size in deployment files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2426
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 23 Jun 2006 08:47:13 +0000 (08:47 +0000)]
Make private structures SD_link and SD_workstation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2425
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Fri, 23 Jun 2006 08:08:24 +0000 (08:08 +0000)]
Use swags to handle task states and change structure s_SD_task_t.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2424
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 22 Jun 2006 09:41:48 +0000 (09:41 +0000)]
Add a function SD_task_dependency_get_data and improve some code
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2423
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 22 Jun 2006 09:23:06 +0000 (09:23 +0000)]
Replace xbt_dynar_get_ptr by xbt_dynar_get_cpy to simplify code
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2422
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Thu, 22 Jun 2006 09:15:59 +0000 (09:15 +0000)]
Implement functions SD_task_dependency_add and SD_task_dependency_remove.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2421
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 22 Jun 2006 08:09:49 +0000 (08:09 +0000)]
STUPID ME. Don't die awfully when pals shutdown there sockets properly. Shame, shame
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2420
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 22 Jun 2006 08:07:04 +0000 (08:07 +0000)]
More verbose on processes interactions
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2419
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 22 Jun 2006 08:04:03 +0000 (08:04 +0000)]
Those sleeps in the run script are redundent with the ones in the code
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2418
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Wed, 21 Jun 2006 14:26:21 +0000 (14:26 +0000)]
Implement SD_task_schedule, SD_task_unschedule and other SimDag stuff.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2417
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
thiery [Wed, 21 Jun 2006 09:46:18 +0000 (09:46 +0000)]
Implement functions SD_get_amount, SD_get_remaining_amount, SD_watch and SD_unwatch
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@2416
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
