Arnaud Giersch [Tue, 29 Nov 2011 13:38:23 +0000 (14:38 +0100)]
Avoid the temporary malloc in xbt_dynar_to_array.
Christophe Thiéry [Tue, 29 Nov 2011 13:39:17 +0000 (14:39 +0100)]
Revert "The latency is constant with the constant network model."
Because actually, it's not.
This reverts commit
123eb94b4e6a8db258c498d38ac1b21656f0f840.
Christophe Thiéry [Tue, 29 Nov 2011 12:57:42 +0000 (13:57 +0100)]
The latency is constant with the constant network model.
Arnaud Giersch [Mon, 28 Nov 2011 22:27:20 +0000 (23:27 +0100)]
Fix uninitialized variable.
Arnaud Giersch [Mon, 28 Nov 2011 17:08:05 +0000 (18:08 +0100)]
Include surf_resource.h, for surf_resource_name().
Arnaud Giersch [Mon, 28 Nov 2011 17:16:14 +0000 (18:16 +0100)]
s/add/free/ in doc.
Arnaud Giersch [Mon, 28 Nov 2011 14:38:21 +0000 (15:38 +0100)]
Don't lie about xbt_xfg_t.
Arnaud Giersch [Tue, 29 Nov 2011 10:36:47 +0000 (11:36 +0100)]
Correclty handle the case p1 < p2 in cmpstringp.
Arnaud Giersch [Fri, 25 Nov 2011 13:58:37 +0000 (14:58 +0100)]
Function does not exist.
Arnaud Giersch [Fri, 25 Nov 2011 17:17:15 +0000 (18:17 +0100)]
Rename dict structs to follow README.coding.
Arnaud Giersch [Mon, 28 Nov 2011 12:21:41 +0000 (13:21 +0100)]
Remove useless internal functions.
Arnaud Giersch [Mon, 28 Nov 2011 12:19:18 +0000 (13:19 +0100)]
Allow NULL for the reset function of mallocators.
Arnaud Giersch [Mon, 28 Nov 2011 12:07:46 +0000 (13:07 +0100)]
Remove unused empty function.
Arnaud Giersch [Thu, 24 Nov 2011 10:57:22 +0000 (11:57 +0100)]
Reorganize structs to save a few bytes.
The general idea is to move the fields, in order to group pointers
apart from integers. Thus, the padding between fields is reduced,
and the size of the struct is reduced too.
It can save up to 32 bytes per node (host and/or process) on 64bits
architectures.
Arnaud Giersch [Wed, 23 Nov 2011 15:25:13 +0000 (16:25 +0100)]
Remove an useless (but optimized!) memcpy.
Arnaud Giersch [Wed, 23 Nov 2011 14:54:14 +0000 (15:54 +0100)]
Avoid to clear mem for the last element.
And save a few calls to memset.
Arnaud Giersch [Wed, 23 Nov 2011 12:55:19 +0000 (13:55 +0100)]
_xbt_clear_mem is used only once. Use memset directly.
Arnaud Giersch [Wed, 23 Nov 2011 12:53:16 +0000 (13:53 +0100)]
Factorize code for xbt_dynar_shrink/_xbt_dynar_expand.
Arnaud Giersch [Wed, 23 Nov 2011 12:51:27 +0000 (13:51 +0100)]
There's no point in clearing mem in _xbt_dynar_expand.
It's already done in _xbt_dynar_set_at_ptr.
Arnaud Giersch [Wed, 23 Nov 2011 12:39:43 +0000 (13:39 +0100)]
Kill dead (commented) code.
Arnaud Giersch [Wed, 23 Nov 2011 12:37:32 +0000 (13:37 +0100)]
Factorize code for xbt_dynar_set/xbt_dynar_set_at_ptr.
Arnaud Giersch [Wed, 23 Nov 2011 12:31:36 +0000 (13:31 +0100)]
Unlock dynar at proper place.
Arnaud Giersch [Wed, 23 Nov 2011 12:22:46 +0000 (13:22 +0100)]
Don't call _xbt_dynar_expand when it's not needed.
Arnaud Giersch [Tue, 22 Nov 2011 12:39:39 +0000 (13:39 +0100)]
Don't clear mem before free.
Use a tool like valgrind to check for memory errors.
Arnaud Giersch [Tue, 22 Nov 2011 14:48:47 +0000 (15:48 +0100)]
Push last_route in global_routing.
Now we are able to free it at exit.
Arnaud Giersch [Mon, 21 Nov 2011 11:23:26 +0000 (12:23 +0100)]
Need to have my own link list.
Else, it is freed while I'm still using it.
Arnaud Giersch [Mon, 21 Nov 2011 11:20:06 +0000 (12:20 +0100)]
Free cost table on finalize.
Arnaud Giersch [Mon, 21 Nov 2011 11:18:59 +0000 (12:18 +0100)]
Strings are already dup'ed by caller.
Arnaud Giersch [Mon, 21 Nov 2011 11:17:02 +0000 (12:17 +0100)]
memcheck_tests.cmake: regenerate
Arnaud Giersch [Mon, 21 Nov 2011 11:15:05 +0000 (12:15 +0100)]
Set path for binary in token ring tesh file.
Arnaud Giersch [Wed, 23 Nov 2011 08:10:08 +0000 (09:10 +0100)]
xbt_dict_new_ext() doesn't exist anymore.
Arnaud Giersch [Wed, 23 Nov 2011 13:13:30 +0000 (14:13 +0100)]
Fix warning about clobbered variable.
Arnaud Giersch [Wed, 23 Nov 2011 15:40:10 +0000 (16:40 +0100)]
Fixup for commit
08ac8cf2900.
Looks like I forgot this changed line.
Martin Quinson [Fri, 18 Nov 2011 16:03:57 +0000 (17:03 +0100)]
make the string large enough to avoid buffer overflows
Martin Quinson [Fri, 18 Nov 2011 16:02:39 +0000 (17:02 +0100)]
get workstation model to pick the right network model, the one given as default one (done the crude way, and LV08 actually breaks some stuff)
Martin Quinson [Fri, 18 Nov 2011 15:35:15 +0000 (16:35 +0100)]
there is no route in vivaldi
Navarrop [Thu, 17 Nov 2011 11:14:28 +0000 (12:14 +0100)]
Add a test for model RuleBased.
Cause now in cluster it is model "cluster".
Christophe Thiéry [Wed, 16 Nov 2011 14:48:06 +0000 (15:48 +0100)]
Lua: change the return values of simgrid.task.send.
Also fix a bug in checking the comm type.
Christophe Thiéry [Wed, 16 Nov 2011 14:47:07 +0000 (15:47 +0100)]
MSG_comm_test and MSG_comm_testany did not update isused
Christophe Thiéry [Wed, 16 Nov 2011 14:41:57 +0000 (15:41 +0100)]
Lua: debug chord again
Arnaud Giersch [Wed, 16 Nov 2011 06:35:16 +0000 (07:35 +0100)]
Regenerated memcheck tests.
Christophe Thiéry [Tue, 15 Nov 2011 22:54:29 +0000 (23:54 +0100)]
Lua: fixing the chord example
Christophe Thiéry [Tue, 15 Nov 2011 22:53:40 +0000 (23:53 +0100)]
Lua: simplify the return values of simgrid.comm.test()
Christophe Thiéry [Tue, 15 Nov 2011 22:01:03 +0000 (23:01 +0100)]
Lua: implement isend, dsend and irecv
Christophe Thiéry [Tue, 15 Nov 2011 22:00:44 +0000 (23:00 +0100)]
Lua: fix Chord example
Christophe Thiéry [Tue, 15 Nov 2011 17:57:13 +0000 (18:57 +0100)]
Nobody uses MSG_get_errno and wants useless calls to SIMIX_process_self
Christophe Thiéry [Tue, 15 Nov 2011 17:22:05 +0000 (18:22 +0100)]
SIMIX: add a redundant parameter to avoid repeated SIMIX_process_self()
Christophe Thiéry [Tue, 15 Nov 2011 15:54:33 +0000 (16:54 +0100)]
Remove a useless function. The default one does the same thing.
Martin Quinson [Tue, 15 Nov 2011 16:15:03 +0000 (17:15 +0100)]
plug some more memleaks. We have no leak on the GOAL example anymore, yuhu
Navarrop [Tue, 15 Nov 2011 15:18:37 +0000 (16:18 +0100)]
Add a platform with a bypass route. Add the test to cmake.
Update flatifier and basic_parsing_test to not put "\t" but spaces.
Martin QUINSON [Tue, 15 Nov 2011 14:54:08 +0000 (15:54 +0100)]
forgot to recreate the option when reviving supernovae
Martin Quinson [Tue, 15 Nov 2011 14:43:55 +0000 (15:43 +0100)]
revert [
a30523e] (remove supernovae) because I need some timings on G5K, and fix what broke in supernovae mode since then
Martin Quinson [Tue, 15 Nov 2011 13:34:45 +0000 (14:34 +0100)]
plug some more memleaks. aint bad anymore
Christophe Thiéry [Tue, 15 Nov 2011 13:28:29 +0000 (14:28 +0100)]
Lua: removing the MSG_process_sleep hack changes the order of output
Christophe Thiéry [Tue, 15 Nov 2011 13:23:43 +0000 (14:23 +0100)]
Lua: copy tasks so that the sender doesn't have to sleep
This will allow to implement asynchronous comms (isend/irecv)
Christophe Thiéry [Tue, 15 Nov 2011 13:18:28 +0000 (14:18 +0100)]
Cosmetics
Martin Quinson [Tue, 15 Nov 2011 12:14:28 +0000 (13:14 +0100)]
plug a bunch of memleaks
Martin Quinson [Tue, 15 Nov 2011 11:04:08 +0000 (12:04 +0100)]
damn it, there is no example testing the bypass mecanism!!
Martin Quinson [Tue, 15 Nov 2011 10:48:41 +0000 (11:48 +0100)]
replace a new/free cycle by a locally stacked object
Martin Quinson [Tue, 15 Nov 2011 10:41:26 +0000 (11:41 +0100)]
work to enforce that link arg is readily usable in r_g_r_a_l
Martin Quinson [Tue, 15 Nov 2011 10:28:19 +0000 (11:28 +0100)]
reduce the dynar creation madness in g_r_a_l in recursive case
Martin Quinson [Tue, 15 Nov 2011 09:43:04 +0000 (10:43 +0100)]
rewrite an error message
Martin Quinson [Tue, 15 Nov 2011 09:36:22 +0000 (10:36 +0100)]
merge get_route and get_latency in routings too
Martin Quinson [Tue, 15 Nov 2011 08:52:40 +0000 (09:52 +0100)]
rename a global to avoid namespace pollution (and other cosmetics)
Martin Quinson [Tue, 15 Nov 2011 08:43:42 +0000 (09:43 +0100)]
kill useless asserts: this is not the way routing logics will take to notice about non-existant routes
Martin Quinson [Tue, 15 Nov 2011 08:40:58 +0000 (09:40 +0100)]
further streamline r_get_route_a_l: now that latency is nicely additive, no need for extra precaution
Martin Quinson [Tue, 15 Nov 2011 08:36:10 +0000 (09:36 +0100)]
simplify _routing_get_route_and_latency wrt latency
- remove duplicated asserts about non negative latency
routing_get_route_and_latency (w/o _) checks it
- give the responsability to initialize it to the caller
- streamline the code
Martin Quinson [Tue, 15 Nov 2011 08:15:09 +0000 (09:15 +0100)]
Change again the calling convention of routing_get_route_and_latency
now, you cannot call it without passing a valid route receiver. It
will be created if pointing to NULL, or simply filled. Actually that
was the de facto calling convention in the code.
This allows to simplify routing_get_route_and_latency by not checking
if the caller is interested in the route at every line.
(plus some useless cosmetics)
Martin Quinson [Tue, 15 Nov 2011 07:45:05 +0000 (08:45 +0100)]
check that AS->get_route don't return NULL on non-existant route, but throw an exception
Martin Quinson [Tue, 15 Nov 2011 01:35:50 +0000 (02:35 +0100)]
kill the now useless 'cleanup' parameter of routing_get_route_and_latency
Instead, if route == NULL we won't bother collecting the route, so
freeing is useless.
If *route == NULL, we create a short-lived dynar (equivalent to the
result of having cleanup=1 previously
else, we fill that route without messing with it
Martin Quinson [Tue, 15 Nov 2011 01:06:53 +0000 (02:06 +0100)]
kill global_routing->get_route_no_cleanup()
use routing_get_route_and_latency(....., 0) instead
Martin Quinson [Tue, 15 Nov 2011 01:01:25 +0000 (02:01 +0100)]
kill routing_get_route(), use routing_get_route_and_latency() instead
Martin Quinson [Tue, 15 Nov 2011 00:35:56 +0000 (01:35 +0100)]
AS->get_route is no longer in charge of creating the route it fills
The whole idea is to reduce the dynar creation, copy, destroy madness
in the router. This step is only the first one on that way.
It may well happen that this change leads to massive memleaks. I'll
check after this later.
Martin Quinson [Mon, 14 Nov 2011 23:31:49 +0000 (00:31 +0100)]
simplification + add comments + reindent of _get_route_and_latency
Martin Quinson [Mon, 14 Nov 2011 23:16:07 +0000 (00:16 +0100)]
useless cosmetics
Martin Quinson [Mon, 14 Nov 2011 16:40:03 +0000 (17:40 +0100)]
pre-parse the vivaldi coordinates
Martin Quinson [Mon, 14 Nov 2011 16:05:55 +0000 (17:05 +0100)]
another empty function murder
Martin Quinson [Mon, 14 Nov 2011 16:01:02 +0000 (17:01 +0100)]
nah, properties in the XML file are not the right place to pass configuration options
Actually, the properties are the right location, but only when
included in a config tag. And in this setting, the right behavior is
already done by the </config> handler, no need to also do it for the
generic <prop> handler. Doing so anyway would prevent to pass
"coordinates" (user-)properties to hosts, links and such.
Martin Quinson [Mon, 14 Nov 2011 15:58:02 +0000 (16:58 +0100)]
don't trust globals to be memset'ed to 0 (and rework a message)
Martin Quinson [Mon, 14 Nov 2011 15:50:13 +0000 (16:50 +0100)]
error message rewrite
Martin Quinson [Mon, 14 Nov 2011 15:48:12 +0000 (16:48 +0100)]
Let's trust the mathematics here: sqrt + fabs + fabs IS positive or nul :)
Martin Quinson [Mon, 14 Nov 2011 15:42:43 +0000 (16:42 +0100)]
tiny cleanups, and add a FIXME
Martin Quinson [Mon, 14 Nov 2011 15:31:41 +0000 (16:31 +0100)]
merge structures route_t and route_extended_t.
All this added complexity to save 2 pointers was a bit too much for
me. Yeah, YMMV, but I prefer it this way.
Martin Quinson [Mon, 14 Nov 2011 14:56:53 +0000 (15:56 +0100)]
tiny function renaming
Martin Quinson [Mon, 14 Nov 2011 19:20:17 +0000 (20:20 +0100)]
fix the breakages that were hidden because some tests were disabled
Martin Quinson [Mon, 14 Nov 2011 16:15:53 +0000 (17:15 +0100)]
pcre is not optional anymore since a while -- reactivate the vivaldi tests
Christophe Thiéry [Mon, 14 Nov 2011 15:48:39 +0000 (16:48 +0100)]
Lua: implementing asynchronous comms (wait, test, isend, dsend, irecv)
Christophe Thiéry [Mon, 14 Nov 2011 14:28:07 +0000 (15:28 +0100)]
Lua: add a function simgrid.process.sleep()
Christophe Thiéry [Mon, 14 Nov 2011 14:10:41 +0000 (15:10 +0100)]
Lua: add a function simgrid.get_clock()
Christophe Thiéry [Mon, 14 Nov 2011 13:51:33 +0000 (14:51 +0100)]
Lua: document the possibility to use numbers for mailboxes
Christophe Thiéry [Mon, 14 Nov 2011 13:38:26 +0000 (14:38 +0100)]
Lua: return errors from execute/send/recv, or nil in case of success
Christophe Thiéry [Mon, 14 Nov 2011 13:03:26 +0000 (14:03 +0100)]
Lua: rename simgrid.task.name() to simgrid.task.get_name()
Christophe Thiéry [Mon, 14 Nov 2011 12:57:18 +0000 (13:57 +0100)]
Lua Chord example (not working yet, the API is still incomplete)
Christophe Thiéry [Mon, 14 Nov 2011 10:22:08 +0000 (11:22 +0100)]
Make MSG_TASK_CANCELLED deprecated, replace it by MSG_TASK_CANCELED
Navarrop [Mon, 14 Nov 2011 16:13:52 +0000 (17:13 +0100)]
Remove ruby example from simgrid.
Arnaud Giersch [Mon, 14 Nov 2011 14:37:41 +0000 (15:37 +0100)]
Use xbt_XXX_is_empty() instead of testing xbt_XXX_length() against 0.
Where XXX is 'dict' or 'dynar'.
(was a request from emptty)
Arnaud Giersch [Mon, 14 Nov 2011 11:01:11 +0000 (12:01 +0100)]
Rewrite xbt libs, using xbt dicts.
And remove lots of code duplication.
Also remove unused functions.
Arnaud Giersch [Mon, 14 Nov 2011 10:23:25 +0000 (11:23 +0100)]
Use xbt_lib_length instead of xbt_lib_size.
Arnaud Giersch [Mon, 14 Nov 2011 10:21:06 +0000 (11:21 +0100)]
Use defined interface for accessing to lib properties.
Arnaud Giersch [Mon, 14 Nov 2011 12:04:16 +0000 (13:04 +0100)]
Remove usage of xbt_dict_size().
xbt_dict_size() can now be safely removed.