Martin Quinson [Tue, 9 Feb 2016 20:38:23 +0000 (21:38 +0100)]
cosmetics
Martin Quinson [Tue, 9 Feb 2016 17:08:44 +0000 (18:08 +0100)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Tue, 9 Feb 2016 17:04:00 +0000 (18:04 +0100)]
also regenerate the MC flex parsers with a recent flexml
Martin Quinson [Tue, 9 Feb 2016 15:20:03 +0000 (16:20 +0100)]
cosmetics: inline a parsing function to unclutter the routing code
degomme [Tue, 9 Feb 2016 14:57:48 +0000 (15:57 +0100)]
Sometimes (ubuntu 15.10 for us) sighup was blocked and made the following line useless.
Make sure we unblock all signals.
Gabriel Corona [Tue, 9 Feb 2016 15:23:23 +0000 (16:23 +0100)]
[smpi] Get rid of "smpi-run -foreground"
Fix https://github.com/mquinson/simgrid/issues/58
Martin Quinson [Tue, 9 Feb 2016 14:17:19 +0000 (15:17 +0100)]
plug another similar memleak. The solution is not nicer
Martin Quinson [Tue, 9 Feb 2016 14:07:23 +0000 (15:07 +0100)]
Plug a memleak -- the ugly way
Gabriel Corona [Mon, 8 Feb 2016 10:22:22 +0000 (11:22 +0100)]
Add .editorconfig file
Gabriel Corona [Tue, 9 Feb 2016 10:58:14 +0000 (11:58 +0100)]
[simix] Use std::function for the maestro callback instead of void(void*)
Gabriel Corona [Tue, 19 Jan 2016 13:49:13 +0000 (14:49 +0100)]
[simix] Attach context
This is expected ot be used in order to use the main caller/thread as
one of the simulated function. This can be used in order to expose an
API where the maestro is hidden and the main thread is the main
process and can create other simulated processes.
Martin Quinson [Tue, 9 Feb 2016 00:29:13 +0000 (01:29 +0100)]
cosmetics in the platform files
Martin Quinson [Tue, 9 Feb 2016 00:25:40 +0000 (01:25 +0100)]
Plug a memleak (free is too robust to my stupidity)
Martin Quinson [Tue, 9 Feb 2016 00:24:15 +0000 (01:24 +0100)]
The memory management of dirname is stupid, so be brutal
http://stackoverflow.com/questions/
26111096/should-i-free-strdup-pointer-after-basename-dirname-in-c
Martin Quinson [Tue, 9 Feb 2016 00:18:28 +0000 (01:18 +0100)]
cosmetics in the ChangeLog
Martin Quinson [Tue, 9 Feb 2016 00:04:50 +0000 (01:04 +0100)]
Commit separately the flex-generated files for DTDv4
They may be rather uncluttered now that flexml sorts its output, they
remain very large.
Martin Quinson [Tue, 9 Feb 2016 00:04:20 +0000 (01:04 +0100)]
Add lua to my eclipse build path
Martin Quinson [Tue, 9 Feb 2016 00:04:03 +0000 (01:04 +0100)]
reflect the DTDv4 in the documentation
Martin Quinson [Mon, 8 Feb 2016 23:51:54 +0000 (00:51 +0100)]
Introduce DTD v4
See ChangeLog
Martin Quinson [Mon, 8 Feb 2016 23:46:33 +0000 (00:46 +0100)]
convert all xml files to the new DTD v4
Martin Quinson [Mon, 8 Feb 2016 21:04:41 +0000 (22:04 +0100)]
make sure that the dtd is at the promised URL
Martin Quinson [Mon, 8 Feb 2016 19:58:23 +0000 (20:58 +0100)]
[appveyor] don't bother installing another Perl, it does not help
Martin Quinson [Mon, 8 Feb 2016 19:17:53 +0000 (20:17 +0100)]
Regenerate all flexml parsers with 1.9.6-4
Now, the tags are sorted alphabetically in the source code, so that
the future diff remain small and readable (hopefully).
Martin Quinson [Mon, 8 Feb 2016 19:16:50 +0000 (20:16 +0100)]
[cmake] fix the rules to fix the output of Flex to please g++
Martin Quinson [Mon, 8 Feb 2016 11:11:39 +0000 (12:11 +0100)]
cosmetics around surf::Resouce::apply_event()
Martin Quinson [Mon, 8 Feb 2016 10:56:04 +0000 (11:56 +0100)]
remove the last remainings of the addTraces() pimple \o/
At least. That was not easy to untangle that.
Martin Quinson [Mon, 8 Feb 2016 10:43:32 +0000 (11:43 +0100)]
make the rest of trace:connect private to the XML parsing
Martin Quinson [Mon, 8 Feb 2016 02:13:15 +0000 (03:13 +0100)]
make trace_connect_list_host_speed private as it should
Martin Quinson [Mon, 8 Feb 2016 01:47:27 +0000 (02:47 +0100)]
[cmake/java] try to reduce the amount of work done over and over
Martin Quinson [Mon, 8 Feb 2016 01:47:00 +0000 (02:47 +0100)]
cosmetics: dont catch an exception you cannot deal
Martin Quinson [Mon, 8 Feb 2016 00:34:55 +0000 (01:34 +0100)]
we won't strip java: we are not robust enough, unfortunately
Martin Quinson [Mon, 8 Feb 2016 00:06:01 +0000 (01:06 +0100)]
kill two tests bypassing the XML parser the ugly way(s)
Martin Quinson [Sun, 7 Feb 2016 23:53:58 +0000 (00:53 +0100)]
VM::next_occuring_event don't need to deal with network
Martin Quinson [Sun, 7 Feb 2016 23:48:34 +0000 (00:48 +0100)]
mark as override a method that does
Martin Quinson [Sun, 7 Feb 2016 23:39:35 +0000 (00:39 +0100)]
make Cpu::speed a surf_metric
Martin Quinson [Sun, 7 Feb 2016 23:28:12 +0000 (00:28 +0100)]
cosmetics
Martin Quinson [Sun, 7 Feb 2016 23:22:28 +0000 (00:22 +0100)]
sanitize surf API: s/updateState/apply_event/
Martin Quinson [Sun, 7 Feb 2016 23:13:42 +0000 (00:13 +0100)]
Protect every lua header with extern 'C'
Martin Quinson [Sun, 7 Feb 2016 23:06:47 +0000 (00:06 +0100)]
sanitize surf API: s/shareResources/next_occuring_event/
This means that we describe the external event (what it does for the
caller) instead of the internal behavior (how it gets to that result)
Martin Quinson [Sun, 7 Feb 2016 22:27:36 +0000 (23:27 +0100)]
sanitize the surf API
updateState() and friends don't need to have the current date passed
as an parameter: they can use surf_get_time() on the rare cases where
they need it.
This makes the surf_solve loop rather ugly as the code is not ready to
jump to the right time to handle events yet. Will soon be fixed.
Martin Quinson [Sun, 7 Feb 2016 21:34:06 +0000 (22:34 +0100)]
Sanitize the handling of the future_event_set
- Make it private to surf
- dont pass it as parameter in public API. That's a singleton after all
Martin Quinson [Sun, 7 Feb 2016 21:33:21 +0000 (22:33 +0100)]
don't load private headers from C tests: they will turn to C++
Martin Quinson [Sun, 7 Feb 2016 21:01:45 +0000 (22:01 +0100)]
revalidate this tesh file, sorry
Martin Quinson [Sun, 7 Feb 2016 20:56:37 +0000 (21:56 +0100)]
sanitize bw and lat in Links: they are now s_surf_metric_t
Martin Quinson [Sun, 7 Feb 2016 20:28:51 +0000 (21:28 +0100)]
first round of cleanups in the NS3 network
Martin Quinson [Sun, 7 Feb 2016 18:17:45 +0000 (19:17 +0100)]
casting an int to void* is not ok
But casting the result of atoi was ok, damn compiler.
Martin Quinson [Sun, 7 Feb 2016 17:40:25 +0000 (18:40 +0100)]
Fix a maybe-uninitialized warning
Martin Quinson [Sun, 7 Feb 2016 17:38:22 +0000 (18:38 +0100)]
Kill some more atof
Martin Quinson [Sun, 7 Feb 2016 17:14:19 +0000 (18:14 +0100)]
kill all atol() from our codebase
Martin Quinson [Sun, 7 Feb 2016 16:51:06 +0000 (17:51 +0100)]
greatly reduce the amount of atoi in our codebase
Some remain and should be hunted down. atol and atof will follow.
Martin Quinson [Sun, 7 Feb 2016 16:15:23 +0000 (17:15 +0100)]
put a deployment file near to its binaries
Martin Quinson [Sun, 7 Feb 2016 15:35:01 +0000 (16:35 +0100)]
New functions: xbt_str_parse_int / xbt_str_parse_double
Martin Quinson [Sun, 7 Feb 2016 14:16:14 +0000 (15:16 +0100)]
fix a potential security hazard
even if a user having a %s in its host name kinda desserves the segfault
Martin Quinson [Sun, 7 Feb 2016 14:09:48 +0000 (15:09 +0100)]
revalidate the NS3 tests
Martin Quinson [Sun, 7 Feb 2016 13:58:27 +0000 (14:58 +0100)]
set the output params even when the input params are invalid
Martin Quinson [Sun, 7 Feb 2016 13:49:01 +0000 (14:49 +0100)]
Nice error message when creating a process on non-existing host
Martin Quinson [Sun, 7 Feb 2016 13:40:15 +0000 (14:40 +0100)]
new function: xbt_dynar_sort_strings(), when the content is char*
And use it when possible
Martin Quinson [Sun, 7 Feb 2016 13:05:29 +0000 (14:05 +0100)]
better name for that struct, and add comments
Martin Quinson [Sun, 7 Feb 2016 12:47:33 +0000 (13:47 +0100)]
rename a struct to obey the C++ One Definition Rule (and please gcc6)
Martin Quinson [Sun, 7 Feb 2016 12:35:28 +0000 (13:35 +0100)]
reindent that one to please gcc6 too
Martin Quinson [Sun, 7 Feb 2016 11:58:03 +0000 (12:58 +0100)]
hunt down other occurences of _Bool
Martin Quinson [Sun, 7 Feb 2016 11:55:11 +0000 (12:55 +0100)]
update my formatting style
Martin Quinson [Sun, 7 Feb 2016 11:49:53 +0000 (12:49 +0100)]
reindent that one to please gcc6 too
Martin Quinson [Sun, 7 Feb 2016 11:24:36 +0000 (12:24 +0100)]
fix indentation to please gcc 6
Martin Quinson [Sun, 7 Feb 2016 11:21:07 +0000 (12:21 +0100)]
clang does not define _Bool in C++, so use bool instead
Martin Quinson [Sun, 7 Feb 2016 11:10:41 +0000 (12:10 +0100)]
compile Lua bindings as C++ code
Martin Quinson [Sun, 7 Feb 2016 10:50:31 +0000 (11:50 +0100)]
compile SMPI as C++ source
Agreed, that's not real C++ code. We would need to objectify
everything, and kill our XBT containers, at the very least.
But this will allow to further objectify the kernel. I'll deal with
the user interfaces later.
Martin Quinson [Sun, 7 Feb 2016 09:31:30 +0000 (10:31 +0100)]
error message improvement
Martin Quinson [Sat, 6 Feb 2016 09:22:20 +0000 (10:22 +0100)]
sed -i -e 's/\t/ /g' [sources] Please people, stop using tabs
Martin Quinson [Sat, 6 Feb 2016 01:02:19 +0000 (02:02 +0100)]
kill that ruby script that nobody will ever use
Martin Quinson [Sat, 6 Feb 2016 00:58:35 +0000 (01:58 +0100)]
clang don't like the register modifer in c++
Martin Quinson [Sat, 6 Feb 2016 00:48:01 +0000 (01:48 +0100)]
convert the last bits of surf to C++
Martin Quinson [Sat, 6 Feb 2016 00:16:10 +0000 (01:16 +0100)]
oops I broke that test, sorry
Martin Quinson [Sat, 6 Feb 2016 00:02:36 +0000 (01:02 +0100)]
Merge pull request #59 from fabienchaix/master
Splitting the LMM constraint element_set into enabled_element_set and disabled_element_set
Martin Quinson [Fri, 5 Feb 2016 23:50:02 +0000 (00:50 +0100)]
that test does not need to be sooo long
Martin Quinson [Fri, 5 Feb 2016 23:48:25 +0000 (00:48 +0100)]
convert Jedule bindings to C++
Martin Quinson [Fri, 5 Feb 2016 23:27:54 +0000 (00:27 +0100)]
kill platform generation from the C sources
This should be reimplemented in Lua anyway. It was not tested, and I
stumbled upon it during another refactoring and got upset. RIP.
Martin Quinson [Fri, 5 Feb 2016 20:55:44 +0000 (21:55 +0100)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Fri, 5 Feb 2016 20:46:18 +0000 (21:46 +0100)]
remove stochastic traces for now.
They were not even tested anyway. They should be reimplemented once
trace_mgr is properly objectified, this way or maybe with lua.
Frederic Suter [Fri, 5 Feb 2016 18:23:21 +0000 (19:23 +0100)]
cleaning some oldies
ok to keep old optorsim files, but not xml that were generated with an
even older version of the provided script.
If one wants to try these GridPP and LCG description circa 2004, just
has to run the script to get the XML.
Martin Quinson [Fri, 5 Feb 2016 16:06:59 +0000 (17:06 +0100)]
Try to make now parameter of Model::shareResources() useless
degomme [Fri, 5 Feb 2016 15:02:50 +0000 (16:02 +0100)]
put back some fix for a warning
degomme [Fri, 5 Feb 2016 14:58:59 +0000 (15:58 +0100)]
(try to) Unify detection of windows. _WIN32 may not be defined for 64 bits systems
degomme [Fri, 5 Feb 2016 14:54:18 +0000 (15:54 +0100)]
This file was not supposed to be still manually privatized
There its _manual version for this case (systems without automatic privatization support).
Frederic Suter [Fri, 5 Feb 2016 14:07:14 +0000 (15:07 +0100)]
put back macros for manual privatization
Frederic Suter [Fri, 5 Feb 2016 13:34:34 +0000 (14:34 +0100)]
attempt to remove oudated coccinelle stuff
Frederic Suter [Fri, 5 Feb 2016 13:00:57 +0000 (14:00 +0100)]
privatization with coccinelle is outdated
Frederic Suter [Fri, 5 Feb 2016 09:28:35 +0000 (10:28 +0100)]
test parsing of parallel task graphs in dot
code was there for a while but not added to the tesh file.
Martin Quinson [Thu, 4 Feb 2016 21:23:07 +0000 (22:23 +0100)]
useless cosmetics
Martin Quinson [Thu, 4 Feb 2016 21:21:03 +0000 (22:21 +0100)]
don't compute the amount of hosts if we don't use it
Martin Quinson [Thu, 4 Feb 2016 20:49:14 +0000 (21:49 +0100)]
cosmetics
Frederic Suter [Thu, 4 Feb 2016 15:54:06 +0000 (16:54 +0100)]
leak--, bad indentation-- too
Frederic Suter [Thu, 4 Feb 2016 14:14:05 +0000 (15:14 +0100)]
leak plug and cosmetics
Frederic Suter [Thu, 4 Feb 2016 10:37:30 +0000 (11:37 +0100)]
coverage madness: use asserts here too
degomme [Thu, 4 Feb 2016 09:22:37 +0000 (10:22 +0100)]
Unify naming of contexts support.
HAVE_RAWCTX -> HAVE_RAW_CONTEXTS
HAVE_BOOST_CONTEXT -> HAVE_BOOST_CONTEXTS
CONTEXT_UCONTEXT -> HAVE_UCONTEXT_CONTEXTS
CONTEXT_THREADS -> HAVE_THREAD_CONTEXTS
degomme [Thu, 4 Feb 2016 09:05:45 +0000 (10:05 +0100)]
As we don't have thread contexts everywhere (windows, for now), don't execute tests using it
Martin Quinson [Wed, 3 Feb 2016 20:41:14 +0000 (21:41 +0100)]
normalize the handling of speed traces in CpuTi
Martin Quinson [Wed, 3 Feb 2016 14:28:17 +0000 (15:28 +0100)]
Further untangle trace:connect. make trace_connect_list_host_avail private
Martin Quinson [Wed, 3 Feb 2016 09:47:12 +0000 (10:47 +0100)]
add the bug number in the test case
Frederic Suter [Wed, 3 Feb 2016 10:29:04 +0000 (11:29 +0100)]
update distrib