Logo AND Algorithmique Numérique Distribuée

Public GIT Repository
simgrid.git
7 years ago[Energy] Fix the power_slope in the energy plugin
Christian Heinrich [Sun, 9 Oct 2016 21:39:30 +0000 (23:39 +0200)]
[Energy] Fix the power_slope in the energy plugin

Since the consumption for one core is now explicitly given,
the slope needs to be computed differently, since we must not add
energy for the first core again.

7 years agofix compilation on windows, bummer
Martin Quinson [Sun, 9 Oct 2016 09:17:17 +0000 (11:17 +0200)]
fix compilation on windows, bummer

7 years agodisable the maestro-set test on windows, nobody will ever use it there
Martin Quinson [Sat, 8 Oct 2016 21:37:40 +0000 (23:37 +0200)]
disable the maestro-set test on windows, nobody will ever use it there

7 years agodrop xbt_os_sem_get_value: unused internally, deprecated on OS X El Capitan
Martin Quinson [Fri, 7 Oct 2016 21:59:00 +0000 (23:59 +0200)]
drop xbt_os_sem_get_value: unused internally, deprecated on OS X El Capitan

7 years agoOS X El Capitan deprecates the sem_init() function
Martin Quinson [Fri, 7 Oct 2016 21:28:01 +0000 (23:28 +0200)]
OS X El Capitan deprecates the sem_init() function

7 years agocosmetics in CMakeFile.txt
Martin Quinson [Fri, 7 Oct 2016 21:20:43 +0000 (23:20 +0200)]
cosmetics in CMakeFile.txt

7 years agosome old linkers (as on Mac) don't like --version
Martin Quinson [Fri, 7 Oct 2016 21:11:47 +0000 (23:11 +0200)]
some old linkers (as on Mac) don't like --version

7 years agoplease sonar by using every parameters (+cosmetics)
Martin Quinson [Fri, 7 Oct 2016 19:37:31 +0000 (21:37 +0200)]
please sonar by using every parameters (+cosmetics)

7 years agodon't test equality to 0.
Martin Quinson [Thu, 6 Oct 2016 23:01:06 +0000 (01:01 +0200)]
don't test equality to 0.

7 years agorevalidate a tesh file after Christian recent changes
Martin Quinson [Thu, 6 Oct 2016 22:51:09 +0000 (00:51 +0200)]
revalidate a tesh file after Christian recent changes

7 years agoUh? I cannot write explicit here??
Martin Quinson [Thu, 6 Oct 2016 21:24:07 +0000 (23:24 +0200)]
Uh? I cannot write explicit here??

7 years agoplease sonar
Martin Quinson [Thu, 6 Oct 2016 20:41:18 +0000 (22:41 +0200)]
please sonar

7 years ago[ENERGY] Introduce idle/1core/maxCore energy consumption
Christian Heinrich [Thu, 6 Oct 2016 09:31:53 +0000 (11:31 +0200)]
[ENERGY] Introduce idle/1core/maxCore energy consumption

Up until now, we only supported idle and max_core consumption.
We then used a linear model to compute the energy that we used.

This is not an accurate model, since the difference between idle and 1 core
and 1 core and 2 cores is not equal. We hence need the first core.

But even that is not 100% accurate; for the moment it will have to suffice, though.

7 years ago[ChangeLog] Correct many -> much
Christian Heinrich [Thu, 8 Sep 2016 12:29:30 +0000 (14:29 +0200)]
[ChangeLog] Correct many -> much

7 years agolet storage use lazy updates, and kill dead code
Martin Quinson [Sun, 2 Oct 2016 19:25:37 +0000 (21:25 +0200)]
let storage use lazy updates, and kill dead code

7 years agorename next_occuring_event_isIdempotent() into nextOccuringEventIsIdempotent() (...
Martin Quinson [Sun, 2 Oct 2016 16:42:34 +0000 (18:42 +0200)]
rename next_occuring_event_isIdempotent() into nextOccuringEventIsIdempotent() (+sort bits around)

7 years agocosmetics: rename object fields
Martin Quinson [Sun, 2 Oct 2016 16:27:23 +0000 (18:27 +0200)]
cosmetics: rename object fields

7 years agokill more dead code
Martin Quinson [Sun, 2 Oct 2016 16:24:46 +0000 (18:24 +0200)]
kill more dead code

7 years agotiny code simplification
Martin Quinson [Sun, 2 Oct 2016 16:22:02 +0000 (18:22 +0200)]
tiny code simplification

7 years agoselectiveUpdate is a bool
Martin Quinson [Sun, 2 Oct 2016 16:17:03 +0000 (18:17 +0200)]
selectiveUpdate is a bool

7 years agocompile one more surf example with cpp
Martin Quinson [Sun, 2 Oct 2016 16:16:40 +0000 (18:16 +0200)]
compile one more surf example with cpp

7 years agorename next_occuring_event() into nextOccuringEvent()
Martin Quinson [Sun, 2 Oct 2016 16:03:29 +0000 (18:03 +0200)]
rename next_occuring_event() into nextOccuringEvent()

7 years agobetter use of inherency around Model::next_occuring_event_full()
Martin Quinson [Sun, 2 Oct 2016 15:59:52 +0000 (17:59 +0200)]
better use of inherency around Model::next_occuring_event_full()

7 years agoadd a solve() fun pointer to maxmin datatype (+minor cleanups)
Martin Quinson [Sun, 2 Oct 2016 15:39:34 +0000 (17:39 +0200)]
add a solve() fun pointer to maxmin datatype (+minor cleanups)

7 years agofurther cleanups to all our next_occuring_event functions
Martin Quinson [Sun, 2 Oct 2016 14:50:45 +0000 (16:50 +0200)]
further cleanups to all our next_occuring_event functions

7 years agomore dupplicated code ritual sacrifice and cleanups
Martin Quinson [Sun, 2 Oct 2016 14:36:56 +0000 (16:36 +0200)]
more dupplicated code ritual sacrifice and cleanups

7 years agodon't make a pure function an object member
Martin Quinson [Sun, 2 Oct 2016 14:32:04 +0000 (16:32 +0200)]
don't make a pure function an object member

7 years agoso much dupplicated code, this is disgusting.
Martin Quinson [Sun, 2 Oct 2016 14:20:01 +0000 (16:20 +0200)]
so much dupplicated code, this is disgusting.

The existing code was dealing specifically with the first matching
action to initialize the minimum. I just update the value
inconditionnaly when it's still at its initial value.

We should use numeric_limits<double>::max() instead of -1 as an
initial value, but we return it when no action can be found, and I
don't feel like fixing the expectations of all calling code here.

7 years agokill (hopefully) dead code
Martin Quinson [Sun, 2 Oct 2016 13:25:43 +0000 (15:25 +0200)]
kill (hopefully) dead code

7 years agopreparatory cleanups in model share computations
Martin Quinson [Sun, 2 Oct 2016 11:31:51 +0000 (13:31 +0200)]
preparatory cleanups in model share computations

7 years agostd::unordered_map are sufficient, no need for boost here
Martin Quinson [Sun, 2 Oct 2016 11:04:39 +0000 (13:04 +0200)]
std::unordered_map are sufficient, no need for boost here

+ cosmetics in Floyd

7 years agonice error message when SD_dotload() is called without graphviz installed
Martin Quinson [Fri, 30 Sep 2016 23:21:38 +0000 (01:21 +0200)]
nice error message when SD_dotload() is called without graphviz installed

7 years agotypo and cosmetics
Martin Quinson [Fri, 30 Sep 2016 09:21:58 +0000 (11:21 +0200)]
typo and cosmetics

7 years agoRevert "killing time"
Martin Quinson [Tue, 27 Sep 2016 20:17:18 +0000 (22:17 +0200)]
Revert "killing time"

Old gcc versions do not allow adding std::string fields to plain old
structures. We could fix the commit, but our time is sparse these days.

This reverts commit 23141ea49d9f51ecda69abe8cd02736b92254726.

7 years agosleep disorder treatment
Martin Quinson [Sun, 25 Sep 2016 23:41:46 +0000 (01:41 +0200)]
sleep disorder treatment

7 years agoReduce the casting madness to please sonar
Martin Quinson [Sun, 25 Sep 2016 23:20:00 +0000 (01:20 +0200)]
Reduce the casting madness to please sonar

7 years agomerge a collapsible if statement to please sonar
Martin Quinson [Sun, 25 Sep 2016 23:17:59 +0000 (01:17 +0200)]
merge a collapsible if statement to please sonar

7 years agoappveyor: only run Java tests
Martin Quinson [Sun, 25 Sep 2016 23:08:30 +0000 (01:08 +0200)]
appveyor: only run Java tests

7 years agoappveyor: cleanups now that ctest works
Martin Quinson [Sun, 25 Sep 2016 22:57:57 +0000 (00:57 +0200)]
appveyor: cleanups now that ctest works

7 years agoinline functions should not be dllexported
Martin Quinson [Sun, 25 Sep 2016 22:53:51 +0000 (00:53 +0200)]
inline functions should not be dllexported

7 years agodo not display a warning when running on AppVeyor -- sorry
Martin Quinson [Sun, 25 Sep 2016 22:43:47 +0000 (00:43 +0200)]
do not display a warning when running on AppVeyor -- sorry

7 years agoDon't use stderr to nicely interleave with outputs on stdout
Martin Quinson [Sun, 25 Sep 2016 22:29:53 +0000 (00:29 +0200)]
Don't use stderr to nicely interleave with outputs on stdout

7 years agoappveyor: cosmetics and remove old cruft
Martin Quinson [Sun, 25 Sep 2016 22:13:12 +0000 (00:13 +0200)]
appveyor: cosmetics and remove old cruft

7 years agoappveyor now accepts to hide its config file as travis does
Martin Quinson [Sun, 25 Sep 2016 22:04:41 +0000 (00:04 +0200)]
appveyor now accepts to hide its config file as travis does

7 years agotesh: fix the regexp to ignore windows+java cruft
Martin Quinson [Sun, 25 Sep 2016 22:04:12 +0000 (00:04 +0200)]
tesh: fix the regexp to ignore windows+java cruft

7 years agoappveyor: Various fixes
Martin Quinson [Sun, 25 Sep 2016 20:27:01 +0000 (22:27 +0200)]
appveyor: Various fixes

- don't install cmake since it's already there on appveyor
- use boost 1.60
- fix a typo when installing perl
- do cache the choco downloads

7 years agocosmetics
Martin Quinson [Thu, 22 Sep 2016 08:35:50 +0000 (10:35 +0200)]
cosmetics

7 years agohide a global data container
Martin Quinson [Thu, 15 Sep 2016 16:24:54 +0000 (18:24 +0200)]
hide a global data container

7 years agokilling time
Frederic Suter [Fri, 16 Sep 2016 17:21:57 +0000 (19:21 +0200)]
killing time

7 years agosimplify and fix cycle detection
Frederic Suter [Fri, 16 Sep 2016 12:03:19 +0000 (14:03 +0200)]
simplify and fix cycle detection
updating tesh as some tasks outside a detected cycle were listed in the output.

7 years agodon't convert host to name to host. Perl epoch is over.
Martin Quinson [Thu, 15 Sep 2016 07:29:42 +0000 (09:29 +0200)]
don't convert host to name to host. Perl epoch is over.

7 years agoanother dynar dies
Martin Quinson [Wed, 14 Sep 2016 19:43:42 +0000 (21:43 +0200)]
another dynar dies

7 years agoashaming spell checking
Martin Quinson [Tue, 13 Sep 2016 19:34:28 +0000 (21:34 +0200)]
ashaming spell checking

7 years agoconvert a dynar into a vector
Martin Quinson [Mon, 12 Sep 2016 20:50:58 +0000 (22:50 +0200)]
convert a dynar into a vector

7 years agoMerge pull request #114 from adfaure/master
Martin Quinson [Mon, 12 Sep 2016 20:41:13 +0000 (22:41 +0200)]
Merge pull request #114 from adfaure/master

fix segmentation fault when process try to unlock a not locked mutex

7 years agofix segmentation fault when process try to unlock a not locked mutex
adfaure [Mon, 12 Sep 2016 09:54:24 +0000 (11:54 +0200)]
fix segmentation fault when process try to unlock a not locked mutex

7 years agokill dead code that annoys sonar
Martin Quinson [Sun, 11 Sep 2016 15:33:40 +0000 (17:33 +0200)]
kill dead code that annoys sonar

- I guess that these names were intended to help debugging
- but there is no way it can be directly used, as we are filling
  strings where surf usually provides resources. So, nothing of the
  logging or whatsoever can use them.
- Sonar don't like them because we have to cast the constness
- so wipe that out and forget about this file for now.

7 years agokeep this readable, even if sonar don't like it this way
Martin Quinson [Sun, 11 Sep 2016 15:03:15 +0000 (17:03 +0200)]
keep this readable, even if sonar don't like it this way

7 years agocosmetics, somehow to please sonar
Martin Quinson [Sun, 11 Sep 2016 15:00:56 +0000 (17:00 +0200)]
cosmetics, somehow to please sonar

7 years agoI should try to compile before comiting. Sorry.
Martin Quinson [Sat, 10 Sep 2016 12:06:08 +0000 (14:06 +0200)]
I should try to compile before comiting. Sorry.

7 years agorestore a warning when doing weird assumptions about the system
Martin Quinson [Fri, 9 Sep 2016 22:58:26 +0000 (00:58 +0200)]
restore a warning when doing weird assumptions about the system

7 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Thu, 8 Sep 2016 15:34:37 +0000 (17:34 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

7 years agomore objects
Frederic Suter [Thu, 8 Sep 2016 15:31:29 +0000 (17:31 +0200)]
more objects

7 years agorequalify tesh
degomme [Thu, 8 Sep 2016 14:28:04 +0000 (16:28 +0200)]
requalify tesh

7 years ago[SIMIX] Indentation to please clang-format ...
Christian Heinrich [Thu, 8 Sep 2016 12:28:04 +0000 (14:28 +0200)]
[SIMIX] Indentation to please clang-format ...

7 years ago[SIMIX] Correct spelling of cfg option
Christian Heinrich [Thu, 8 Sep 2016 12:27:30 +0000 (14:27 +0200)]
[SIMIX] Correct spelling of cfg option

7 years ago[SURF/S4U] Rename Cpu::getCore() to Cpu::getCoreCount()
Christian Heinrich [Thu, 8 Sep 2016 12:26:31 +0000 (14:26 +0200)]
[SURF/S4U] Rename Cpu::getCore() to Cpu::getCoreCount()

7 years agosimplifications
Frederic Suter [Thu, 8 Sep 2016 11:03:03 +0000 (13:03 +0200)]
simplifications

7 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Thu, 8 Sep 2016 10:10:14 +0000 (12:10 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

7 years agofurther objectification of jedule bindings
Frederic Suter [Thu, 8 Sep 2016 08:53:56 +0000 (10:53 +0200)]
further objectification of jedule bindings

7 years agoplug a "leak"
degomme [Thu, 8 Sep 2016 09:16:19 +0000 (11:16 +0200)]
plug a "leak"

7 years agoproperly ignore this line
degomme [Thu, 8 Sep 2016 09:15:36 +0000 (11:15 +0200)]
properly ignore this line

7 years agoplug leak
Frederic Suter [Thu, 8 Sep 2016 08:52:37 +0000 (10:52 +0200)]
plug leak

7 years agoquickly review the code of Fred, as requested. Nice job, pal
Martin Quinson [Wed, 7 Sep 2016 21:22:47 +0000 (23:22 +0200)]
quickly review the code of Fred, as requested. Nice job, pal

7 years agoMerge pull request #113 from adfaure/master
Martin Quinson [Wed, 7 Sep 2016 18:58:04 +0000 (20:58 +0200)]
Merge pull request #113 from adfaure/master

[s4u]fix s4u::Comm::wait_any_for bugs when the timeout is below the c…

7 years agoMerge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Wed, 7 Sep 2016 18:53:57 +0000 (20:53 +0200)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid

7 years ago[s4u]fix s4u::Comm::wait_any_for bugs when the timeout is below the current clock
adfaure [Wed, 7 Sep 2016 15:23:54 +0000 (17:23 +0200)]
[s4u]fix s4u::Comm::wait_any_for bugs when the timeout is below the current clock

7 years agosimplifactions
Frederic Suter [Wed, 7 Sep 2016 12:27:57 +0000 (14:27 +0200)]
simplifactions

7 years agodiff on generated file is likely to fail
Frederic Suter [Wed, 7 Sep 2016 12:27:00 +0000 (14:27 +0200)]
diff on generated file is likely to fail
... as soon as jedule is not built

7 years agoplease sonar
Martin Quinson [Tue, 6 Sep 2016 19:47:04 +0000 (21:47 +0200)]
please sonar

7 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Tue, 6 Sep 2016 17:20:32 +0000 (19:20 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

7 years agoobjectify jedule
Frederic Suter [Tue, 6 Sep 2016 17:02:55 +0000 (19:02 +0200)]
objectify jedule
still some cleanings to do but this commit might not go red

7 years agoadding the expected output of jedule
Frederic Suter [Tue, 6 Sep 2016 09:24:50 +0000 (11:24 +0200)]
adding the expected output of jedule
make a diff with the file produced

7 years agofix test
degomme [Tue, 6 Sep 2016 08:59:51 +0000 (10:59 +0200)]
fix test

7 years agoTry to fix python tesh on Windows...
degomme [Tue, 6 Sep 2016 08:46:13 +0000 (10:46 +0200)]
Try to fix python tesh on Windows...
This adds a dirty bit for smpirun, as python popen struggles with executing scripts, so sh needs to be explicitely added in this case.

7 years agotry to get tesh ignoring the cruft of Jenkins and friends
Martin Quinson [Mon, 5 Sep 2016 16:36:17 +0000 (18:36 +0200)]
try to get tesh ignoring the cruft of Jenkins and friends

7 years agoMerge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Mon, 5 Sep 2016 14:24:02 +0000 (16:24 +0200)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid

7 years agoinitialize the trace flags in S4U too
Martin Quinson [Mon, 5 Sep 2016 14:16:43 +0000 (16:16 +0200)]
initialize the trace flags in S4U too

7 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Sun, 4 Sep 2016 15:19:17 +0000 (17:19 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

7 years agokill (unindent) an unamed block in a function
Martin Quinson [Sun, 4 Sep 2016 14:48:54 +0000 (16:48 +0200)]
kill (unindent) an unamed block in a function

7 years agoplug a memleak
Martin Quinson [Sun, 4 Sep 2016 09:08:18 +0000 (11:08 +0200)]
plug a memleak

7 years agodon't leak test files on disk
Martin Quinson [Sun, 4 Sep 2016 08:59:45 +0000 (10:59 +0200)]
don't leak test files on disk

7 years agotesh: add a fucking \n at the end of file in mkfile command
Martin Quinson [Sat, 3 Sep 2016 21:33:44 +0000 (23:33 +0200)]
tesh: add a fucking \n at the end of file in mkfile command

This missing char got smpi_replay mad (believing that there is only
one trace file for all where there was another trace file on the
second line, that went undetected because not properly terminated).

I lost my evenning because of that stupid little \n. I'm not happy.

7 years agoignore less informative messages
Martin Quinson [Sat, 3 Sep 2016 21:33:09 +0000 (23:33 +0200)]
ignore less informative messages

7 years agocosmetics
Martin Quinson [Sat, 3 Sep 2016 21:28:17 +0000 (23:28 +0200)]
cosmetics

7 years agounlink the file from the subjob
Martin Quinson [Sat, 3 Sep 2016 16:38:27 +0000 (18:38 +0200)]
unlink the file from the subjob

- This test often fails because the main job does find the file to cleanup
- I suspect synchronization issue, so try to remove it from perl to ease things
- Also have perl close the file before removing it, some OS don't like
  multiple access to the same file

7 years agoTravis: Try to force python3 on MacOSX (and disable sudo)
Martin Quinson [Sat, 3 Sep 2016 16:01:10 +0000 (18:01 +0200)]
Travis: Try to force python3 on MacOSX (and disable sudo)

7 years agotesh.py: also work when there is no wrapper :(
Martin Quinson [Sat, 3 Sep 2016 15:29:00 +0000 (17:29 +0200)]
tesh.py: also work when there is no wrapper :(

7 years agotesh.py: if there is a wrapper, increase the timeout value
Martin Quinson [Sat, 3 Sep 2016 15:21:16 +0000 (17:21 +0200)]
tesh.py: if there is a wrapper, increase the timeout value

7 years agotest memcheck builds with valgrind
Martin Quinson [Sat, 3 Sep 2016 14:51:56 +0000 (16:51 +0200)]
test memcheck builds with valgrind

- Add a --wrapper option to valgrind
- Setup this option with TESH_WRAPPER in cmake
- Setup TESH_WRAPPER to the right content when valgrind is detected