Martin Quinson [Sun, 11 Oct 2015 09:58:08 +0000 (11:58 +0200)]
fix make distcheck, as usual
Martin Quinson [Sat, 10 Oct 2015 19:14:07 +0000 (21:14 +0200)]
[cmake] cleanup: mv AddTests Tests
Martin Quinson [Fri, 9 Oct 2015 19:51:58 +0000 (21:51 +0200)]
[tesh] fix a borken test -- sorry for the noise
Martin Quinson [Fri, 9 Oct 2015 19:25:30 +0000 (21:25 +0200)]
Merge branch 'master' of github.com:mquinson/simgrid
Martin Quinson [Fri, 9 Oct 2015 19:25:23 +0000 (21:25 +0200)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Fri, 9 Oct 2015 19:22:32 +0000 (21:22 +0200)]
Merge pull request #30 from fernandodeperto/bugfix-ptp_messages
Fix duplicate PTP messages when tracing/smpi/internals is enabled
Fernando Mendonca [Fri, 9 Oct 2015 14:06:14 +0000 (16:06 +0200)]
Fix duplicate PTP messages when tracing/smpi/internals is enabled
When the tracing/smpi/internals option is not enabled, some PTP messages
are showed for send and receive operations, both on tracing and
replaying. When the option is enabled, additional messages are
generated, so duplicate messages for send and receive operations are
duplicated.
Initially I thought that the idea was to only show PTP messages when the
tracing/smpi/internals option is enabled. After talking to Arnaud it
seems that is not the case. So I changed the code so both funcionalities
are present: PTP messages for send and receive operations when the
option is disabled and all PTP messages when the option is enabled, as
expected.
Nicolas Bonichon [Fri, 9 Oct 2015 13:44:01 +0000 (15:44 +0200)]
Update limitations of PSG
Gabriel Corona [Fri, 9 Oct 2015 11:35:12 +0000 (13:35 +0200)]
Fix ISP test classification
Gabriel Corona [Fri, 9 Oct 2015 09:32:58 +0000 (11:32 +0200)]
[mc] Move MC_important_snapshot() as a method of ModelChecker
Gabriel Corona [Thu, 8 Oct 2015 13:27:10 +0000 (15:27 +0200)]
[mc] Move some code out of is_visited_state()
Gabriel Corona [Thu, 8 Oct 2015 12:39:43 +0000 (14:39 +0200)]
[mc] Reduce the scope of some variables in mc_visited.cpp
Gabriel Corona [Thu, 8 Oct 2015 12:35:05 +0000 (14:35 +0200)]
[mc] Factorisation in mc_visited.cpp
Gabriel Corona [Tue, 6 Oct 2015 11:54:43 +0000 (13:54 +0200)]
[mc] Create a Member for for structure/class/union member
instead of using Type (which is now only used for types).
Gabriel Corona [Tue, 6 Oct 2015 10:21:13 +0000 (12:21 +0200)]
[mc] Add AddressSpace::process() method
Gabriel Corona [Tue, 6 Oct 2015 09:54:47 +0000 (11:54 +0200)]
[mc] Make Process and Snapshot classes final
Gabriel Corona [Tue, 6 Oct 2015 09:51:26 +0000 (11:51 +0200)]
[mc] Remove Process.is_self()
This was needed for single-process mode.
Gabriel Corona [Tue, 6 Oct 2015 09:39:25 +0000 (11:39 +0200)]
[mc] Move typedef in correct header
Gabriel Corona [Tue, 6 Oct 2015 09:33:21 +0000 (11:33 +0200)]
[mc] Rename mc_process.{h,cpp} into Process.{h,c}pp
Gabriel Corona [Tue, 6 Oct 2015 08:37:22 +0000 (10:37 +0200)]
[mc] Cleanup mc_hash.cpp
* Remove commened-out code;
* OO-ify the hasing method;
* Make it a noop: it was broken since the process separation. I'll add
some basic hashing later on.
Gabriel Corona [Tue, 6 Oct 2015 08:36:06 +0000 (10:36 +0200)]
[mc] Remove commented-out code in mc_diff.cpp
Gabriel Corona [Fri, 2 Oct 2015 11:35:40 +0000 (13:35 +0200)]
[mc] Cleanup #includes
Gabriel Corona [Mon, 5 Oct 2015 13:57:42 +0000 (15:57 +0200)]
[mc] Fix location expression initialization
Gabriel Corona [Mon, 5 Oct 2015 13:10:57 +0000 (15:10 +0200)]
[mc] Fix memory leak in MC_get_memory_regions()
Gabriel Corona [Mon, 5 Oct 2015 11:45:05 +0000 (13:45 +0200)]
[mc] Fix loop bound checking using heaplimit
Christian Heinrich [Fri, 9 Oct 2015 10:21:19 +0000 (12:21 +0200)]
[Doc] Corrected smpi/display_timing paragraph
Christian Heinrich [Fri, 9 Oct 2015 09:57:02 +0000 (11:57 +0200)]
[SMPI] Fixed output alignment for display_timing option
Martin Quinson [Fri, 9 Oct 2015 00:07:12 +0000 (02:07 +0200)]
[tesh] cosmetics on outputs when an error is detected
Martin Quinson [Thu, 8 Oct 2015 23:27:17 +0000 (01:27 +0200)]
[tesh] cleanups in the file parsing
Martin Quinson [Thu, 8 Oct 2015 23:04:09 +0000 (01:04 +0200)]
[tesh] cleanups in the handling of child processes
Martin Quinson [Thu, 8 Oct 2015 05:48:55 +0000 (07:48 +0200)]
[tesh] TODO++
Martin Quinson [Wed, 7 Oct 2015 19:57:51 +0000 (21:57 +0200)]
Appveyor cannot do in-macro macro expension. Bummer.
Martin Quinson [Wed, 7 Oct 2015 19:21:14 +0000 (21:21 +0200)]
[appveyor] try to use tesh without cmake to test the java
Martin Quinson [Wed, 7 Oct 2015 16:12:56 +0000 (18:12 +0200)]
try harder to close stdin on appveyor tests
Martin Quinson [Wed, 7 Oct 2015 13:13:25 +0000 (15:13 +0200)]
[appveyor] close stdin to avoid issues (hopefully)
Martin Quinson [Wed, 7 Oct 2015 09:52:49 +0000 (11:52 +0200)]
[appveyor] try to download the dependency file twice
It sometimes fails for no good reason and trying again seems to be
effective.
(+typo fix)
Martin Quinson [Wed, 7 Oct 2015 09:45:37 +0000 (11:45 +0200)]
Reduce verbosity of the appveyor manual test, and fix ctest invocation
Martin Quinson [Wed, 7 Oct 2015 09:25:20 +0000 (11:25 +0200)]
[appveyor] reactivate our tests now that the jarfile should work
Martin Quinson [Tue, 6 Oct 2015 21:07:26 +0000 (23:07 +0200)]
[java] strip the windows version now that it works
Martin Quinson [Tue, 6 Oct 2015 21:06:26 +0000 (23:06 +0200)]
[appveyor] typo?
Martin Quinson [Tue, 6 Oct 2015 19:43:34 +0000 (21:43 +0200)]
add the ISP/UMPIRE tests to the dist in any case
Martin Quinson [Tue, 6 Oct 2015 19:31:57 +0000 (21:31 +0200)]
[windows] typo breaking the build
Martin Quinson [Tue, 6 Oct 2015 19:17:40 +0000 (21:17 +0200)]
ignore generated tesh files
Martin Quinson [Tue, 6 Oct 2015 19:16:13 +0000 (21:16 +0200)]
set a SimGrid identity to maestro on windows too
Martin Quinson [Mon, 5 Oct 2015 23:48:24 +0000 (01:48 +0200)]
[win] I fail to disable winpthread + I think it's harmless, so re-enable it
Martin Quinson [Mon, 5 Oct 2015 23:44:23 +0000 (01:44 +0200)]
[appveyor] use full path to test folder
Martin Quinson [Mon, 5 Oct 2015 22:49:34 +0000 (00:49 +0200)]
properly disable pthreads on windows (+cleanups)
Martin Quinson [Mon, 5 Oct 2015 22:43:09 +0000 (00:43 +0200)]
[win+java] we need to need CMAKE_SHARED_LIBRARY_LINK_C_FLAGS too
Martin Quinson [Mon, 5 Oct 2015 22:34:13 +0000 (00:34 +0200)]
[windows] pass -m64 when needed
Martin Quinson [Mon, 5 Oct 2015 22:00:26 +0000 (00:00 +0200)]
[appveyor] make the build more verbose to see the used flags
Martin Quinson [Mon, 5 Oct 2015 21:59:46 +0000 (23:59 +0200)]
[win+java] sort out the linker flags that we pass
Martin Quinson [Sun, 4 Oct 2015 23:46:06 +0000 (01:46 +0200)]
[MC] ensure that the main ISP/Umpire tests are activated
All 74 tests of http://formalverification.cs.utah.edu/ISP_Tests/ are
activated. The many other tests of the directory that were not listed
on that page are not activated.
The tesh files are generated on the fly as we don't check the output
anyway. cmake coding is such a pleasure, I could not resist.
We should probably add a way to use tesh without tesh files but only
with command line flags. But not tonight.
/\ /\
/!!\ These tests are currently broken /!!\
/_!!_\ /_!!_\
Martin Quinson [Sun, 4 Oct 2015 21:22:32 +0000 (23:22 +0200)]
don't clutter cmake with tests intended for appveyor only
and increase the verbosity of this test to see what happens
Martin Quinson [Sun, 4 Oct 2015 14:07:02 +0000 (16:07 +0200)]
Try to help the cmake of oldstable to find the java stuff we need
Martin Quinson [Sun, 4 Oct 2015 13:07:13 +0000 (15:07 +0200)]
[win+java] maybe libwinpthread is causing the segfaults on JNI startup that we see
So disable completely the use of pthread on windows, just to see if it
helps. If so, this will still have to be cleaned up properly.
Martin Quinson [Sun, 4 Oct 2015 08:21:39 +0000 (10:21 +0200)]
kill the most outdated parts of our TODO file
Martin Quinson [Sat, 3 Oct 2015 20:58:31 +0000 (22:58 +0200)]
[windows] We don't use winsock2 since the death of GRAS, a few years ago
Martin Quinson [Sat, 3 Oct 2015 20:32:24 +0000 (22:32 +0200)]
[windows] don't mess with -m32 or -m64 gcc flags
Martin Quinson [Sat, 3 Oct 2015 20:23:51 +0000 (22:23 +0200)]
[windows] pexport is a dead project anyway
We should use gendef, the mingw-w64 equivalent, if we want to produce
DLL that can be used in external projects. But we don't want that.
Martin Quinson [Sat, 3 Oct 2015 20:20:00 +0000 (22:20 +0200)]
[java] don't strip libs on windows, to track down the issue
Martin Quinson [Sat, 3 Oct 2015 19:52:52 +0000 (21:52 +0200)]
[cmake] kill some more unused tests
Martin Quinson [Sat, 3 Oct 2015 15:27:04 +0000 (17:27 +0200)]
I meant that Java v1.7 is REQUIRED
Martin Quinson [Sat, 3 Oct 2015 15:06:13 +0000 (17:06 +0200)]
add the NEWS file to the jarfile
Martin Quinson [Sat, 3 Oct 2015 15:05:05 +0000 (17:05 +0200)]
[Java] bump dependency to Java 7
Martin Quinson [Fri, 2 Oct 2015 21:45:21 +0000 (23:45 +0200)]
ignore the lib stripping errors that seem to occure on Mac OSX
Martin Quinson [Fri, 2 Oct 2015 21:24:30 +0000 (23:24 +0200)]
Strip the libs in the jarfile
Martin Quinson [Fri, 2 Oct 2015 21:14:45 +0000 (23:14 +0200)]
[appveyor] don't fail on broken tests (artefact not uploaded on error)
Martin Quinson [Fri, 2 Oct 2015 21:13:42 +0000 (23:13 +0200)]
Don't load DLLs when org.simgrid.NativeLib is loaded: it's called by cmake before the jarfile is ready
Martin Quinson [Fri, 2 Oct 2015 21:00:36 +0000 (23:00 +0200)]
normalize a bit the java test names
Martin Quinson [Fri, 2 Oct 2015 20:57:56 +0000 (22:57 +0200)]
[java] unpack dll/so to a tempDir instead of messing with their name
Martin Quinson [Fri, 2 Oct 2015 14:47:37 +0000 (16:47 +0200)]
[windows] typo in java bare test
Martin Quinson [Fri, 2 Oct 2015 14:25:13 +0000 (16:25 +0200)]
Merge branch 'master' of github.com:mquinson/simgrid
Martin Quinson [Fri, 2 Oct 2015 14:21:55 +0000 (16:21 +0200)]
[appveyor] run the test that I've cooked for windows
Gabriel Corona [Fri, 2 Oct 2015 14:08:10 +0000 (16:08 +0200)]
[mc] Really remove mc_libdw for now
Some bits were not removed properly.
Martin Quinson [Fri, 2 Oct 2015 13:51:42 +0000 (15:51 +0200)]
[windows] try to do a very basic test of java, without tesh
Martin Quinson [Fri, 2 Oct 2015 13:38:09 +0000 (15:38 +0200)]
[win+java] fix the name of libwinpthread-1.dll
Martin Quinson [Fri, 2 Oct 2015 13:25:10 +0000 (15:25 +0200)]
and now, fix the distcheck target
Martin Quinson [Fri, 2 Oct 2015 13:22:29 +0000 (15:22 +0200)]
oops, fix jenkins after moving its conf files
Martin Quinson [Fri, 2 Oct 2015 12:59:25 +0000 (14:59 +0200)]
empty the buildtool directory
Martin Quinson [Fri, 2 Oct 2015 12:40:54 +0000 (14:40 +0200)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Fri, 2 Oct 2015 12:40:45 +0000 (14:40 +0200)]
[jenkins] better location for the jenkins configuration files
Martin Quinson [Fri, 2 Oct 2015 12:34:45 +0000 (14:34 +0200)]
[jenkins] remove the java-created files on startup
Martin Quinson [Fri, 2 Oct 2015 09:09:34 +0000 (11:09 +0200)]
[cmake] remove some unused tests
and various little cleanups
Gabriel Corona [Fri, 2 Oct 2015 10:12:26 +0000 (12:12 +0200)]
Revert "[mc] Move libdw binding code in its own file"
This reverts commit
b81d0b9c9c365d2b195c8a189e2315adbdaa6af6.
Conflicts:
src/mc/mc_dwarf.cpp
src/mc/mc_dwarf.hpp
src/mc/mc_libdw.cpp
src/mc/mc_location.h
Gabriel Corona [Fri, 2 Oct 2015 10:04:10 +0000 (12:04 +0200)]
Revert "[mc] Fix initialisation of dwarf expression"
This reverts commit
4a00aeabd069cdbd1927a3e8a433bb10f49362be.
Gabriel Corona [Fri, 2 Oct 2015 10:04:02 +0000 (12:04 +0200)]
Revert "[mc] Remove remaining bits of libdw outside of mc_libdw.{h,c}pp"
This reverts commit
97cc0fc3eb2ed9773ecd1d612da40bee60727b32.
Gabriel Corona [Mon, 28 Sep 2015 14:19:26 +0000 (16:19 +0200)]
[mc] Bring back soft-dirty tracking from the dead 💀
We want to compare the different snapshoting options on the new
cross-process architecture.
Gabriel Corona [Tue, 29 Sep 2015 11:26:53 +0000 (13:26 +0200)]
[mc] Quick hack to display system memory information (optional)
Gabriel Corona [Mon, 28 Sep 2015 10:20:51 +0000 (12:20 +0200)]
[mc] Bring back KSM (MADV_MERGEABLE) support
For KSM support, we allocate the memory for the snapshot regions
directly with mmap() instead of using malloc(). This way we can safely
mark them as MERGEABLE without impacting the heap at all.
Another solution would be to use posix_memalign() (now that we are
using a full-featured malloc) to allocate page-aligned buffers and
remove MADV_UNMERGEABLE when deallocating.
Gabriel Corona [Thu, 1 Oct 2015 08:45:19 +0000 (10:45 +0200)]
[mc] Support for mmap-able snapthots
This will be used for KSM-friendly snapshots.
Martin Quinson [Fri, 2 Oct 2015 09:10:24 +0000 (11:10 +0200)]
Revert "don't cleanup java directory for now (debugging time)"
Debugging time is over, and it fills partitions up to failures.
This reverts commit
f1e5f8f653f822da96a873fc0a0062fcd4faf3c8.
Martin Quinson [Fri, 2 Oct 2015 08:06:19 +0000 (10:06 +0200)]
[cmake] correctly quote the c:\ that windows requests
Martin Quinson [Fri, 2 Oct 2015 07:27:51 +0000 (09:27 +0200)]
Only try to load winpthread on windows
Previously, it was loaded in any case, and errors were ignored (in an
ineffective way leading to failures on non-windows).
Martin Quinson [Fri, 2 Oct 2015 06:21:34 +0000 (08:21 +0200)]
Revert "[windows] also take the static version of libpthread.dll"
It does not help anyway.
This reverts commit
ae2703f5cf2c00470b492092bb6188c38466b5b5.
Martin Quinson [Fri, 2 Oct 2015 06:21:24 +0000 (08:21 +0200)]
also include winpthread.dll in the jarfile
Martin Quinson [Fri, 2 Oct 2015 05:28:14 +0000 (07:28 +0200)]
[cmake] remove tests for which we don't use the result
Martin Quinson [Fri, 2 Oct 2015 05:23:57 +0000 (07:23 +0200)]
[cmake] cleanup: mv MakeJava.cmake Java.cmake
Martin Quinson [Thu, 1 Oct 2015 20:18:49 +0000 (22:18 +0200)]
[windows] also take the static version of libpthread.dll
GOSH, this is ugly.
Martin Quinson [Thu, 1 Oct 2015 19:44:29 +0000 (21:44 +0200)]
don't cleanup java directory for now (debugging time)
Martin Quinson [Thu, 1 Oct 2015 19:41:32 +0000 (21:41 +0200)]
[windows] request mingw-gcc to put the content of libgcc_s_seh-1.dll statically in our libs