Augustin Degomme [Wed, 18 Jun 2014 15:00:41 +0000 (17:00 +0200)]
try to have all these mpich temporary libs sent to the right folder while building
Arnaud Giersch [Wed, 18 Jun 2014 10:30:19 +0000 (12:30 +0200)]
Update tesh files after commit
2c2afb4d73ea196ea97ca46bc89dd512451222db (untested).
Gabriel Corona [Tue, 17 Jun 2014 14:44:42 +0000 (16:44 +0200)]
[mc] Do not use offsetof() in mc_snapshot_get_heap_end
We don't need this. It's much clearer like this.
Gabriel Corona [Tue, 17 Jun 2014 14:40:25 +0000 (16:40 +0200)]
[mc] Fix mc_snapshot_get_heap_end() for per-page snapshots
Gabriel Corona [Tue, 17 Jun 2014 14:19:28 +0000 (16:19 +0200)]
[mc] Fix mc_translate_address_region() for per-page snapshots
It was completely broken :/
Gabriel Corona [Tue, 17 Jun 2014 13:52:32 +0000 (15:52 +0200)]
[mc] Remove mc_untranslate_address() and mc_member_snapshot_resolve()
Those were working in the snapshot address space so we do not use them
anymore (it is not very useful to work in the snapshot space when the
snapshot space is broken in pages).
We work in the process address space instead (mc_member_resolve).
Gabriel Corona [Tue, 17 Jun 2014 13:45:39 +0000 (15:45 +0200)]
[mc] Use mc_snapshot_read in snapshot_compare (compatibility with per-page snapshots)
Gabriel Corona [Tue, 17 Jun 2014 13:29:24 +0000 (15:29 +0200)]
[mc] mremap does not like MAP_PRIVATE|MAP_ANONYMOUS regions
Use a MAP_PRIVATE one instead.
Gabriel Corona [Tue, 17 Jun 2014 12:42:38 +0000 (14:42 +0200)]
[mc] Add option --cfg=model-check/sparse-checkpoint:yes to enable per page snapshot
Gabriel Corona [Tue, 17 Jun 2014 12:21:50 +0000 (14:21 +0200)]
[mc] Fix bugs in mc_page_snapshot
Gabriel Corona [Tue, 17 Jun 2014 12:06:04 +0000 (14:06 +0200)]
[mc] Use mc_snapshot_read in mc_diff (compatibility with per-page snapshots)
Gabriel Corona [Tue, 17 Jun 2014 10:56:32 +0000 (12:56 +0200)]
[mc] Use mc_snapshot_read in dwarf expression evaluation (compatibility with per-page snapshots)
Gabriel Corona [Tue, 17 Jun 2014 10:54:29 +0000 (12:54 +0200)]
[mc] Fix bug in mc_snapshot_read
It was broken when no snapshot was used.
Gabriel Corona [Tue, 17 Jun 2014 10:12:57 +0000 (12:12 +0200)]
[mc] Remove useless bits
Gabriel Corona [Tue, 17 Jun 2014 08:40:37 +0000 (10:40 +0200)]
Add some informations/help for the user on SIGSEGV
Martin Quinson [Tue, 17 Jun 2014 07:26:43 +0000 (09:26 +0200)]
tiny doc update
Martin Quinson [Tue, 17 Jun 2014 07:23:23 +0000 (09:23 +0200)]
cosmetics on some error messages
Stéphane Castelli [Fri, 13 Jun 2014 16:18:37 +0000 (18:18 +0200)]
This should ease the future implementation of graphes in smpi
Augustin Degomme [Mon, 16 Jun 2014 17:41:12 +0000 (19:41 +0200)]
remove simgrid_documentation target from distcheck
Augustin Degomme [Mon, 16 Jun 2014 17:40:24 +0000 (19:40 +0200)]
add missing file for distcheck
Augustin Degomme [Mon, 16 Jun 2014 14:39:28 +0000 (16:39 +0200)]
ignore sigpipes (as tesh.c used to do)
Gabriel Corona [Mon, 16 Jun 2014 14:10:26 +0000 (16:10 +0200)]
[mc] Remove useless parameters in compare_local_variables
Gabriel Corona [Mon, 16 Jun 2014 13:59:34 +0000 (15:59 +0200)]
[mc] Group MC_ignore_global_variable("mc_diff_info"); its fiends
Gabriel Corona [Mon, 16 Jun 2014 13:29:26 +0000 (15:29 +0200)]
[mc] Remove useless parts in mmalloc_compare_heap, MC_take_snapshot_stacks, snapshot_compare
Gabriel Corona [Mon, 16 Jun 2014 12:50:42 +0000 (14:50 +0200)]
[mc] Use process addresses in mc_compare
This is necessary in order to work work with per-page snapshots.
Gabriel Corona [Fri, 13 Jun 2014 14:03:58 +0000 (16:03 +0200)]
[mc] Partial integration of per-page snapshot address translation (wip)
Gabriel Corona [Mon, 16 Jun 2014 11:00:15 +0000 (13:00 +0200)]
[mc] Simplify s_mc_snapshot_stack
Some fields were not used and have been removed:
* stack_pointer (SP in the snapshot)
* real_address (address of the stack)
The first one does not make sense with page-fragmented snapshot.
Simplify the computation of snapshot->stack_sizes[cursor]. The old
computation did not make sense with page fragmented snapshots.
Stéphane Castelli [Mon, 16 Jun 2014 09:35:38 +0000 (11:35 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Stéphane Castelli [Mon, 16 Jun 2014 09:22:37 +0000 (11:22 +0200)]
Some boilerplate around IB networking
Augustin Degomme [Mon, 16 Jun 2014 09:12:44 +0000 (11:12 +0200)]
change symbol name to match one declared in mpif.h
Stéphane Castelli [Mon, 16 Jun 2014 08:54:56 +0000 (10:54 +0200)]
Uninstall recently added manpages
Fix typos in documentation
degomme [Fri, 13 Jun 2014 13:40:28 +0000 (15:40 +0200)]
Requalify one more test
degomme [Fri, 13 Jun 2014 13:05:32 +0000 (15:05 +0200)]
Revalidate tesh files
Gabriel Corona [Fri, 13 Jun 2014 09:17:52 +0000 (11:17 +0200)]
[mc] Update doc for page_store
Gabriel Corona [Tue, 27 May 2014 10:27:47 +0000 (12:27 +0200)]
[mc] Page-level sparse snapshot: work-in-progress, working page_store
Implement page_store which stores immutables/shared/refcounted copies
of memory pages.
Gabriel Corona [Fri, 13 Jun 2014 12:07:46 +0000 (14:07 +0200)]
[mc] In MC_ignore, keep the zeroed data out of the main snapshot
The previous implementation was zeroing out the snapshot data
* so that it will be automatically ignored by the state comparator;
* but not for the first snapshot in order to be able to backtrack
to the initial state;
In preparation for the per-page snapshoting code, we
1. zero out on the main memory;
2. keep the zero-ed bytes out of the main snapshot regions;
3. restore the zero-ed bytes.
This way the state comparison code and the per-page snapshoting code
do not have to worry about this. Moreover by normalizing, the
"ignored" bytes, we increase the possibility of sharing memory pages
in per-page snapshots.
Martin Quinson [Fri, 13 Jun 2014 00:21:40 +0000 (02:21 +0200)]
example simplification and cosmetics
Arnaud Giersch [Thu, 12 Jun 2014 10:55:10 +0000 (12:55 +0200)]
In tesh.pl, with "output display", print contents of array, and not a reference to it.
Arnaud Giersch [Thu, 12 Jun 2014 09:30:31 +0000 (11:30 +0200)]
Increase size of junk area to 32 KiB in mm_legacy.c.
It is apparently needed when running Fortran tests through tesh.pl.
Arnaud Giersch [Fri, 6 Jun 2014 11:42:36 +0000 (13:42 +0200)]
More updates to smpi* manpages.
Paul Bédaride [Thu, 12 Jun 2014 13:47:07 +0000 (15:47 +0200)]
Rpath active for make install on osX
Paul Bédaride [Thu, 12 Jun 2014 09:28:00 +0000 (11:28 +0200)]
Doc to install libboost on osX
degomme [Wed, 11 Jun 2014 21:10:05 +0000 (23:10 +0200)]
add option to output the right name in smpiff
it used to compile with the temporary name instead, for example with some NAS
Paul Bédaride [Wed, 11 Jun 2014 15:14:36 +0000 (17:14 +0200)]
No rpath in libsimgird-java and libsurf-java
Paul Bédaride [Wed, 11 Jun 2014 13:26:45 +0000 (15:26 +0200)]
Network::shareResourcesFull revival
degomme [Wed, 11 Jun 2014 13:09:17 +0000 (15:09 +0200)]
activate tests using resized and handle some corner cases
scastelli [Wed, 11 Jun 2014 06:56:57 +0000 (08:56 +0200)]
Implementation of MPI_Type_create_resized
degomme [Tue, 10 Jun 2014 23:13:22 +0000 (01:13 +0200)]
add missing prototype
scastelli [Tue, 10 Jun 2014 22:44:15 +0000 (00:44 +0200)]
We should'nt allow the use of uncommited datatypes in communications.
Small fix to follow more closely the specs in mpi_type_free
degomme [Tue, 10 Jun 2014 13:09:35 +0000 (15:09 +0200)]
torus should provide better results if latency is taken into account ..
scastelli [Fri, 6 Jun 2014 07:42:56 +0000 (09:42 +0200)]
Small correction in a manpage
Arnaud Giersch [Thu, 5 Jun 2014 20:43:08 +0000 (22:43 +0200)]
Fix memory leaks.
Arnaud Giersch [Thu, 5 Jun 2014 20:35:42 +0000 (22:35 +0200)]
Update smpi* manpages.
Arnaud Giersch [Thu, 5 Jun 2014 20:35:16 +0000 (22:35 +0200)]
Install additional manpages.
Arnaud Giersch [Thu, 5 Jun 2014 20:31:45 +0000 (22:31 +0200)]
Fix distcheck.
degomme [Thu, 5 Jun 2014 18:20:46 +0000 (20:20 +0200)]
Fortran SMPI is not working for now, so deactivate the build in cmake.
It would need a dirty hack as seen in smpi_main.h, but in F77 and F90, to begin with ...
degomme [Thu, 5 Jun 2014 17:57:50 +0000 (19:57 +0200)]
some cmake calls were forgotten here
Gabriel Corona [Thu, 5 Jun 2014 17:24:38 +0000 (19:24 +0200)]
Remove comment which has not been accurate for a long long time
Arnaud Giersch [Thu, 5 Jun 2014 14:41:51 +0000 (16:41 +0200)]
Remove temporary files.
Arnaud Giersch [Thu, 5 Jun 2014 14:14:08 +0000 (16:14 +0200)]
Change non-portable echo -e for printf.
Arnaud Giersch [Thu, 5 Jun 2014 14:04:51 +0000 (16:04 +0200)]
Typo.
Gabriel Corona [Wed, 4 Jun 2014 22:00:53 +0000 (00:00 +0200)]
Fix broken tesh file (with the Perl tesh)
It was working with the C tesh.
scastelli [Wed, 4 Jun 2014 21:23:44 +0000 (23:23 +0200)]
Some man pages
scastelli [Wed, 4 Jun 2014 18:32:59 +0000 (20:32 +0200)]
Well, that happens sometime (bad merge)
scastelli [Wed, 4 Jun 2014 18:27:06 +0000 (20:27 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Conflicts:
src/smpi/smpiff.in
degomme [Wed, 4 Jun 2014 18:20:45 +0000 (20:20 +0200)]
Fix tesh files
scastelli [Wed, 4 Jun 2014 18:17:18 +0000 (20:17 +0200)]
Some minor documentation fixes
scastelli [Wed, 4 Jun 2014 18:16:19 +0000 (20:16 +0200)]
Update of the smpicc man page
scastelli [Wed, 4 Jun 2014 18:15:46 +0000 (20:15 +0200)]
Add a smpicxx man page
scastelli [Wed, 4 Jun 2014 18:14:44 +0000 (20:14 +0200)]
Add version notes to smpiXXX scripts
degomme [Wed, 4 Jun 2014 18:02:34 +0000 (20:02 +0200)]
Deactivate temporarily one test.
It works with automatic privatization on, but would fail on Windows as we no longer provide a way to privatize global variables.
We can't either just use dynamic memory as in its F90 counterpart ...
degomme [Wed, 4 Jun 2014 17:51:58 +0000 (19:51 +0200)]
Kill f2c. Finally. I hope.
degomme [Wed, 4 Jun 2014 16:19:37 +0000 (18:19 +0200)]
make smpiff behave like smpif90, so we don't use f2c anymore
degomme [Wed, 4 Jun 2014 16:18:55 +0000 (18:18 +0200)]
Privatization : switch data segments when copying inside a single node
Takahiro Hirofuchi [Wed, 4 Jun 2014 17:23:42 +0000 (19:23 +0200)]
fix spell and remove trailing space
scastelli [Wed, 4 Jun 2014 16:54:56 +0000 (18:54 +0200)]
Indentation
Paul Bédaride [Wed, 4 Jun 2014 15:40:08 +0000 (17:40 +0200)]
Add enable_documentation option
Marion Guthmuller [Wed, 4 Jun 2014 15:15:21 +0000 (17:15 +0200)]
model-checker : use new variable _sg_mc_liveness instead of _sg_mc_property_file
Marion Guthmuller [Wed, 4 Jun 2014 15:13:23 +0000 (17:13 +0200)]
model-checker : remove useless debug message and add another one
Marion Guthmuller [Wed, 4 Jun 2014 15:08:47 +0000 (17:08 +0200)]
model-checker : compare the pid of enabled processes before memory introspection
Takahiro Hirofuchi [Wed, 4 Jun 2014 14:22:52 +0000 (16:22 +0200)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Takahiro Hirofuchi [Wed, 4 Jun 2014 14:13:50 +0000 (16:13 +0200)]
add a hopefully simple and clear example of MSG_vm
This is revival of masterslave_virtual_machines. It should be simpler
and clearer than the old one. Because the other examples of MSG_vm are
rather oriented to regression test, it would be nice that we have a
tutorial-oriented example here.
scastelli [Wed, 4 Jun 2014 13:28:54 +0000 (15:28 +0200)]
Is seems as right as before, only shorter
Paul Bédaride [Wed, 4 Jun 2014 13:12:40 +0000 (15:12 +0200)]
Use tesh perl for all platforms
scastelli [Wed, 4 Jun 2014 10:48:43 +0000 (12:48 +0200)]
Some documentation to NetworkModel.
scastelli [Tue, 3 Jun 2014 18:03:46 +0000 (20:03 +0200)]
Minor corrections in doc
Martin Quinson [Wed, 4 Jun 2014 08:37:42 +0000 (10:37 +0200)]
let's build the doc when 'make doc' is called, just in case someone would plan to it
Marion Guthmuller [Wed, 4 Jun 2014 08:35:10 +0000 (10:35 +0200)]
model-checker : fix backtracking with comm determinism verification
degomme [Wed, 4 Jun 2014 08:33:49 +0000 (10:33 +0200)]
As this is done only with java compilation, use the right variable (find_path misses javadoc.exe on win)
Martin Quinson [Tue, 3 Jun 2014 17:33:28 +0000 (19:33 +0200)]
more informative error message
Marion Guthmuller [Tue, 3 Jun 2014 17:00:53 +0000 (19:00 +0200)]
model-checker : comment erroneous ignore (not completely sure ...)
Marion Guthmuller [Tue, 3 Jun 2014 16:14:09 +0000 (18:14 +0200)]
model-checker : use the right heap
Marion Guthmuller [Tue, 3 Jun 2014 16:13:18 +0000 (18:13 +0200)]
model-checker : reword debug message
Marion Guthmuller [Tue, 3 Jun 2014 16:10:33 +0000 (18:10 +0200)]
model-checker : mc_dpor.c -> mc_safety.c
Marion Guthmuller [Tue, 3 Jun 2014 15:45:43 +0000 (17:45 +0200)]
model-checker : remove useless ignore
degomme [Tue, 3 Jun 2014 15:34:51 +0000 (17:34 +0200)]
Requalify teshes, as we really wanted to send 1 byte for this one.
degomme [Tue, 3 Jun 2014 15:24:13 +0000 (17:24 +0200)]
Revert "the users are not the only ones lost with 10e0 :)"
Indeed :p
This reverts commit
cec110d828d0c4d14c6264e79e520a30653c2a4a.
Martin Quinson [Tue, 3 Jun 2014 15:12:36 +0000 (17:12 +0200)]
the users are not the only ones lost with 10e0 :)
Martin Quinson [Tue, 3 Jun 2014 14:46:44 +0000 (16:46 +0200)]
don't fool the users. Don't write 1 as 10e0
Paul Bédaride [Tue, 3 Jun 2014 14:30:36 +0000 (16:30 +0200)]
Fix javadoc into simgrid_full.jar