Logo AND Algorithmique Numérique Distribuée

Public GIT Repository
simgrid.git
9 years ago[mc] Remove useless parameters in compare_local_variables
Gabriel Corona [Mon, 16 Jun 2014 14:10:26 +0000 (16:10 +0200)]
[mc] Remove useless parameters in compare_local_variables

9 years ago[mc] Group MC_ignore_global_variable("mc_diff_info"); its fiends
Gabriel Corona [Mon, 16 Jun 2014 13:59:34 +0000 (15:59 +0200)]
[mc] Group MC_ignore_global_variable("mc_diff_info"); its fiends

9 years ago[mc] Remove useless parts in mmalloc_compare_heap, MC_take_snapshot_stacks, snapshot_...
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

9 years ago[mc] Use process addresses in mc_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.

9 years ago[mc] Partial integration of per-page snapshot address translation (wip)
Gabriel Corona [Fri, 13 Jun 2014 14:03:58 +0000 (16:03 +0200)]
[mc] Partial integration of per-page snapshot address translation (wip)

9 years ago[mc] Simplify s_mc_snapshot_stack
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.

9 years ago[mc] Update doc for page_store
Gabriel Corona [Fri, 13 Jun 2014 09:17:52 +0000 (11:17 +0200)]
[mc] Update doc for page_store

9 years ago[mc] Page-level sparse snapshot: work-in-progress, working 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.

9 years ago[mc] In MC_ignore, keep the zeroed data out of the main snapshot
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.

9 years agoRemove comment which has not been accurate for a long long time
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

9 years agomodel-checker : use new variable _sg_mc_liveness instead of _sg_mc_property_file
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

9 years agomodel-checker : remove useless debug message and add another one
Marion Guthmuller [Wed, 4 Jun 2014 15:13:23 +0000 (17:13 +0200)]
model-checker : remove useless debug message and add another one

9 years agomodel-checker : compare the pid of enabled processes before memory introspection
Marion Guthmuller [Wed, 4 Jun 2014 15:08:47 +0000 (17:08 +0200)]
model-checker : compare the pid of enabled processes before memory introspection

9 years agomodel-checker : fix backtracking with comm determinism verification
Marion Guthmuller [Wed, 4 Jun 2014 08:35:10 +0000 (10:35 +0200)]
model-checker : fix backtracking with comm determinism verification

9 years agomodel-checker : comment erroneous ignore (not completely sure ...)
Marion Guthmuller [Tue, 3 Jun 2014 17:00:53 +0000 (19:00 +0200)]
model-checker : comment erroneous ignore (not completely sure ...)

9 years agomodel-checker : use the right heap
Marion Guthmuller [Tue, 3 Jun 2014 16:14:09 +0000 (18:14 +0200)]
model-checker : use the right heap

9 years agomodel-checker : reword debug message
Marion Guthmuller [Tue, 3 Jun 2014 16:13:18 +0000 (18:13 +0200)]
model-checker : reword debug message

9 years agomodel-checker : mc_dpor.c -> mc_safety.c
Marion Guthmuller [Tue, 3 Jun 2014 16:10:33 +0000 (18:10 +0200)]
model-checker : mc_dpor.c -> mc_safety.c

9 years agomodel-checker : remove useless ignore
Marion Guthmuller [Tue, 3 Jun 2014 15:45:43 +0000 (17:45 +0200)]
model-checker : remove useless ignore

9 years agomodel-checker : new tesh examples about communications determinism verification
Marion Guthmuller [Tue, 3 Jun 2014 08:42:48 +0000 (10:42 +0200)]
model-checker : new tesh examples about communications determinism verification

9 years agomodel-checker : fix comm determinism detection mechanisms
Marion Guthmuller [Fri, 30 May 2014 21:51:41 +0000 (23:51 +0200)]
model-checker : fix comm determinism detection mechanisms

9 years agomodel-checker : forget file
Marion Guthmuller [Fri, 30 May 2014 21:30:34 +0000 (23:30 +0200)]
model-checker : forget file

9 years agomodel-checker : cleanups, refactoring and apply indent script
Marion Guthmuller [Fri, 30 May 2014 21:28:56 +0000 (23:28 +0200)]
model-checker : cleanups, refactoring and apply indent script

9 years agomodel-checker : cosmetics in log message
Marion Guthmuller [Fri, 30 May 2014 10:04:10 +0000 (12:04 +0200)]
model-checker : cosmetics in log message

9 years agomodel-checker : improve error message
Marion Guthmuller [Fri, 30 May 2014 07:37:44 +0000 (09:37 +0200)]
model-checker : improve error message

9 years ago[mc] Temporary fix for smpi_process_finalize leading to graph explosion in MC
Gabriel Corona [Mon, 26 May 2014 10:12:39 +0000 (12:12 +0200)]
[mc] Temporary fix for smpi_process_finalize leading to graph explosion in MC

9 years agoMerge branch 'master' into mc
Gabriel Corona [Mon, 26 May 2014 10:05:12 +0000 (12:05 +0200)]
Merge branch 'master' into mc

9 years agoFix a compilation error regarding logging and strings in fat trees
Stéphane Castelli [Mon, 26 May 2014 09:42:17 +0000 (11:42 +0200)]
Fix a compilation error regarding logging and strings in fat trees

9 years agoLots of bugfixes for the fat trees, it should at least not crash when using it
Stéphane Castelli [Mon, 26 May 2014 08:55:34 +0000 (10:55 +0200)]
Lots of bugfixes for the fat trees, it should at least not crash when using it

9 years agosmall cosmetics to AsClusterFatTree::generateDotFile()
Stéphane Castelli [Thu, 22 May 2014 14:53:21 +0000 (16:53 +0200)]
small cosmetics to AsClusterFatTree::generateDotFile()

9 years agoFix some bugs in fat trees
Stéphane Castelli [Thu, 22 May 2014 12:36:10 +0000 (14:36 +0200)]
Fix some bugs in fat trees

9 years agomake the xbt_os_time module public
Martin Quinson [Sat, 24 May 2014 20:38:15 +0000 (22:38 +0200)]
make the xbt_os_time module public

9 years agoupdate the comments of that file
Martin Quinson [Sat, 24 May 2014 20:13:27 +0000 (22:13 +0200)]
update the comments of that file

9 years agolisting new files
suter [Fri, 23 May 2014 18:34:01 +0000 (20:34 +0200)]
listing new files

9 years agostart testing remote I/Os
suter [Fri, 23 May 2014 18:27:50 +0000 (20:27 +0200)]
start testing remote I/Os

9 years agoleak--
suter [Fri, 23 May 2014 18:03:57 +0000 (20:03 +0200)]
leak--

9 years agoallow several hosts to open the same file simultaneously
suter [Fri, 23 May 2014 17:47:49 +0000 (19:47 +0200)]
allow several hosts to open the same file simultaneously

9 years agoThrow an exception only when it's really not implemented
suter [Fri, 23 May 2014 10:44:54 +0000 (12:44 +0200)]
Throw an exception only when it's really not implemented

9 years agoMove unachieved release goals from ChangeLog to TODO file.
Arnaud Giersch [Fri, 23 May 2014 09:46:07 +0000 (11:46 +0200)]
Move unachieved release goals from ChangeLog to TODO file.

9 years agoCosmetics: keep alphabetical order.
Arnaud Giersch [Thu, 22 May 2014 09:40:23 +0000 (11:40 +0200)]
Cosmetics: keep alphabetical order.

9 years agoUpdate doc for surf++
Paul Bédaride [Thu, 22 May 2014 15:23:43 +0000 (17:23 +0200)]
Update doc for surf++

9 years agoRevert "[mmalloc] Use mremap to expand heaps (heap collision prevention)"
Gabriel Corona [Thu, 22 May 2014 12:09:16 +0000 (14:09 +0200)]
Revert "[mmalloc] Use mremap to expand heaps (heap collision prevention)"

This is really how is should be handled. However, this does not work
if someone has split the VMA in parts: this is done when creating
stacks on the heap without MC, mprotect() is used to create a guard
page for the stack.

This reverts commit ba3185d8a4f0a56804d3b1ff24e596096f06a37c.

9 years agoDocumentation for stack size, stack guard and context factory
Gabriel Corona [Thu, 22 May 2014 11:45:55 +0000 (13:45 +0200)]
Documentation for stack size, stack guard and context factory

9 years agoAdd some words about %e fore spaces in logs.
Arnaud Giersch [Wed, 21 May 2014 13:22:16 +0000 (15:22 +0200)]
Add some words about %e fore spaces in logs.

9 years agoWhen the parameter parser is jammed, display the full parsed string
Martin Quinson [Wed, 21 May 2014 08:32:08 +0000 (10:32 +0200)]
When the parameter parser is jammed, display the full parsed string

9 years agoadd some doc about the trap I just felt into
Martin Quinson [Wed, 21 May 2014 08:30:58 +0000 (10:30 +0200)]
add some doc about the trap I just felt into

9 years agoFree previously set process data.
Arnaud Giersch [Tue, 20 May 2014 16:00:35 +0000 (18:00 +0200)]
Free previously set process data.

9 years agoFree smpi_instances.
Arnaud Giersch [Tue, 20 May 2014 15:26:27 +0000 (17:26 +0200)]
Free smpi_instances.

9 years agoFree powerPeak which is unused here.
Arnaud Giersch [Tue, 20 May 2014 15:03:22 +0000 (17:03 +0200)]
Free powerPeak which is unused here.

9 years agoSome tests have different outputs when !HAVE_TRACING. Fix them.
Arnaud Giersch [Tue, 20 May 2014 14:02:10 +0000 (16:02 +0200)]
Some tests have different outputs when !HAVE_TRACING.  Fix them.

9 years agoThis field definitely has sense only when tracing is compiled in.
Arnaud Giersch [Tue, 20 May 2014 13:14:29 +0000 (15:14 +0200)]
This field definitely has sense only when tracing is compiled in.

9 years agoFix race condition.
Arnaud Giersch [Tue, 20 May 2014 13:02:22 +0000 (15:02 +0200)]
Fix race condition.

Avoid to launch and cancel the task at the same time.  It seems to work with
HAVE_TRACING, but fails without.  In fact, the task cancelation happens before
the task starts to execute.  As a workaround, add a small delay before firing
up the cancel.

9 years agoSkip simcall when HAVE_TRACING.
Arnaud Giersch [Tue, 20 May 2014 08:07:14 +0000 (10:07 +0200)]
Skip simcall when HAVE_TRACING.

And try to keep the same visible behavior with and without tracing.

9 years agoCosmetics: reuse already defined variable.
Arnaud Giersch [Tue, 20 May 2014 08:05:14 +0000 (10:05 +0200)]
Cosmetics: reuse already defined variable.

9 years agotask->counter is initialized only when HAVE_TRACING.
Arnaud Giersch [Mon, 19 May 2014 09:29:53 +0000 (11:29 +0200)]
task->counter is initialized only when HAVE_TRACING.

Use task pointer instead to compute the key.

9 years agoDisable tracing tests when !HAVE_TRACING.
Arnaud Giersch [Mon, 19 May 2014 08:17:01 +0000 (10:17 +0200)]
Disable tracing tests when !HAVE_TRACING.

9 years agoUse boost::intrusive in CpuTi
Paul Bédaride [Mon, 19 May 2014 13:37:46 +0000 (15:37 +0200)]
Use boost::intrusive in CpuTi

9 years agoFix a a wrong use of cmake, which makes linking fail when smpi is disabled
Stéphane Castelli [Mon, 19 May 2014 08:51:59 +0000 (10:51 +0200)]
Fix a a wrong use of cmake, which makes linking fail when smpi is disabled

9 years agofurther document the parmap mess. Next step should be refactoring/simplifying
Martin Quinson [Sun, 18 May 2014 18:33:40 +0000 (20:33 +0200)]
further document the parmap mess. Next step should be refactoring/simplifying

9 years agogreatly improve the documentation of the parallel execution in simix
Martin Quinson [Sun, 18 May 2014 17:02:27 +0000 (19:02 +0200)]
greatly improve the documentation of the parallel execution in simix

9 years agoprevious version was initializing retval=0 and then testing retval!=MPI_ERR_REQUEST...
Martin Quinson [Sun, 18 May 2014 17:01:09 +0000 (19:01 +0200)]
previous version was initializing retval=0 and then testing retval!=MPI_ERR_REQUEST. This seems dubious to me

9 years agomove a declaration. Should fix compilation bug on MacOS X .
degomme [Fri, 16 May 2014 16:31:51 +0000 (18:31 +0200)]
move a declaration. Should fix compilation bug on MacOS X .

9 years agoFix compil without tracing, which was broken multiple times ...
degomme [Fri, 16 May 2014 16:20:52 +0000 (18:20 +0200)]
Fix compil without tracing, which was broken multiple times ...
Warning : some tests are broken when tracing is disabled

9 years agoMerge branch 'mc_without_ksm' into mc
Gabriel Corona [Fri, 16 May 2014 13:56:26 +0000 (15:56 +0200)]
Merge branch 'mc_without_ksm' into mc

Conflicts:
src/mc/mc_dpor.c
src/mc/mc_global.c
src/mc/mc_memory.c

9 years agoRevert "Replace xbt_heap by boost::heap::fibonacci_heap"
Paul Bédaride [Fri, 16 May 2014 13:48:07 +0000 (15:48 +0200)]
Revert "Replace xbt_heap by boost::heap::fibonacci_heap"

This reverts commit fb588fbb8c119a8f04e6d2cb15b663257786ba70.

9 years ago[mc] Avoid memsetting twice
Gabriel Corona [Fri, 16 May 2014 13:45:16 +0000 (15:45 +0200)]
[mc] Avoid memsetting twice

9 years agoReplace xbt_heap by boost::heap::fibonacci_heap
Paul Bédaride [Fri, 16 May 2014 12:20:54 +0000 (14:20 +0200)]
Replace xbt_heap by boost::heap::fibonacci_heap

9 years agoRemove unnecessary casts
Paul Bédaride [Mon, 12 May 2014 11:39:35 +0000 (13:39 +0200)]
Remove unnecessary casts

9 years agoUnify models apis
Paul Bédaride [Mon, 12 May 2014 09:00:41 +0000 (11:00 +0200)]
Unify models apis

9 years ago[mc] Use the actual heap size when testing if an address is in the heap (in mm_diff...
Gabriel Corona [Fri, 16 May 2014 09:54:38 +0000 (11:54 +0200)]
[mc] Use the actual heap size when testing if an address is in the heap (in mm_diff, linear_compare_heap)

9 years ago[mc] Move mm_diff in mc/
Gabriel Corona [Fri, 16 May 2014 08:54:08 +0000 (10:54 +0200)]
[mc] Move mm_diff in mc/

mm_diff relies heavily on MC and is not relevant in the xbt/mmalloc
layer anymore. Moreover we probably want to merge it with mc_compare.

9 years ago[mc] Use the actual heap size when testing if an address is in the heap in mm_diff
Gabriel Corona [Fri, 16 May 2014 08:19:59 +0000 (10:19 +0200)]
[mc] Use the actual heap size when testing if an address is in the heap in mm_diff

9 years ago[mc] Use the actual heap size when testing if an address is in the heap in mc_compare
Gabriel Corona [Fri, 16 May 2014 07:53:46 +0000 (09:53 +0200)]
[mc] Use the actual heap size when testing if an address is in the heap in mc_compare

9 years ago[mc] Comments/documentation of memory related code
Gabriel Corona [Fri, 16 May 2014 07:42:55 +0000 (09:42 +0200)]
[mc] Comments/documentation of memory related code

9 years agoFix a bug in SMPI that leads to a segfault when one of the request in MPI_StartAll...
Stéphane Castelli [Wed, 14 May 2014 15:49:44 +0000 (17:49 +0200)]
Fix a bug in SMPI that leads to a segfault when one of the request in MPI_StartAll is MPI_REQUEST_NULL

9 years agoFat trees may actually be used
Stéphane Castelli [Wed, 14 May 2014 14:39:42 +0000 (16:39 +0200)]
Fat trees may actually be used

9 years agoAdd a routing for fat trees
Stéphane Castelli [Tue, 13 May 2014 14:41:57 +0000 (16:41 +0200)]
Add a routing for fat trees

9 years agothird time's the charm ?
Augustin Degomme [Tue, 13 May 2014 18:56:10 +0000 (20:56 +0200)]
third time's the charm ?

9 years agomodel-checker : handle waitany simcall
Marion Guthmuller [Tue, 13 May 2014 16:18:36 +0000 (18:18 +0200)]
model-checker : handle waitany simcall

9 years ago[mc] Rename raw_heap into mc_heap
Gabriel Corona [Tue, 13 May 2014 13:23:13 +0000 (15:23 +0200)]
[mc] Rename raw_heap into mc_heap

9 years agoThe tag cluster accepts the topology fat_tree
Stéphane Castelli [Tue, 13 May 2014 12:56:59 +0000 (14:56 +0200)]
The tag cluster accepts the topology fat_tree

9 years agoAdd a function returning an instance of fat tree
Stéphane Castelli [Tue, 13 May 2014 08:53:37 +0000 (10:53 +0200)]
Add a function returning an instance of fat tree

9 years ago[mmalloc] Use mremap to expand heaps (heap collision prevention)
Gabriel Corona [Fri, 25 Apr 2014 08:05:10 +0000 (10:05 +0200)]
[mmalloc] Use mremap to expand heaps (heap collision prevention)

mremap() will fail instead of overwriting an existing memory mapping :
the first heap could silently overflow on the second one.

9 years agomerge Fred and Henri description of the same stuff
suter [Tue, 13 May 2014 07:24:58 +0000 (09:24 +0200)]
merge Fred and Henri description of the same stuff

9 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Henri Casanova [Mon, 12 May 2014 19:30:55 +0000 (09:30 -1000)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

9 years agoAdded documentation of the --cfg="smpi/simulation_computation" feature of mpirun
Henri Casanova [Mon, 12 May 2014 19:30:23 +0000 (09:30 -1000)]
Added documentation of the  --cfg="smpi/simulation_computation" feature of mpirun

9 years agoupdating ignore patterns-list should be complete now
Luka Stanisic [Mon, 12 May 2014 16:43:51 +0000 (18:43 +0200)]
updating ignore patterns-list should be complete now

9 years agodocument the smpi/simulation_computation option and list it in the
suter [Mon, 12 May 2014 12:44:51 +0000 (14:44 +0200)]
document the smpi/simulation_computation option and list it in the
index (with the one on privatization too)

9 years agomodel-checker : complete comm pattern only after a comm wait
Marion Guthmuller [Mon, 12 May 2014 11:53:59 +0000 (13:53 +0200)]
model-checker : complete comm pattern only after a comm wait

9 years agoSet return value.
Arnaud Giersch [Fri, 9 May 2014 13:01:42 +0000 (15:01 +0200)]
Set return value.

9 years agoDon't try to kill non-existing process (*process == NULL).
Arnaud Giersch [Fri, 9 May 2014 12:15:22 +0000 (14:15 +0200)]
Don't try to kill non-existing process (*process == NULL).

9 years agoFix dead assignments.
Arnaud Giersch [Fri, 9 May 2014 12:09:41 +0000 (14:09 +0200)]
Fix dead assignments.

That's when a value stored in a variable is never used.

9 years agoAdd check for error.
Arnaud Giersch [Fri, 9 May 2014 11:11:38 +0000 (13:11 +0200)]
Add check for error.

9 years agoFix warning about duplicate typedef.
Arnaud Giersch [Fri, 9 May 2014 11:10:53 +0000 (13:10 +0200)]
Fix warning about duplicate typedef.

9 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Stéphane Castelli [Mon, 12 May 2014 07:07:35 +0000 (09:07 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

9 years agoFix use of wrong variable
Stéphane Castelli [Mon, 12 May 2014 07:06:55 +0000 (09:06 +0200)]
Fix use of wrong variable

9 years agoAdded a --cfg="smpi/simulation_computation" to smpirun. The default
Henri Casanova [Fri, 9 May 2014 01:17:37 +0000 (15:17 -1000)]
Added a --cfg="smpi/simulation_computation" to smpirun. The default
value is "yes", meaning that SMPI will benchmark computational parts
of the simulation application and simulate them using smpi_execute(), as
it's always been done in the past.

Disabling the simulation of computation is useful when smpirunning an "application"
that is in fact doing a "live replay" of another MPI app (e.g., ScalaTrace's replay tool, various on-line simulators that run an app at scale). In this case the computation due to the replay logic
should be be simulated by SMPI.

For the time being, for smpirunning such an "application", the approach is:
1. use --cfg="smpi/simulation_computation:no"
2. add calls to smpi_execute() in the "application"

9 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Henri Casanova [Thu, 8 May 2014 20:12:11 +0000 (10:12 -1000)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

9 years agoRewrote smpi_sleep() and smpi_usleep() so that:
Henri Casanova [Thu, 8 May 2014 20:10:44 +0000 (10:10 -1000)]
Rewrote smpi_sleep() and smpi_usleep() so that:
1. the code is not duplicated but shared in a static private function
2. sleeping is done via an actual simcall_process_sleep rather than a bogus computation

Augmented SMPI tracing to allow tracing of sleep events

9 years agoforgot a comma ...
suter [Thu, 8 May 2014 20:10:02 +0000 (22:10 +0200)]
forgot a comma ...