Martin Quinson [Sun, 30 Apr 2017 20:14:50 +0000 (22:14 +0200)]
reduce the amount of static_cast in smx_network
Martin Quinson [Sat, 29 Apr 2017 22:00:21 +0000 (00:00 +0200)]
fix some doxygen errors
Martin Quinson [Sat, 29 Apr 2017 21:42:06 +0000 (23:42 +0200)]
Hide a gcc attribute that drives doxygen nuts
Martin Quinson [Sat, 29 Apr 2017 21:38:38 +0000 (23:38 +0200)]
doc improvements: mainly intro of platf and deployment
Martin Quinson [Fri, 28 Apr 2017 08:09:29 +0000 (10:09 +0200)]
commit the script I use to compile simgrid
Martin Quinson [Thu, 27 Apr 2017 20:10:08 +0000 (22:10 +0200)]
improve a tiny bit the doc introduction
I wish I will find the time to improve it further soon.
Martin Quinson [Wed, 26 Apr 2017 14:50:57 +0000 (16:50 +0200)]
Strengthen the memory ordering order on that refcounting to work around a bug
Martin Quinson [Wed, 26 Apr 2017 11:56:10 +0000 (13:56 +0200)]
comment++ on that code that I barely understand but exhibits bugs
Frederic Suter [Wed, 26 Apr 2017 07:16:17 +0000 (09:16 +0200)]
this is now useless
Martin Quinson [Mon, 24 Apr 2017 23:29:07 +0000 (01:29 +0200)]
sort logically the fields of ActorImpl
Martin Quinson [Mon, 24 Apr 2017 23:24:09 +0000 (01:24 +0200)]
introduce type aid_t for Actor's ID (ie, PID)
Frederic Suter [Mon, 24 Apr 2017 20:07:38 +0000 (22:07 +0200)]
one xbt_lib less in storage
still clumsy
Frederic Suter [Mon, 24 Apr 2017 15:45:16 +0000 (17:45 +0200)]
simplify way files are handled by MSG
Martin Quinson [Mon, 24 Apr 2017 13:43:33 +0000 (15:43 +0200)]
try to use more standard, less obscure synchro for refcounting
Frederic Suter [Mon, 24 Apr 2017 13:17:06 +0000 (15:17 +0200)]
xbt_strbuff to std::string in cpp files
Martin Quinson [Mon, 24 Apr 2017 12:46:28 +0000 (14:46 +0200)]
New function xbt_replay_action_get(): Retrieve the function previously associated to an event type.
Martin Quinson [Mon, 24 Apr 2017 06:35:13 +0000 (08:35 +0200)]
fix my late sunday commit
Martin Quinson [Sun, 23 Apr 2017 20:56:35 +0000 (22:56 +0200)]
reduce the verbosity of that message
Martin Quinson [Sat, 22 Apr 2017 19:25:06 +0000 (21:25 +0200)]
32bits I hate you.
Martin Quinson [Sat, 22 Apr 2017 13:25:00 +0000 (15:25 +0200)]
damn 32bits
Martin Quinson [Sat, 22 Apr 2017 07:32:32 +0000 (09:32 +0200)]
try to avoid that others fall into the trap where I was yesterday
not quite sucessfull: the failure occures before main :(
Martin Quinson [Fri, 21 Apr 2017 16:20:24 +0000 (18:20 +0200)]
check that the sendfile worked correctly or die verbosely if not
Martin Quinson [Fri, 21 Apr 2017 13:20:50 +0000 (15:20 +0200)]
inline a function (to please sonar)
Martin Quinson [Thu, 20 Apr 2017 22:34:58 +0000 (00:34 +0200)]
more informative msg when the dlopen of mmap priv fails
as it appears to be the case on freebsd
Martin Quinson [Thu, 20 Apr 2017 14:45:41 +0000 (16:45 +0200)]
please sonar a tiny bit
Martin Quinson [Thu, 20 Apr 2017 14:44:29 +0000 (16:44 +0200)]
don't suppose anythin on privatization when threads/ucontextes are not found
Martin Quinson [Thu, 20 Apr 2017 13:45:29 +0000 (15:45 +0200)]
fix a broken condition in the tests
Martin Quinson [Thu, 20 Apr 2017 13:43:58 +0000 (15:43 +0200)]
dlopen: give one fdin per thread to avoid stupid races resulting in empty copied files
Martin Quinson [Thu, 20 Apr 2017 13:43:01 +0000 (15:43 +0200)]
document the dlopen thing in the changelog
Martin Quinson [Thu, 20 Apr 2017 12:52:59 +0000 (14:52 +0200)]
I cannot even use xbt_die correctly anymore, bummer
Martin Quinson [Thu, 20 Apr 2017 12:40:27 +0000 (14:40 +0200)]
Even more informative error messages when dlopen fails
Martin Quinson [Thu, 20 Apr 2017 12:35:52 +0000 (14:35 +0200)]
be more verbose when dlopen fails
degomme [Thu, 20 Apr 2017 12:30:32 +0000 (14:30 +0200)]
typo --
Martin Quinson [Thu, 20 Apr 2017 11:01:47 +0000 (13:01 +0200)]
missed these spots while renaming an option
Martin Quinson [Thu, 20 Apr 2017 11:00:04 +0000 (13:00 +0200)]
test dlopen privatization too
Martin Quinson [Thu, 20 Apr 2017 10:57:46 +0000 (12:57 +0200)]
fix the dlopen privatization (fix #157)
Thanks to @Ezibenroc for the report and the patch
Martin Quinson [Thu, 20 Apr 2017 10:27:48 +0000 (12:27 +0200)]
rename option smpi/privatize-global-variables to smpi/privatization
degomme [Thu, 20 Apr 2017 08:54:34 +0000 (10:54 +0200)]
fix typos in cmake, preventing include paths from being correctly added
Martin Quinson [Wed, 19 Apr 2017 08:07:48 +0000 (10:07 +0200)]
improve the documentation of smpi/privatize-global-variables
Martin Quinson [Tue, 18 Apr 2017 13:41:19 +0000 (15:41 +0200)]
please sonar a tiny bit
Martin Quinson [Mon, 17 Apr 2017 07:27:33 +0000 (09:27 +0200)]
SMPI cleanups: rename a symbol and remove unused parameters
Martin Quinson [Mon, 17 Apr 2017 07:27:06 +0000 (09:27 +0200)]
ignore newly activated mpich tests
Martin Quinson [Mon, 17 Apr 2017 07:25:10 +0000 (09:25 +0200)]
small sonar-inspired cleanups
Martin Quinson [Sun, 16 Apr 2017 20:33:33 +0000 (22:33 +0200)]
please sonar and cosmetics
Martin Quinson [Sun, 16 Apr 2017 20:16:00 +0000 (22:16 +0200)]
make the mpich3 tests compile with the warnings on my machine
Martin Quinson [Sat, 15 Apr 2017 22:11:45 +0000 (00:11 +0200)]
please sonar: kill unused parameters
and also declare each variable on its own line, even if that smell is
more supurious
Martin Quinson [Sat, 15 Apr 2017 21:55:47 +0000 (23:55 +0200)]
fix the copyright date on these files
Tom Cornebize [Sat, 15 Apr 2017 21:18:01 +0000 (23:18 +0200)]
Reduce the size of partial shared malloc tests.
Martin Quinson [Fri, 14 Apr 2017 17:13:06 +0000 (19:13 +0200)]
damn, there is more of these nasty %lx around there
Martin Quinson [Fri, 14 Apr 2017 16:34:20 +0000 (18:34 +0200)]
try to fix 32bits builds
Augustin Degomme [Fri, 14 Apr 2017 15:39:28 +0000 (17:39 +0200)]
fix 32 bits builds ... probably
Augustin Degomme [Fri, 14 Apr 2017 14:57:09 +0000 (16:57 +0200)]
fix dist
Martin Quinson [Fri, 14 Apr 2017 13:26:13 +0000 (15:26 +0200)]
Merge branch 'master' of github.com:simgrid/simgrid
Martin Quinson [Fri, 14 Apr 2017 13:25:40 +0000 (15:25 +0200)]
this static function is now useless
Martin Quinson [Fri, 14 Apr 2017 13:25:07 +0000 (15:25 +0200)]
cosmetics
Martin Quinson [Fri, 14 Apr 2017 13:10:07 +0000 (15:10 +0200)]
Merge pull request #156 from simgrid/munmap_size
Call munmap with the size in smpi_shared_free
Tom Cornebize [Fri, 14 Apr 2017 09:23:49 +0000 (11:23 +0200)]
Call munmap with the size in smpi_shared_free.
degomme [Fri, 14 Apr 2017 08:18:44 +0000 (10:18 +0200)]
Patch up smpi process creation ...
Dirty, as this will call smpi_init_options twice, so add a return for the second call.
Martin Quinson [Thu, 13 Apr 2017 20:13:37 +0000 (22:13 +0200)]
Merge pull request #154 from simgrid/partial_shared_malloc
Partial shared malloc
degomme [Thu, 13 Apr 2017 07:34:23 +0000 (09:34 +0200)]
seems needed here.
Martin Quinson [Thu, 13 Apr 2017 06:00:48 +0000 (08:00 +0200)]
Portability fun with FreeBSD
Make sure that the lambda capture parameter is actually used when
sendfile(2) is not usable, to avoid a fatal warning: FreeBSD is both
super advanced (with clang being very uptodate) and somewhat
prehistorical.
And no, I do not plan to deal myself with the FreeBSD-specific
sendfile prototype to avoid this problem in a cleaner way. But patches
are welcome as usual.
Martin Quinson [Thu, 13 Apr 2017 01:17:35 +0000 (03:17 +0200)]
portable implementation of sendfile
Martin Quinson [Wed, 12 Apr 2017 23:18:15 +0000 (01:18 +0200)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Wed, 12 Apr 2017 23:12:03 +0000 (01:12 +0200)]
dl_open priv: manually copy the file w/o using cp
Martin Quinson [Wed, 12 Apr 2017 22:07:39 +0000 (00:07 +0200)]
make smpirun use SMPI_init to unify call paths
Martin Quinson [Wed, 12 Apr 2017 21:52:12 +0000 (23:52 +0200)]
use surf init/fini callbacks, no need for TRACE ones
Tom Cornebize [Wed, 12 Apr 2017 21:39:23 +0000 (23:39 +0200)]
Fix tesh tests.
Martin Quinson [Wed, 12 Apr 2017 21:22:03 +0000 (23:22 +0200)]
Increase a timeout to help TSan converging
Tom Cornebize [Wed, 12 Apr 2017 21:19:41 +0000 (23:19 +0200)]
Remove useless code.
Tom Cornebize [Wed, 12 Apr 2017 21:13:02 +0000 (23:13 +0200)]
Merge branch 'master' into partial_shared_malloc
Christian Heinrich [Wed, 12 Apr 2017 09:48:31 +0000 (11:48 +0200)]
[SMPI] Updated the SmpiHost class
- Now supports oisend, osend, orecv values on a per-host basis
- Removed cruft
Christian Heinrich [Fri, 6 Jan 2017 13:38:23 +0000 (14:38 +0100)]
[SMPI] Use SmpiHost::orecv().
I moved the computation of the orecv configs to the SmpiHost class.
It's currently a crude implementation and does not (yet) allow for a
per-host configuration, but this is planned and outlined in the code.
For that, all we need is an option to parse and set orecv values on
a per-node basis. (See the orecv_ property.)
Christian Heinrich [Fri, 6 Jan 2017 13:31:08 +0000 (14:31 +0100)]
[SMPI] Added SmpiHost.cpp and .hpp
The SmpiHost is not yet fully implemented but is
the first step towards a more configurable SMPI.
Martin Quinson [Wed, 12 Apr 2017 09:14:06 +0000 (11:14 +0200)]
http://abstrusegoose.com/432
Martin Quinson [Wed, 12 Apr 2017 08:37:47 +0000 (10:37 +0200)]
MC: further snapshot comparison simplifications
Martin Quinson [Wed, 12 Apr 2017 08:29:02 +0000 (10:29 +0200)]
This is a for loop
Martin Quinson [Wed, 12 Apr 2017 08:21:08 +0000 (10:21 +0200)]
MC: snapshot comparison simplification
Frederic Suter [Wed, 12 Apr 2017 07:25:22 +0000 (09:25 +0200)]
fix inconsistencies between smpi_replay and instr_TI_trace.cpp
Frederic Suter [Wed, 12 Apr 2017 06:44:48 +0000 (08:44 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Martin Quinson [Wed, 12 Apr 2017 05:44:06 +0000 (07:44 +0200)]
SMPI: simplify process creation
finalization_barrier is either created when the processes are created
from smpirun, or taken from the instance if any. So there is no need
for a setter
This could be further simplified by making a default instance when
starting from smpirun. I just need to understand how to make this move
in the current swamp of corner cases...
Martin Quinson [Tue, 11 Apr 2017 22:09:06 +0000 (00:09 +0200)]
SMPI: simplifies the process destruction
Martin Quinson [Tue, 11 Apr 2017 21:38:19 +0000 (23:38 +0200)]
regenerate the smpi_location thingy
Martin Quinson [Tue, 11 Apr 2017 21:37:47 +0000 (23:37 +0200)]
ignore the newly created test files
degomme [Tue, 11 Apr 2017 21:53:08 +0000 (23:53 +0200)]
Avoid trailing whitespaces when dl not found
... which is an issue by itself, but too late, martin will fix it tomorrow 8 a.m.
degomme [Tue, 11 Apr 2017 21:32:20 +0000 (23:32 +0200)]
Ooooops ²
degomme [Tue, 11 Apr 2017 21:25:51 +0000 (23:25 +0200)]
Fix .. something ?
degomme [Tue, 11 Apr 2017 17:09:10 +0000 (19:09 +0200)]
fix mc build
degomme [Tue, 11 Apr 2017 15:21:08 +0000 (17:21 +0200)]
ooooops
degomme [Tue, 11 Apr 2017 13:43:29 +0000 (15:43 +0200)]
fix smpi tests
degomme [Tue, 11 Apr 2017 13:22:41 +0000 (15:22 +0200)]
Yet another dlopen merge
Martin Quinson [Tue, 11 Apr 2017 20:53:01 +0000 (22:53 +0200)]
kill useless code
Martin Quinson [Tue, 11 Apr 2017 20:31:36 +0000 (22:31 +0200)]
simplify SMPI process creation a tiny bit
Martin Quinson [Tue, 11 Apr 2017 17:17:45 +0000 (19:17 +0200)]
try to cleanup how MPI process are created. First act.
Martin Quinson [Tue, 11 Apr 2017 16:00:33 +0000 (18:00 +0200)]
travis: activate optimizations, since the tests are taking for ever
Tom Cornebize [Tue, 11 Apr 2017 14:40:41 +0000 (16:40 +0200)]
Remove warnings.
Martin Quinson [Tue, 11 Apr 2017 13:30:02 +0000 (15:30 +0200)]
please sonar by killing useless parameters and cleanups
Martin Quinson [Tue, 11 Apr 2017 13:09:25 +0000 (15:09 +0200)]
typos
degomme [Tue, 11 Apr 2017 12:28:55 +0000 (14:28 +0200)]
Iprobe was actually ignored if MPI_STATUS_IGNORE was set, which is a bug.
Martin Quinson [Mon, 10 Apr 2017 23:26:26 +0000 (01:26 +0200)]
MC: this bug was disabling state equality detection
The variable is an integer being the maximal amount of state that we
save at the same time (typically 1000). I'm not sure which int value
is equal to "true", but I guess it's not 1000.
Martin Quinson [Mon, 10 Apr 2017 23:21:38 +0000 (01:21 +0200)]
typos