Logo AND Algorithmique Numérique Distribuée

Public GIT Repository
simgrid.git
10 years agoPut larger fields first (reduce total size for struct).
Arnaud Giersch [Thu, 31 Oct 2013 22:02:17 +0000 (23:02 +0100)]
Put larger fields first (reduce total size for struct).

10 years agoMake documentation match the code.
Arnaud Giersch [Thu, 31 Oct 2013 21:26:18 +0000 (22:26 +0100)]
Make documentation match the code.

10 years agoUse infinity (-1) for smpi/cpu_threshold in tesh files.
Arnaud Giersch [Thu, 31 Oct 2013 20:08:09 +0000 (21:08 +0100)]
Use infinity (-1) for smpi/cpu_threshold in tesh files.

10 years agoA value of -1 for smpi/cpu_threshold means infinity.
Arnaud Giersch [Thu, 31 Oct 2013 20:24:38 +0000 (21:24 +0100)]
A value of -1 for smpi/cpu_threshold means infinity.

10 years agoCache cfg parameters smpi/{cpu_threshold,running_power}.
Arnaud Giersch [Thu, 31 Oct 2013 20:24:12 +0000 (21:24 +0100)]
Cache cfg parameters smpi/{cpu_threshold,running_power}.

10 years agoCompare files.
Arnaud Giersch [Thu, 31 Oct 2013 20:47:18 +0000 (21:47 +0100)]
Compare files.

Looks like it was forgotten in commit e9befaced39d9f668fb9de3bfbc37dc7ac7cf2de.

10 years agoIt's ok to pass NULL to xbt_free.
Arnaud Giersch [Thu, 31 Oct 2013 20:36:49 +0000 (21:36 +0100)]
It's ok to pass NULL to xbt_free.

10 years agoleaks --
Augustin Degomme [Thu, 31 Oct 2013 17:08:50 +0000 (18:08 +0100)]
leaks --

10 years agoadd gatherv replay action (bigdft can be replayed easily now)
Augustin Degomme [Thu, 31 Oct 2013 17:01:52 +0000 (18:01 +0100)]
add gatherv replay action (bigdft can be replayed easily now)

10 years agoConnect new log categories.
Arnaud Giersch [Thu, 31 Oct 2013 13:37:45 +0000 (14:37 +0100)]
Connect new log categories.

10 years agoUse yes/no for tracing options instead of 1/0.
Arnaud Giersch [Thu, 31 Oct 2013 13:07:49 +0000 (14:07 +0100)]
Use yes/no for tracing options instead of 1/0.

10 years agoUse booleans instead of ints for tracing options.
Arnaud Giersch [Thu, 31 Oct 2013 13:06:57 +0000 (14:06 +0100)]
Use booleans instead of ints for tracing options.

10 years agoCosmetics.
Arnaud Giersch [Thu, 31 Oct 2013 09:35:04 +0000 (10:35 +0100)]
Cosmetics.

10 years agoDon't rely on assert for error handling.
Arnaud Giersch [Thu, 31 Oct 2013 08:53:08 +0000 (09:53 +0100)]
Don't rely on assert for error handling.

10 years agoSimplify error handling.
Arnaud Giersch [Thu, 31 Oct 2013 08:35:48 +0000 (09:35 +0100)]
Simplify error handling.

10 years agoFix computation of NUMBER_OF_PROCESSORS.
Arnaud Giersch [Thu, 31 Oct 2013 08:22:38 +0000 (09:22 +0100)]
Fix computation of NUMBER_OF_PROCESSORS.

10 years agoFree process_id, plus reindent.
Arnaud Giersch [Wed, 30 Oct 2013 20:54:36 +0000 (21:54 +0100)]
Free process_id, plus reindent.

10 years agoIgnore memory leaks in any binary from /bin or /usr/bin.
Arnaud Giersch [Wed, 30 Oct 2013 20:31:37 +0000 (21:31 +0100)]
Ignore memory leaks in any binary from /bin or /usr/bin.

10 years agoleak-- , potential segfault--
Augustin Degomme [Wed, 30 Oct 2013 17:09:37 +0000 (18:09 +0100)]
leak-- , potential segfault--

10 years agoSlightly change the sleep duration.
Arnaud Giersch [Wed, 30 Oct 2013 15:50:48 +0000 (16:50 +0100)]
Slightly change the sleep duration.

The goal is to have the same results on 32 and 64 bits, despite the different precision
for floating point operations.  To achieve this, we choose a sleep_delay that is not a
multiple of the other delays of the algorithm.  Hope that it will be sufficient.

Regenerate tesh files accordingly.

10 years agoleaks --
Augustin Degomme [Wed, 30 Oct 2013 15:54:37 +0000 (16:54 +0100)]
leaks --

10 years agorequalify a tesh
Augustin Degomme [Wed, 30 Oct 2013 14:34:31 +0000 (15:34 +0100)]
requalify a tesh

10 years agoset extra data to 0 at allocation time to avoid bad surprises
Augustin Degomme [Wed, 30 Oct 2013 14:21:26 +0000 (15:21 +0100)]
set extra data to 0 at allocation time to avoid bad surprises

10 years agochange size of sent messages for replay testing, and requalify tesh
Augustin Degomme [Wed, 30 Oct 2013 14:05:09 +0000 (15:05 +0100)]
change size of sent messages for replay testing, and requalify tesh

10 years agoAdd an Inception tesh to test TI output.
Augustin Degomme [Wed, 30 Oct 2013 13:46:16 +0000 (14:46 +0100)]
Add an Inception tesh to test TI output.
It uses pingpong executable to generate a TI trace.
Then it replays this TI trace, generating a second one.
They are then compared one to each other (after sorting), and should be identical

10 years agoadd Time independent output trace format ( TI ) to output traces that can then be...
Augustin Degomme [Wed, 30 Oct 2013 13:42:49 +0000 (14:42 +0100)]
add Time independent output trace format ( TI ) to output traces that can then be replayed
For now only handles SMPI
Using -trace-ti flag with smpirun should output a smpi_simgrid.txt file by default,
which could be then used for replay. All outputs are for now merged into only one file.

10 years agotrace MPI_Init and MPI_Finalize as well
Augustin Degomme [Fri, 25 Oct 2013 09:06:34 +0000 (11:06 +0200)]
trace MPI_Init and MPI_Finalize as well

10 years agochange the way computation tracing is handled in smpi
Augustin Degomme [Fri, 25 Oct 2013 09:05:44 +0000 (11:05 +0200)]
change the way computation tracing is handled in smpi
Now only smpi_execute_flops is surrounded by tracing calls.

10 years agoadd tracing/smpi/format option (only smpi for now), to specify which output we want
Augustin Degomme [Fri, 25 Oct 2013 08:50:30 +0000 (10:50 +0200)]
add tracing/smpi/format option (only smpi for now), to specify which output we want

10 years agoreally make replay case insensitive
Augustin Degomme [Fri, 25 Oct 2013 08:30:04 +0000 (10:30 +0200)]
really make replay case insensitive

10 years agosort output for reduce tesh
Augustin Degomme [Fri, 25 Oct 2013 08:29:05 +0000 (10:29 +0200)]
sort output for reduce tesh

10 years agoadd -trace-ti option to smpirun to generate time independant traces
Augustin Degomme [Tue, 22 Oct 2013 15:37:12 +0000 (17:37 +0200)]
add -trace-ti option to smpirun to generate time independant traces

10 years agoas replay uses a weird case for labelling its inputs, make it case insensitive
Augustin Degomme [Tue, 22 Oct 2013 15:35:01 +0000 (17:35 +0200)]
as replay uses a weird case for labelling its inputs, make it case insensitive
This will make TI traces generation easier for simgrid, without breaking existing replay

10 years agoreplace withsize variant by more generic withextra version, because we will want...
Augustin Degomme [Tue, 22 Oct 2013 12:35:02 +0000 (14:35 +0200)]
replace withsize variant by more generic withextra version, because we will want more data here

10 years agoFix another memory leak in chord example.
Arnaud Giersch [Wed, 30 Oct 2013 10:53:45 +0000 (11:53 +0100)]
Fix another memory leak in chord example.

10 years agoAvoid code duplication.
Arnaud Giersch [Wed, 30 Oct 2013 08:36:04 +0000 (09:36 +0100)]
Avoid code duplication.

10 years agoDefine symbols for builds w/o MC.
Arnaud Giersch [Wed, 30 Oct 2013 08:35:02 +0000 (09:35 +0100)]
Define symbols for builds w/o MC.

10 years agoSpurious ~.
Arnaud Giersch [Tue, 29 Oct 2013 16:49:55 +0000 (17:49 +0100)]
Spurious ~.

10 years agoFix memory leak in smpi/colls/allgather-SMP-NTS.
Arnaud Giersch [Tue, 29 Oct 2013 14:00:40 +0000 (15:00 +0100)]
Fix memory leak in smpi/colls/allgather-SMP-NTS.

Wait for isends.  Also use a dynamically allocated array for requests, and use
MPI_STATUS_IGNORE instead of an unused variable.

10 years agoOutput of "wc" is not reliable (variable number of spaces).
Arnaud Giersch [Tue, 29 Oct 2013 07:52:54 +0000 (08:52 +0100)]
Output of "wc" is not reliable (variable number of spaces).

Did not find any handy replacement for "wc -c", so only count the
number of lines in IO-bigsize.

10 years agoAdd calls to smpi_bench_{end,begin} in PMPI_Wtime.
Arnaud Giersch [Mon, 28 Oct 2013 14:33:41 +0000 (15:33 +0100)]
Add calls to smpi_bench_{end,begin} in PMPI_Wtime.

Without this change, doing something like the following always
gives a duration of zero:

    ...
    start = MPI_Wtime();
    // big computation...
    end = MPI_Wtime();
    duration = end - start;
    ...

Make sure that no calls to smpi_bench are made when process is not
between init and finalize, nor during sample blocks.

10 years agoAlign address on page boundary (fails on kfreebsd otherwise).
Arnaud Giersch [Mon, 28 Oct 2013 13:32:24 +0000 (14:32 +0100)]
Align address on page boundary (fails on kfreebsd otherwise).

10 years agoTesh sort removes empty lines, do it also with tesh.pl.
Arnaud Giersch [Mon, 28 Oct 2013 10:24:38 +0000 (11:24 +0100)]
Tesh sort removes empty lines, do it also with tesh.pl.

10 years agoAdd suport for sort prefix length in tesh.pl.
Arnaud Giersch [Mon, 28 Oct 2013 10:02:48 +0000 (11:02 +0100)]
Add suport for sort prefix length in tesh.pl.

10 years agoTypo.
Arnaud Giersch [Mon, 28 Oct 2013 09:42:18 +0000 (10:42 +0100)]
Typo.

10 years agoThere is no reason to trim the output in tesh.pl too.
Arnaud Giersch [Mon, 28 Oct 2013 09:41:19 +0000 (10:41 +0100)]
There is no reason to trim the output in tesh.pl too.

10 years agoRemove spurious spaces in tesh output.
Arnaud Giersch [Mon, 28 Oct 2013 09:40:19 +0000 (10:40 +0100)]
Remove spurious spaces in tesh output.

10 years agoAdd some tests for tesh "output sort".
Arnaud Giersch [Fri, 25 Oct 2013 07:58:05 +0000 (09:58 +0200)]
Add some tests for tesh "output sort".

10 years agoAdd more multiline tests for tesh. Also check for failures.
Arnaud Giersch [Fri, 25 Oct 2013 07:58:04 +0000 (09:58 +0200)]
Add more multiline tests for tesh.  Also check for failures.

10 years agoRename file for consistency.
Arnaud Giersch [Fri, 25 Oct 2013 07:58:04 +0000 (09:58 +0200)]
Rename file for consistency.

10 years agoPut tests for tesh output sort in a separate file.
Arnaud Giersch [Fri, 25 Oct 2013 07:58:04 +0000 (09:58 +0200)]
Put tests for tesh output sort in a separate file.

10 years agoRun tesh self tests in ${CMAKE_BINARY_DIR}/bin.
Arnaud Giersch [Fri, 25 Oct 2013 10:56:46 +0000 (12:56 +0200)]
Run tesh self tests in ${CMAKE_BINARY_DIR}/bin.

10 years agoFew updates to INSTALL file.
Arnaud Giersch [Sun, 27 Oct 2013 15:58:47 +0000 (16:58 +0100)]
Few updates to INSTALL file.

SVN replaced by Git, Ruby bindings don't belong to the main archive,
add options for Scala bindings and model-checker.

10 years agoFix spacing issues in tesh files.
Arnaud Giersch [Fri, 25 Oct 2013 07:58:05 +0000 (09:58 +0200)]
Fix spacing issues in tesh files.

10 years agoThere is no reason to trim the output.
Arnaud Giersch [Fri, 25 Oct 2013 07:58:05 +0000 (09:58 +0200)]
There is no reason to trim the output.

10 years agoCorrectly remember when "contexts/stack_size" was not set by the user.
Arnaud Giersch [Thu, 24 Oct 2013 21:00:49 +0000 (23:00 +0200)]
Correctly remember when "contexts/stack_size" was not set by the user.

The main effect is to use the system default for thread stack size.

10 years agoImprove xbt_os_thread_setstacksize().
Arnaud Giersch [Thu, 24 Oct 2013 19:55:18 +0000 (21:55 +0200)]
Improve xbt_os_thread_setstacksize().

If it fails with the requested size, try again with a multiple of PTHREAD_STACK_MIN.

10 years agoMinor corrections in ChangeLog.
Arnaud Giersch [Thu, 24 Oct 2013 15:51:09 +0000 (17:51 +0200)]
Minor corrections in ChangeLog.

10 years agoAdd name of failing function in error message.
Arnaud Giersch [Thu, 24 Oct 2013 15:29:46 +0000 (17:29 +0200)]
Add name of failing function in error message.

10 years agoTry to fix mess with exit status.
Arnaud Giersch [Thu, 24 Oct 2013 15:12:30 +0000 (17:12 +0200)]
Try to fix mess with exit status.

10 years agotesh.pl: tesh input can be received on stdin, even with no args.
Arnaud Giersch [Thu, 24 Oct 2013 08:41:07 +0000 (10:41 +0200)]
tesh.pl: tesh input can be received on stdin, even with no args.

10 years agoWith smpiff, give output file a sensible name.
Arnaud Giersch [Wed, 23 Oct 2013 16:20:30 +0000 (18:20 +0200)]
With smpiff, give output file a sensible name.

10 years agoLink fortran programs with libm.
Arnaud Giersch [Wed, 23 Oct 2013 16:18:24 +0000 (18:18 +0200)]
Link fortran programs with libm.

10 years agoAdd tests for ucontext and raw factory (storage, masterslave-virtual-machines).
Arnaud Giersch [Thu, 24 Oct 2013 14:09:09 +0000 (16:09 +0200)]
Add tests for ucontext and raw factory (storage, masterslave-virtual-machines).

10 years agoRemove debug messages from test, allow it to pass even with enable_debug=OFF.
Arnaud Giersch [Thu, 24 Oct 2013 13:57:42 +0000 (15:57 +0200)]
Remove debug messages from test, allow it to pass even with enable_debug=OFF.

Also add suffix "-thread" to tests with the thread context factory.

10 years agomodel-checker : update tesh files
Marion Guthmuller [Thu, 24 Oct 2013 13:43:18 +0000 (15:43 +0200)]
model-checker : update tesh files

10 years agomodel-checker : print depth of the counter-example in model checking statistics
Marion Guthmuller [Wed, 23 Oct 2013 13:33:50 +0000 (15:33 +0200)]
model-checker : print depth of the counter-example in model checking statistics

10 years agoDo not use the "debug" loglevel when built with NDEBUG.
Arnaud Giersch [Thu, 24 Oct 2013 12:28:30 +0000 (14:28 +0200)]
Do not use the "debug" loglevel when built with NDEBUG.

10 years agoAdding code for benchmarking code block v3_9_90
Luka Stanisic [Tue, 22 Oct 2013 08:33:02 +0000 (10:33 +0200)]
Adding code for benchmarking code block

10 years agowhen parallel ctests are performed, using the default tracing filename may cause...
Augustin Degomme [Fri, 18 Oct 2013 09:39:16 +0000 (11:39 +0200)]
when parallel ctests are performed, using the default tracing filename may cause trouble.

10 years agotry to speed up a bit testing by using all available cores in the slaves
Augustin Degomme [Thu, 17 Oct 2013 15:54:15 +0000 (17:54 +0200)]
try to speed up a bit testing by using all available cores in the slaves

10 years agowe now only have one build mode for win, as mc is irrelevant for now
Augustin Degomme [Thu, 17 Oct 2013 15:40:46 +0000 (17:40 +0200)]
we now only have one build mode for win, as mc is irrelevant for now

10 years agonot implicit on all systems
Augustin Degomme [Thu, 17 Oct 2013 12:36:53 +0000 (14:36 +0200)]
not implicit on all systems

10 years agoadd background commands support + correct error code for output mismatch
Augustin Degomme [Wed, 16 Oct 2013 16:26:19 +0000 (18:26 +0200)]
add background commands support + correct error code for output mismatch

10 years agoUse temporary files to store output of the command in perl tesh
Augustin Degomme [Tue, 15 Oct 2013 15:33:57 +0000 (17:33 +0200)]
Use temporary files to store output of the command in perl tesh
Avoids deadlocking when large data is sent to/from pipes (see tesh test IO-bigsize)

10 years agowin is not yet ready for coverage either
Augustin Degomme [Mon, 14 Oct 2013 19:15:44 +0000 (21:15 +0200)]
win is not yet ready for coverage either

10 years agowin is not yet ready for warning removal
Augustin Degomme [Mon, 14 Oct 2013 19:02:28 +0000 (21:02 +0200)]
win is not yet ready for warning removal

10 years agoAdd error checking to the script
Augustin Degomme [Mon, 14 Oct 2013 16:13:15 +0000 (18:13 +0200)]
Add error checking to the script

10 years agoavoid failing the tar extraction and working on previous version instead
Augustin Degomme [Mon, 14 Oct 2013 15:19:45 +0000 (17:19 +0200)]
avoid failing the tar extraction and working on previous version instead

10 years agofix run script
Augustin Degomme [Mon, 14 Oct 2013 14:03:32 +0000 (16:03 +0200)]
fix run script

10 years agounify version name of executable
Augustin Degomme [Mon, 14 Oct 2013 13:54:07 +0000 (15:54 +0200)]
unify version name of executable

10 years agoLet's try to unify build rules for jenkins
Augustin Degomme [Mon, 14 Oct 2013 13:18:12 +0000 (15:18 +0200)]
Let's try to unify build rules for jenkins

10 years agoWin + cmake wants perl to be explicit here.. why not?
Augustin Degomme [Mon, 14 Oct 2013 12:29:08 +0000 (14:29 +0200)]
Win + cmake wants perl to be explicit here.. why not?

10 years agoWindows error handling some errors for you and changing them to return codes, these...
Augustin Degomme [Mon, 14 Oct 2013 11:47:02 +0000 (13:47 +0200)]
Windows error handling some errors for you and changing them to return codes, these won't work on win simply

10 years agoThis can actually be built and run in win
Augustin Degomme [Mon, 14 Oct 2013 11:44:06 +0000 (13:44 +0200)]
This can actually be built and run in win

10 years agochomp may not be enough in some cases + tell win that we want 2 digits only for exponents
Augustin Degomme [Thu, 10 Oct 2013 15:17:35 +0000 (17:17 +0200)]
chomp may not be enough in some cases + tell win that we want 2 digits only for exponents

10 years agomore windows compliancy fun
Augustin Degomme [Thu, 10 Oct 2013 14:41:42 +0000 (16:41 +0200)]
more windows compliancy fun

10 years agooops
Augustin Degomme [Thu, 10 Oct 2013 13:31:04 +0000 (15:31 +0200)]
oops

10 years agoForgotten by commit 8c9500c.
Arnaud Giersch [Thu, 10 Oct 2013 13:54:18 +0000 (15:54 +0200)]
Forgotten by commit 8c9500c.

10 years agogenerate this file at testtime, as it failed with out of source windows builds
Augustin Degomme [Thu, 10 Oct 2013 12:58:20 +0000 (14:58 +0200)]
generate this file at testtime, as it failed with out of source windows builds

10 years agoreplace signal number by its name, for better portability
Augustin Degomme [Thu, 10 Oct 2013 09:41:15 +0000 (11:41 +0200)]
replace signal number by its name, for better portability

10 years agowe wait for 255 in the tesh, so be it.
Augustin Degomme [Thu, 10 Oct 2013 09:18:10 +0000 (11:18 +0200)]
we wait for 255 in the tesh, so be it.

10 years agothis was not built on win
Augustin Degomme [Thu, 10 Oct 2013 08:56:10 +0000 (10:56 +0200)]
this was not built on win

10 years agoOnce again, Linux does not follow the real way of doing things.
Arnaud Giersch [Thu, 10 Oct 2013 12:39:30 +0000 (14:39 +0200)]
Once again, Linux does not follow the real way of doing things.

The default buffering mode for output streams seems different on
Windows (lines don't mix in the same order between stdout and stderr).
Use unbuffered mode, since line buffer mode is not supported on these
systems.

See http://msdn.microsoft.com/en-us/library/86cebhfs%28v=vs.71%29.aspx
   _IOLBF
     For some systems, this provides line buffering. However, for Win32,
     the behavior is the same as _IOFBF - Full Buffering."

Let's wait for the cdash to confirm that it works...

10 years agoat least this one works. leak--
suter [Thu, 10 Oct 2013 10:15:14 +0000 (12:15 +0200)]
at least this one works. leak--

10 years agoChasing more leaks.
Arnaud Giersch [Thu, 10 Oct 2013 10:10:52 +0000 (12:10 +0200)]
Chasing more leaks.

10 years agokill the timeout process not only for last command, but for each one
Augustin Degomme [Thu, 10 Oct 2013 08:23:52 +0000 (10:23 +0200)]
kill the timeout process not only for last command, but for each one

10 years agoLength modifier 'l' is indeed wrong in %l[efg] for printf.
Arnaud Giersch [Wed, 9 Oct 2013 20:10:33 +0000 (22:10 +0200)]
Length modifier 'l' is indeed wrong in %l[efg] for printf.

See e.g. http://c-faq.com/stdio/scanfvsprintf.html
Change it everywhere.

10 years agoturns out %le=%e and %lf=%f ... except on windows
Augustin Degomme [Wed, 9 Oct 2013 14:19:16 +0000 (16:19 +0200)]
turns out %le=%e and %lf=%f ... except on windows

10 years agoUse sg_storage_size_t in Java bindings too.
Arnaud Giersch [Wed, 9 Oct 2013 12:40:26 +0000 (14:40 +0200)]
Use sg_storage_size_t in Java bindings too.