Logo AND Algorithmique Numérique Distribuée

Public GIT Repository
simgrid.git
2 years ago[SMPI] Added SmpiHost.cpp and .hpp
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.

2 years agohttp://abstrusegoose.com/432
Martin Quinson [Wed, 12 Apr 2017 09:14:06 +0000 (11:14 +0200)]
http://abstrusegoose.com/432

2 years agoMC: further snapshot comparison simplifications
Martin Quinson [Wed, 12 Apr 2017 08:37:47 +0000 (10:37 +0200)]
MC: further snapshot comparison simplifications

2 years agoThis is a for loop
Martin Quinson [Wed, 12 Apr 2017 08:29:02 +0000 (10:29 +0200)]
This is a for loop

2 years agoMC: snapshot comparison simplification
Martin Quinson [Wed, 12 Apr 2017 08:21:08 +0000 (10:21 +0200)]
MC: snapshot comparison simplification

2 years agofix inconsistencies between smpi_replay and instr_TI_trace.cpp
Frederic Suter [Wed, 12 Apr 2017 07:25:22 +0000 (09:25 +0200)]
fix inconsistencies between smpi_replay and instr_TI_trace.cpp

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
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

2 years agoSMPI: simplify process creation
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...

2 years agoSMPI: simplifies the process destruction
Martin Quinson [Tue, 11 Apr 2017 22:09:06 +0000 (00:09 +0200)]
SMPI: simplifies the process destruction

2 years agoregenerate the smpi_location thingy
Martin Quinson [Tue, 11 Apr 2017 21:38:19 +0000 (23:38 +0200)]
regenerate the smpi_location thingy

2 years agoignore the newly created test files
Martin Quinson [Tue, 11 Apr 2017 21:37:47 +0000 (23:37 +0200)]
ignore the newly created test files

2 years agoAvoid trailing whitespaces when dl not found
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.

2 years agoOoooops ²
degomme [Tue, 11 Apr 2017 21:32:20 +0000 (23:32 +0200)]
Ooooops ²

2 years agoFix .. something ?
degomme [Tue, 11 Apr 2017 21:25:51 +0000 (23:25 +0200)]
Fix .. something ?

2 years agofix mc build
degomme [Tue, 11 Apr 2017 17:09:10 +0000 (19:09 +0200)]
fix mc build

2 years agoooooops
degomme [Tue, 11 Apr 2017 15:21:08 +0000 (17:21 +0200)]
ooooops

2 years agofix smpi tests
degomme [Tue, 11 Apr 2017 13:43:29 +0000 (15:43 +0200)]
fix smpi tests

2 years agoYet another dlopen merge
degomme [Tue, 11 Apr 2017 13:22:41 +0000 (15:22 +0200)]
Yet another dlopen merge

2 years agokill useless code
Martin Quinson [Tue, 11 Apr 2017 20:53:01 +0000 (22:53 +0200)]
kill useless code

2 years agosimplify SMPI process creation a tiny bit
Martin Quinson [Tue, 11 Apr 2017 20:31:36 +0000 (22:31 +0200)]
simplify SMPI process creation a tiny bit

2 years agotry to cleanup how MPI process are created. First act.
Martin Quinson [Tue, 11 Apr 2017 17:17:45 +0000 (19:17 +0200)]
try to cleanup how MPI process are created. First act.

2 years agotravis: activate optimizations, since the tests are taking for ever
Martin Quinson [Tue, 11 Apr 2017 16:00:33 +0000 (18:00 +0200)]
travis: activate optimizations, since the tests are taking for ever

2 years agoplease sonar by killing useless parameters and cleanups
Martin Quinson [Tue, 11 Apr 2017 13:30:02 +0000 (15:30 +0200)]
please sonar by killing useless parameters and cleanups

2 years agotypos
Martin Quinson [Tue, 11 Apr 2017 13:09:25 +0000 (15:09 +0200)]
typos

2 years agoIprobe was actually ignored if MPI_STATUS_IGNORE was set, which is a bug.
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.

2 years agoMC: this bug was disabling state equality detection
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.

2 years agotypos
Martin Quinson [Mon, 10 Apr 2017 23:21:38 +0000 (01:21 +0200)]
typos

2 years agobetter handling of the MC debug on unknown simcalls
Martin Quinson [Mon, 10 Apr 2017 22:21:34 +0000 (00:21 +0200)]
better handling of the MC debug on unknown simcalls

2 years agoDue to recent "improvements" in the smpi mpich rma tests, netbsd timeouts. Avoid...
degomme [Mon, 10 Apr 2017 16:28:54 +0000 (18:28 +0200)]
Due to recent "improvements" in the smpi mpich rma tests, netbsd timeouts. Avoid this.

2 years agoAdd two new mpich tests.
degomme [Mon, 10 Apr 2017 16:06:17 +0000 (18:06 +0200)]
Add two new mpich tests.
Taken from https://github.com/pmodels/mpich/commit/c4d2d5b88d530f1f8d3a74fe725b8fa30f57137e and https://github.com/pmodels/mpich/commit/cc7f28a1442f8bf731e174df334ab5e770fc88b9 .
They are rather simple and useless for us, but I'm tracking all changes to mpich test tree now to ease updates.

2 years ago[SMX] Fixed wrong termination timing when daemons are present
Christian Heinrich [Fri, 7 Apr 2017 12:20:18 +0000 (14:20 +0200)]
[SMX] Fixed wrong termination timing when daemons are present

The check whether daemons need to be killed because no real process is present
anymore needs to be performed every time a process could potentially have exited.

2 years agofix a dead link
Martin Quinson [Fri, 7 Apr 2017 08:56:43 +0000 (10:56 +0200)]
fix a dead link

2 years agofix some newly introduced smells
Frederic Suter [Fri, 7 Apr 2017 08:21:28 +0000 (10:21 +0200)]
fix some newly introduced smells

2 years agobe more robust to plugins creating processes a bit early
Martin Quinson [Thu, 6 Apr 2017 14:57:19 +0000 (16:57 +0200)]
be more robust to plugins creating processes a bit early

2 years agotry to appease ubsan a bit
degomme [Thu, 6 Apr 2017 09:03:23 +0000 (11:03 +0200)]
try to appease ubsan a bit

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Thu, 6 Apr 2017 06:47:09 +0000 (08:47 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agoplug some leak by delete last message before leaving
Frederic Suter [Thu, 6 Apr 2017 06:46:36 +0000 (08:46 +0200)]
plug some leak by delete last message before leaving

2 years agoactivate some more rma tests
degomme [Wed, 5 Apr 2017 22:39:03 +0000 (00:39 +0200)]
activate some more rma tests

2 years agocornercases handling
degomme [Wed, 5 Apr 2017 22:28:10 +0000 (00:28 +0200)]
cornercases handling

2 years ago[mc] Install simgrid-mc
Gabriel Corona [Wed, 5 Apr 2017 22:06:49 +0000 (00:06 +0200)]
[mc] Install simgrid-mc

2 years agotry to please MC
Frederic Suter [Wed, 5 Apr 2017 18:25:53 +0000 (20:25 +0200)]
try to please MC

2 years agonew s4u example: chord
Frederic Suter [Wed, 5 Apr 2017 16:56:53 +0000 (18:56 +0200)]
new s4u example: chord
leaking a lot, not perfect, but hopefully better than the C version

2 years agoSuper lazily avoid globals in this test.
degomme [Wed, 5 Apr 2017 16:24:20 +0000 (18:24 +0200)]
Super lazily avoid globals in this test.

2 years agoActivate some more tests.
degomme [Wed, 5 Apr 2017 15:06:25 +0000 (17:06 +0200)]
Activate some more tests.
We almost have everything in MPI3 RMAs, now.
I will try to activate more tests soon.

2 years agoPotential crash --
degomme [Wed, 5 Apr 2017 15:05:22 +0000 (17:05 +0200)]
Potential crash --

2 years agoHave lock and unlock actually behave like they should (terminating both remote and...
degomme [Wed, 5 Apr 2017 15:05:08 +0000 (17:05 +0200)]
Have lock and unlock actually behave like they should (terminating both remote and local requests)

2 years agoPotential crash --
degomme [Wed, 5 Apr 2017 15:04:19 +0000 (17:04 +0200)]
Potential crash --

2 years agoActually we can have locks and fences (erk) in the same program.
degomme [Wed, 5 Apr 2017 15:04:05 +0000 (17:04 +0200)]
Actually we can have locks and fences (erk) in the same program.

2 years agoAdd MPI_Rput, Rget, Raccumulate and Rget_accumulate calls.
degomme [Wed, 5 Apr 2017 15:03:26 +0000 (17:03 +0200)]
Add MPI_Rput, Rget, Raccumulate and Rget_accumulate calls.
They return the request for separate termination instead of putting it in the queue.

2 years agoAdd MPI_Compare_and_swap (normally atomic) call
degomme [Wed, 5 Apr 2017 15:01:59 +0000 (17:01 +0200)]
Add MPI_Compare_and_swap (normally atomic) call

2 years agoWe need this order of parameters, as the user may want to override the default privat...
degomme [Wed, 5 Apr 2017 13:11:14 +0000 (15:11 +0200)]
We need this order of parameters, as the user may want to override the default privatization command.

2 years ago[FIX] Fix broken commit 770f1cc88b3b35c90e505470753b70b95a326b91
Christian Heinrich [Wed, 5 Apr 2017 12:50:44 +0000 (14:50 +0200)]
[FIX] Fix broken commit 770f1cc88b3b35c90e505470753b70b95a326b91

2 years ago[PLUGINS] Added HostLoad::getAverageLoad() to the HostLoad plugin
Christian Heinrich [Wed, 5 Apr 2017 12:32:55 +0000 (14:32 +0200)]
[PLUGINS] Added HostLoad::getAverageLoad() to the HostLoad plugin

This functions returns the average load since the last reset

2 years agoactivate tests
degomme [Tue, 4 Apr 2017 08:06:04 +0000 (10:06 +0200)]
activate tests

2 years agoAdd MPI_Win_lock_all MPI_Win_unlock_all MPI_Win_flush MPI_Win_flush_local MPI_Win_flu...
degomme [Tue, 4 Apr 2017 08:05:58 +0000 (10:05 +0200)]
Add MPI_Win_lock_all MPI_Win_unlock_all MPI_Win_flush MPI_Win_flush_local MPI_Win_flush_all MPI_Win_flush_local_all calls

2 years agoOops, this has to be secured by the mutex
degomme [Tue, 4 Apr 2017 00:27:50 +0000 (02:27 +0200)]
Oops, this has to be secured by the mutex

2 years agoAdd MPI_Fetch_and_op call, which is just a get_accumulate with specific params
degomme [Mon, 3 Apr 2017 12:42:22 +0000 (14:42 +0200)]
Add MPI_Fetch_and_op call, which is just a get_accumulate with specific params

2 years agoOops, we need to store a pointer to the simix process in the smpi process now.
degomme [Mon, 3 Apr 2017 12:41:23 +0000 (14:41 +0200)]
Oops, we need to store a pointer to the simix process in the smpi process now.

2 years agoNo_op has a specific effect, actually, and turns the get_accumulate in a get.
degomme [Mon, 3 Apr 2017 12:38:20 +0000 (14:38 +0200)]
No_op has a specific effect, actually, and turns the get_accumulate in a get.
The standard says it may differ a bit internally.. Or not.

2 years agoWe can start the recv requests directly now, and don't have to wait for a synchro.
degomme [Mon, 3 Apr 2017 12:37:19 +0000 (14:37 +0200)]
We can start the recv requests directly now, and don't have to wait for a synchro.
Doesn't change much as they were async anyway, but it's simpler.

2 years agocleanup, and allow an external process to start a MPI request.
degomme [Mon, 3 Apr 2017 12:36:13 +0000 (14:36 +0200)]
cleanup, and allow an external process to start a MPI request.

2 years agoclose the window if we explicitely tell it's closed
degomme [Mon, 3 Apr 2017 12:35:19 +0000 (14:35 +0200)]
close the window if we explicitely tell it's closed

2 years agono fun if you don't break evertyhing 'cos you're dumb...
Frederic Suter [Tue, 4 Apr 2017 07:45:28 +0000 (09:45 +0200)]
no fun if you don't break evertyhing 'cos you're dumb...

2 years agoadd send with timeout and irecv to this_actor API
Frederic Suter [Tue, 4 Apr 2017 07:30:18 +0000 (09:30 +0200)]
add send with timeout and irecv to this_actor API

2 years agofix bug in asynchronous receive
Frederic Suter [Tue, 4 Apr 2017 07:03:53 +0000 (09:03 +0200)]
fix bug in asynchronous receive
(https://github.com/simgrid/simgrid/issues/144)

2 years agoReally don't do anything in the copy callback if we are in a shared zone.
degomme [Fri, 31 Mar 2017 13:03:39 +0000 (15:03 +0200)]
Really don't do anything in the copy callback if we are in a shared zone.
Attempt to fix https://github.com/simgrid/simgrid/issues/147

2 years agoIncrease our max limit for tag.. because 1M was low, indeed.
degomme [Fri, 31 Mar 2017 12:01:59 +0000 (14:01 +0200)]
Increase our max limit for tag.. because 1M was low, indeed.

2 years agoHide xbt_cfg cruft from smpi tesh files.
degomme [Fri, 31 Mar 2017 09:45:08 +0000 (11:45 +0200)]
Hide xbt_cfg cruft from smpi tesh files.
We need this as privatization may not be activated on some systems, thus outputting a different thing.

2 years agoStop lying. Really activate privatization by default (when supported)
degomme [Fri, 31 Mar 2017 09:44:05 +0000 (11:44 +0200)]
Stop lying. Really activate privatization by default (when supported)

2 years agokill dead code
Frederic Suter [Fri, 31 Mar 2017 07:19:00 +0000 (09:19 +0200)]
kill dead code
heterogenous dicts disappeared a while ago

2 years agofix a bunch of recently introduced smells
Frederic Suter [Fri, 31 Mar 2017 07:08:08 +0000 (09:08 +0200)]
fix a bunch of recently introduced smells

2 years agoanother leak plugged
Frederic Suter [Thu, 30 Mar 2017 14:48:48 +0000 (16:48 +0200)]
another leak plugged

2 years agoleak plug
Frederic Suter [Thu, 30 Mar 2017 14:46:32 +0000 (16:46 +0200)]
leak plug

2 years agodo not initialize surf for nothing
Frederic Suter [Thu, 30 Mar 2017 14:06:53 +0000 (16:06 +0200)]
do not initialize surf for nothing

2 years agoattempt to plug all leaks for simdag.
Frederic Suter [Thu, 30 Mar 2017 12:17:10 +0000 (14:17 +0200)]
attempt to plug all leaks for simdag.
only fill a dynar if you need it, and in this case allocate and free
it yourself

2 years agoremove dead code
Frederic Suter [Thu, 30 Mar 2017 11:19:56 +0000 (13:19 +0200)]
remove dead code

2 years agomodernize two simcalls
Frederic Suter [Thu, 30 Mar 2017 10:49:54 +0000 (12:49 +0200)]
modernize two simcalls
and use a third in msg_io

2 years agoRevert "Revert "cast once for all at surf level and not in the APIs""
Frederic Suter [Thu, 30 Mar 2017 10:37:10 +0000 (12:37 +0200)]
Revert "Revert "cast once for all at surf level and not in the APIs""

This reverts commit fa900b346af7c55195fd90c3345cc4107c2afc6d.

2 years agoRevert "cast once for all at surf level and not in the APIs"
Frederic Suter [Thu, 30 Mar 2017 10:35:31 +0000 (12:35 +0200)]
Revert "cast once for all at surf level and not in the APIs"

This reverts commit 14696475e1a6145501ae0f9d6104a82ab2ffa64d.

2 years agocast once for all at surf level and not in the APIs
Frederic Suter [Thu, 30 Mar 2017 10:15:59 +0000 (12:15 +0200)]
cast once for all at surf level and not in the APIs

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Thu, 30 Mar 2017 08:56:31 +0000 (10:56 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agocomplete s4u::Storage interface, what a mess...
Frederic Suter [Thu, 30 Mar 2017 08:53:03 +0000 (10:53 +0200)]
complete s4u::Storage interface, what a mess...
not all storages are put in the static storages_ map

2 years agoReplace linear search by a faster one.
degomme [Wed, 29 Mar 2017 20:17:17 +0000 (22:17 +0200)]
Replace linear search by a faster one.

2 years agooops
degomme [Wed, 29 Mar 2017 11:08:35 +0000 (13:08 +0200)]
oops

2 years agoAdd a smpi/shared-malloc-blocksize option. This is relevant only when global shared...
degomme [Wed, 29 Mar 2017 10:51:38 +0000 (12:51 +0200)]
Add a smpi/shared-malloc-blocksize option. This is relevant only when global shared mallocs mode is used.
This allows to change the size of the fake file used (default 1MB), to potentially limit the number of mappings for large runs.

2 years agotiny optim
degomme [Wed, 29 Mar 2017 10:49:37 +0000 (12:49 +0200)]
tiny optim

2 years agoignore some more memcpys/serialization when a buffer is detected as shared
degomme [Wed, 29 Mar 2017 10:30:51 +0000 (12:30 +0200)]
ignore some more memcpys/serialization when a buffer is detected as shared

2 years agoconnect logs
degomme [Wed, 29 Mar 2017 10:30:10 +0000 (12:30 +0200)]
connect logs

2 years agonothing to see here.
degomme [Tue, 28 Mar 2017 16:59:52 +0000 (18:59 +0200)]
nothing to see here.

2 years agook, let's mess up the includes a bit more.
degomme [Tue, 28 Mar 2017 16:51:01 +0000 (18:51 +0200)]
ok, let's mess up the includes a bit more.

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Tue, 28 Mar 2017 15:37:13 +0000 (17:37 +0200)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agosome compilers don't like unused variables
Frederic Suter [Tue, 28 Mar 2017 15:36:44 +0000 (17:36 +0200)]
some compilers don't like unused variables

2 years agoSplit up smpi_bench.cpp .
degomme [Tue, 28 Mar 2017 11:55:37 +0000 (13:55 +0200)]
Split up smpi_bench.cpp .
Privatization is sent to smpi_memory.cpp (will be renamed later).
smpi_shared* functions are sent to smpi_shared.cpp.
I don't know what to do with trace_call_location yet. bench should only be timing related functions, smpi_sample* and sleep stuff.

2 years agoI think I just killed a simcall
Frederic Suter [Tue, 28 Mar 2017 15:28:40 +0000 (17:28 +0200)]
I think I just killed a simcall

2 years agoleak-- smell++ (C cast to avoid const_cast)
Frederic Suter [Tue, 28 Mar 2017 15:26:33 +0000 (17:26 +0200)]
leak-- smell++ (C cast to avoid const_cast)

2 years agopleasing sonar was a bad idea there
Frederic Suter [Tue, 28 Mar 2017 15:12:28 +0000 (17:12 +0200)]
pleasing sonar was a bad idea there

2 years agotest conversion
Frederic Suter [Tue, 28 Mar 2017 11:56:06 +0000 (13:56 +0200)]
test conversion
not completely satisfying, but storage is a bit messy

2 years agoleak plug and test AS properties
Frederic Suter [Tue, 28 Mar 2017 08:48:22 +0000 (10:48 +0200)]
leak plug and test AS properties

2 years agotypos and leak
Frederic Suter [Tue, 28 Mar 2017 08:01:56 +0000 (10:01 +0200)]
typos and leak

2 years agotracing TODO marks
Lucas M. Schnorr [Tue, 28 Mar 2017 07:39:28 +0000 (09:39 +0200)]
tracing TODO marks