Logo AND Algorithmique Numérique Distribuée

Public GIT Repository
simgrid.git
7 years agoWe never supported this one, but it was doing nothing without MPI2.2
degomme [Sat, 11 Feb 2017 17:53:42 +0000 (18:53 +0100)]
We never supported this one, but it was doing nothing without MPI2.2

7 years agoLet's pretend we support MPI 2.2.
degomme [Sat, 11 Feb 2017 17:48:46 +0000 (18:48 +0100)]
Let's pretend we support MPI 2.2.
We don't have everything, but we really have enough support to handle most of the cases.
I hope.
This activates plenty of tests that were doing nothing before.

7 years agodon't go looking for something that does not exist.
degomme [Sat, 11 Feb 2017 17:41:07 +0000 (18:41 +0100)]
don't go looking for something that does not exist.

7 years agofix reduce_scatter in place
degomme [Sat, 11 Feb 2017 17:36:45 +0000 (18:36 +0100)]
fix reduce_scatter in place

7 years agoSupport MPI_IN_PLACE in MPI_Alltoall, MPI_Alltoallv and MPI_Exscan
degomme [Sat, 11 Feb 2017 17:12:58 +0000 (18:12 +0100)]
Support MPI_IN_PLACE in MPI_Alltoall, MPI_Alltoallv and MPI_Exscan
This was introduced in MPI 2.2

7 years agowarnings
degomme [Sat, 11 Feb 2017 16:59:18 +0000 (17:59 +0100)]
warnings

7 years agoAdd MPI_Comm_group_create ... which is basically MPI_Comm_create
degomme [Sat, 11 Feb 2017 13:31:12 +0000 (14:31 +0100)]
Add MPI_Comm_group_create ... which is basically MPI_Comm_create

7 years agoUpdate mpich comm tests
degomme [Sat, 11 Feb 2017 13:30:36 +0000 (14:30 +0100)]
Update mpich comm tests

7 years agoreplace failing algo with a new version from mpich
degomme [Sat, 11 Feb 2017 01:42:02 +0000 (02:42 +0100)]
replace failing algo with a new version from mpich

7 years agoUpgrade coll mpich testlist to new mpich
degomme [Sat, 11 Feb 2017 01:41:42 +0000 (02:41 +0100)]
Upgrade coll mpich testlist to new mpich

7 years agooops
degomme [Fri, 10 Feb 2017 23:30:58 +0000 (00:30 +0100)]
oops

7 years agoUpdate mpich pt2pt tests
degomme [Fri, 10 Feb 2017 23:29:25 +0000 (00:29 +0100)]
Update mpich pt2pt tests

7 years agoUpdate mpich test files.
degomme [Fri, 10 Feb 2017 23:28:21 +0000 (00:28 +0100)]
Update mpich test files.

7 years agofix some potential memory issues with _block MPI datatypes
degomme [Fri, 10 Feb 2017 23:27:03 +0000 (00:27 +0100)]
fix some potential memory issues with _block MPI datatypes

7 years agoAdd MPI_Type_size_x call
degomme [Fri, 10 Feb 2017 23:26:30 +0000 (00:26 +0100)]
Add MPI_Type_size_x call

7 years agoeasily please sonar
Martin Quinson [Fri, 10 Feb 2017 23:21:57 +0000 (00:21 +0100)]
easily please sonar

7 years agosplit simix::MailboxImpl to its own files
Martin Quinson [Fri, 10 Feb 2017 21:50:37 +0000 (22:50 +0100)]
split simix::MailboxImpl to its own files

7 years agorename simix::Mailbox to simix::MailboxImpl for sake of consistency
Martin Quinson [Fri, 10 Feb 2017 21:25:35 +0000 (22:25 +0100)]
rename simix::Mailbox to simix::MailboxImpl for sake of consistency

7 years agorename a header file now that it's clean
Martin Quinson [Fri, 10 Feb 2017 21:20:10 +0000 (22:20 +0100)]
rename a header file now that it's clean

7 years agoobey our naming conventions
Martin Quinson [Fri, 10 Feb 2017 20:59:42 +0000 (21:59 +0100)]
obey our naming conventions

7 years agocosmetics
Martin Quinson [Fri, 10 Feb 2017 14:33:02 +0000 (15:33 +0100)]
cosmetics

7 years agoignore new lib for freebsd (used by libevent)
degomme [Fri, 10 Feb 2017 11:03:10 +0000 (12:03 +0100)]
ignore new lib for freebsd (used by libevent)

7 years agosome captured params for llambda were unused, triggering new warning in clang 5
degomme [Fri, 10 Feb 2017 10:03:12 +0000 (11:03 +0100)]
some captured params for llambda were unused, triggering new warning in clang 5

7 years agoexplicitly capture some other callback's scopes
Martin Quinson [Fri, 10 Feb 2017 01:07:01 +0000 (02:07 +0100)]
explicitly capture some other callback's scopes

7 years agogosh, sonar still chokes on this single function. Better now?
Martin Quinson [Fri, 10 Feb 2017 00:09:18 +0000 (01:09 +0100)]
gosh, sonar still chokes on this single function. Better now?

7 years agosonar complains that these cases are redundant. Fair enough
Martin Quinson [Thu, 9 Feb 2017 23:40:32 +0000 (00:40 +0100)]
sonar complains that these cases are redundant. Fair enough

7 years agoexplicitly capture the variables to please sonar
Martin Quinson [Thu, 9 Feb 2017 22:31:51 +0000 (23:31 +0100)]
explicitly capture the variables to please sonar

7 years agosome cosmetics to please sonar
Martin Quinson [Thu, 9 Feb 2017 20:49:05 +0000 (21:49 +0100)]
some cosmetics to please sonar

7 years agocosmetics
Martin Quinson [Wed, 8 Feb 2017 22:26:04 +0000 (23:26 +0100)]
cosmetics

7 years agomore info to the user.
Martin Quinson [Tue, 7 Feb 2017 15:56:43 +0000 (16:56 +0100)]
more info to the user.

7 years agobe even more informative to the user
Martin Quinson [Tue, 7 Feb 2017 15:18:43 +0000 (16:18 +0100)]
be even more informative to the user

7 years agosort output of test
degomme [Tue, 7 Feb 2017 16:40:59 +0000 (17:40 +0100)]
sort output of test

7 years agotry to fix build for gcc 7
degomme [Tue, 7 Feb 2017 16:39:50 +0000 (17:39 +0100)]
try to fix build for gcc 7

7 years agoand now fix NS3 builds, bummer
Martin Quinson [Tue, 7 Feb 2017 13:39:24 +0000 (14:39 +0100)]
and now fix NS3 builds, bummer

7 years agocosmetics
Martin Quinson [Tue, 7 Feb 2017 13:16:17 +0000 (14:16 +0100)]
cosmetics

7 years agosmpirun: gracefuly handle the case where the user forgets to give a program to execute
Martin Quinson [Tue, 7 Feb 2017 10:24:09 +0000 (11:24 +0100)]
smpirun: gracefuly handle the case where the user forgets to give a program to execute

7 years agofix lua builds
Martin Quinson [Tue, 7 Feb 2017 09:54:50 +0000 (10:54 +0100)]
fix lua builds

7 years agocosmetics
Martin Quinson [Tue, 7 Feb 2017 09:54:40 +0000 (10:54 +0100)]
cosmetics

7 years agoAdd a s4u::Link class, at least
Martin Quinson [Tue, 7 Feb 2017 01:03:01 +0000 (02:03 +0100)]
Add a s4u::Link class, at least

7 years agorename surf::Link into surf::LinkImpl to make room for s4u
Martin Quinson [Mon, 6 Feb 2017 23:16:24 +0000 (00:16 +0100)]
rename surf::Link into surf::LinkImpl to make room for s4u

7 years agorename an option and improve its doc: maxmin/concurrency-limit
Martin Quinson [Mon, 6 Feb 2017 21:02:57 +0000 (22:02 +0100)]
rename an option and improve its doc: maxmin/concurrency-limit

7 years agotest that joining a terminated process still works
Martin Quinson [Mon, 6 Feb 2017 20:47:51 +0000 (21:47 +0100)]
test that joining a terminated process still works

This patch was provided by Samuel Thibault and (fix #135).

As noted by Samuel, this feature used to fail for SimGrid 3.12 and
3.13 but seems to work since 3.14. This new test intends to make sure
that it remains this way.

Many thanks for that Sam!

7 years agoThis test requires concurrency_limit to 100 (but works with MC)
Martin Quinson [Mon, 6 Feb 2017 07:37:42 +0000 (08:37 +0100)]
This test requires concurrency_limit to 100 (but works with MC)

7 years agochange the default value of maxmin/concurrency_limit to -1 (+cosmetics)
Martin Quinson [Sun, 5 Feb 2017 21:21:11 +0000 (22:21 +0100)]
change the default value of maxmin/concurrency_limit to -1 (+cosmetics)

7 years agomake this file usable with smpi
Martin Quinson [Sun, 5 Feb 2017 21:03:31 +0000 (22:03 +0100)]
make this file usable with smpi

7 years agouse NAS/ep as an example of SMPI_SHARED_MALLOC and SMPI_SAMPLE_GLOBAL
Martin Quinson [Sat, 4 Feb 2017 20:21:06 +0000 (21:21 +0100)]
use NAS/ep as an example of SMPI_SHARED_MALLOC and SMPI_SAMPLE_GLOBAL

7 years agocosmetics
Martin Quinson [Fri, 3 Feb 2017 20:14:49 +0000 (21:14 +0100)]
cosmetics

7 years agoplug a huge memleak that I created while optimizing for speed
Martin Quinson [Thu, 2 Feb 2017 14:26:42 +0000 (15:26 +0100)]
plug a huge memleak that I created while optimizing for speed

7 years agoDon't specify RuntimeExceptions in throws clauses
Martin Quinson [Thu, 2 Feb 2017 02:31:31 +0000 (03:31 +0100)]
Don't specify RuntimeExceptions in throws clauses

7 years agoplease sonar by marking C functions as such
Martin Quinson [Thu, 2 Feb 2017 01:46:20 +0000 (02:46 +0100)]
please sonar by marking C functions as such

7 years agoplease sonar by writing lighter C++ code
Martin Quinson [Thu, 2 Feb 2017 01:39:06 +0000 (02:39 +0100)]
please sonar by writing lighter C++ code

7 years agoimprove the doc of the context factories
Martin Quinson [Thu, 2 Feb 2017 00:32:06 +0000 (01:32 +0100)]
improve the doc of the context factories

7 years agoFix MC builds, bummer
Martin Quinson [Wed, 1 Feb 2017 23:58:06 +0000 (00:58 +0100)]
Fix MC builds, bummer

7 years agotry to speed up some some traversals
Martin Quinson [Wed, 1 Feb 2017 21:39:56 +0000 (22:39 +0100)]
try to speed up some some traversals

7 years agoNew callbacks: simgrid::s4u::onSimulationEnd (right after the main simulation loop)
Martin Quinson [Wed, 1 Feb 2017 19:16:37 +0000 (20:16 +0100)]
New callbacks: simgrid::s4u::onSimulationEnd (right after the main simulation loop)

7 years agoNew callback: simgrid::s4u::onPlatformCreated (right before the simulation start)
Martin Quinson [Wed, 1 Feb 2017 19:02:02 +0000 (20:02 +0100)]
New callback: simgrid::s4u::onPlatformCreated (right before the simulation start)

7 years agoyeah, eclipse. Whatever
Martin Quinson [Wed, 1 Feb 2017 18:59:43 +0000 (19:59 +0100)]
yeah, eclipse. Whatever

7 years agoAdd some missing types for MPI operators.
degomme [Wed, 1 Feb 2017 17:05:54 +0000 (18:05 +0100)]
Add some missing types for MPI operators.
Ugly, just ugly.

7 years agooptimize this vector traversal
Martin Quinson [Wed, 1 Feb 2017 09:49:46 +0000 (10:49 +0100)]
optimize this vector traversal

7 years agotry to gather more info on this error situation
Martin Quinson [Wed, 1 Feb 2017 08:17:36 +0000 (09:17 +0100)]
try to gather more info on this error situation

7 years agoComment a particularly ... inventive code path
Martin Quinson [Wed, 1 Feb 2017 07:57:57 +0000 (08:57 +0100)]
Comment a particularly ... inventive code path

7 years agofix NS3 builds
Martin Quinson [Wed, 1 Feb 2017 07:48:43 +0000 (08:48 +0100)]
fix NS3 builds

7 years agocosmetics
Martin Quinson [Wed, 1 Feb 2017 01:33:29 +0000 (02:33 +0100)]
cosmetics

7 years agodon't malloc tons of dynars in mpi_waitany
Martin Quinson [Wed, 1 Feb 2017 01:10:12 +0000 (02:10 +0100)]
don't malloc tons of dynars in mpi_waitany

7 years agolet the users know about config changes concerning the contextes
Martin Quinson [Tue, 31 Jan 2017 23:25:04 +0000 (00:25 +0100)]
let the users know about config changes concerning the contextes

7 years agofinish the transition netcards->netpoints in the code
Martin Quinson [Tue, 31 Jan 2017 08:33:30 +0000 (09:33 +0100)]
finish the transition netcards->netpoints in the code

7 years agonetcards were renamed to netpoints recently
Martin Quinson [Mon, 30 Jan 2017 21:32:27 +0000 (22:32 +0100)]
netcards were renamed to netpoints recently

7 years agoadapt simcall.py to recent code reorg in MC
Martin Quinson [Sat, 28 Jan 2017 15:19:33 +0000 (16:19 +0100)]
adapt simcall.py to recent code reorg in MC

7 years agoour semaphore cannot be unlimited since GRAS is gone...
Martin Quinson [Sat, 28 Jan 2017 23:07:45 +0000 (00:07 +0100)]
our semaphore cannot be unlimited since GRAS is gone...

7 years agotuto improvements, and add a link to the SMPI CourseWare
Martin Quinson [Sat, 28 Jan 2017 13:57:47 +0000 (14:57 +0100)]
tuto improvements, and add a link to the SMPI CourseWare

7 years agofurther cleanup the files provided with the tuto
Martin Quinson [Fri, 27 Jan 2017 21:34:50 +0000 (22:34 +0100)]
further cleanup the files provided with the tuto

And update the tuto accordingly

7 years agojava: kill obscure NativeException
Martin Quinson [Fri, 27 Jan 2017 17:44:29 +0000 (18:44 +0100)]
java: kill obscure NativeException

7 years agodocument the new MSG example I recently added
Martin Quinson [Fri, 27 Jan 2017 16:41:24 +0000 (17:41 +0100)]
document the new MSG example I recently added

7 years agoimprove a FAQ entry
Martin Quinson [Fri, 27 Jan 2017 07:04:37 +0000 (08:04 +0100)]
improve a FAQ entry

7 years agosonar: these checks are just wrong in our context
Martin Quinson [Thu, 26 Jan 2017 23:40:09 +0000 (00:40 +0100)]
sonar: these checks are just wrong in our context

7 years agoMake MSG_process_yield visible from java, and test it
Martin Quinson [Thu, 26 Jan 2017 22:29:42 +0000 (23:29 +0100)]
Make MSG_process_yield visible from java, and test it

7 years agofix make distcheck
Martin Quinson [Thu, 26 Jan 2017 21:42:45 +0000 (22:42 +0100)]
fix make distcheck

7 years agoNew example for MSG_process_yield()
Martin Quinson [Thu, 26 Jan 2017 21:41:36 +0000 (22:41 +0100)]
New example for MSG_process_yield()

7 years agonew function MSG_process_yield()
Martin Quinson [Thu, 26 Jan 2017 21:23:31 +0000 (22:23 +0100)]
new function MSG_process_yield()

7 years agoDocument last change
Martin Quinson [Thu, 26 Jan 2017 21:22:36 +0000 (22:22 +0100)]
Document last change

7 years agotell eclipse where to find the arch-dependent java files (on my machine)
Martin Quinson [Thu, 26 Jan 2017 21:21:00 +0000 (22:21 +0100)]
tell eclipse where to find the arch-dependent java files (on my machine)

7 years agoMake these example files compile
Martin Quinson [Thu, 26 Jan 2017 21:19:44 +0000 (22:19 +0100)]
Make these example files compile

7 years agoallow java actors to kill themselves
Martin Quinson [Thu, 26 Jan 2017 17:23:25 +0000 (18:23 +0100)]
allow java actors to kill themselves

7 years agocheck that suiciding a process actually works in C
Martin Quinson [Thu, 26 Jan 2017 15:14:54 +0000 (16:14 +0100)]
check that suiciding a process actually works in C

7 years agoI'm using openjdk-8 nowadays
Martin Quinson [Thu, 26 Jan 2017 15:05:03 +0000 (16:05 +0100)]
I'm using openjdk-8 nowadays

This should reduce the amount of errors I get in eclipse on the JNI
files.

7 years agofix a memleak in Java VM when the used did not manually destroy VM
Martin Quinson [Thu, 26 Jan 2017 11:14:51 +0000 (12:14 +0100)]
fix a memleak in Java VM when the used did not manually destroy VM

The C world was cleaned on a vm.destroy() and not when the GC
finalizes the Java world.

- Pb1: any java call to a destroy()ed vm could segfault
- Pb2: C world memleaked when forgetting destroy()

So now, the VM is not really destroyed when calling destroy(), but
only when all references to the object are gone. It tastes much more
Java to me that the previous version where our users had to to do the
memory management manually.

Arguably, this differs a bit more from libvirt, but I think I can live
with it.

7 years agoMC: cosmetics
Martin Quinson [Thu, 26 Jan 2017 10:51:31 +0000 (11:51 +0100)]
MC: cosmetics

7 years agoMC: comment, rename stuff, don't change anything profund
Martin Quinson [Wed, 25 Jan 2017 14:14:06 +0000 (15:14 +0100)]
MC: comment, rename stuff, don't change anything profund

7 years agoreduce the amount of mallocs and use space on the stack
Martin Quinson [Wed, 25 Jan 2017 13:56:22 +0000 (14:56 +0100)]
reduce the amount of mallocs and use space on the stack

7 years agoone algo was missing from the doc
degomme [Wed, 25 Jan 2017 13:29:44 +0000 (14:29 +0100)]
one algo was missing from the doc

7 years agoMerge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Mon, 23 Jan 2017 21:52:33 +0000 (22:52 +0100)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid

7 years agoRename NetCards to NetPoints
Martin Quinson [Mon, 23 Jan 2017 21:51:47 +0000 (22:51 +0100)]
Rename NetCards to NetPoints

7 years agodarwin: (preliminary) memory mappings + smpi variable privatization
Matthieu Volat [Mon, 23 Jan 2017 18:30:30 +0000 (19:30 +0100)]
darwin: (preliminary) memory mappings + smpi variable privatization

7 years agoFix a typo in smpirun.
Matthieu Volat [Mon, 23 Jan 2017 18:28:55 +0000 (19:28 +0100)]
Fix a typo in smpirun.

7 years agoremove weird blank characters
Augustin Degomme [Mon, 23 Jan 2017 19:57:09 +0000 (20:57 +0100)]
remove weird blank characters

7 years agocosmetics
Martin Quinson [Sun, 22 Jan 2017 21:27:33 +0000 (22:27 +0100)]
cosmetics

7 years agotry to make SafetyChecker a bit easier to read
Martin Quinson [Sat, 21 Jan 2017 08:52:16 +0000 (09:52 +0100)]
try to make SafetyChecker a bit easier to read

Mostly adding comments and renaming symbols, but not changing to the
logic (yet)

7 years agoplug a memleak
Martin Quinson [Fri, 20 Jan 2017 10:53:21 +0000 (11:53 +0100)]
plug a memleak

This one is vicious and I should cleanup the VM and Host interactions.
one day.

7 years agoinstall the tools from the packages when possible; forget about Paje
Martin Quinson [Fri, 20 Jan 2017 09:34:45 +0000 (10:34 +0100)]
install the tools from the packages when possible; forget about Paje

7 years agorevalidate a tesh file after a fix due to please sonar
Martin Quinson [Fri, 20 Jan 2017 00:03:18 +0000 (01:03 +0100)]
revalidate a tesh file after a fix due to please sonar

+ remove yet another annoying sonar warning (on the exact line that I
  fixed in the previous run: the second task was never run in Java)