Logo AND Algorithmique Numérique Distribuée

Public GIT Repository
simgrid.git
2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Tue, 28 Feb 2017 09:37:58 +0000 (10:37 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agosonar: ignore coverage on XML files (and some others)
Martin Quinson [Tue, 28 Feb 2017 09:32:21 +0000 (10:32 +0100)]
sonar: ignore coverage on XML files (and some others)

2 years agoundynarify Torus
Frederic Suter [Tue, 28 Feb 2017 09:31:28 +0000 (10:31 +0100)]
undynarify Torus

2 years agoanother try at having coverage info on travis, for sonar
Martin Quinson [Tue, 28 Feb 2017 02:10:10 +0000 (03:10 +0100)]
another try at having coverage info on travis, for sonar

2 years agoTry to get coverage files on travis
Martin Quinson [Tue, 28 Feb 2017 01:43:52 +0000 (02:43 +0100)]
Try to get coverage files on travis

2 years agoplease sonar
Martin Quinson [Mon, 27 Feb 2017 20:17:56 +0000 (21:17 +0100)]
please sonar

2 years agosimplify the way tesh deals with sorted diffs
Martin Quinson [Sun, 26 Feb 2017 00:11:19 +0000 (01:11 +0100)]
simplify the way tesh deals with sorted diffs

2 years agorename some C type for consistency
Martin Quinson [Sat, 25 Feb 2017 23:14:29 +0000 (00:14 +0100)]
rename some C type for consistency

2 years agosonar: allow 'Pointers should not be cast to integral types' everywhere in source
Martin Quinson [Sat, 25 Feb 2017 15:12:49 +0000 (16:12 +0100)]
sonar: allow 'Pointers should not be cast to integral types' everywhere in source

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Mon, 27 Feb 2017 16:48:18 +0000 (17:48 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agoless free in Dragonfly and Torus
Frederic Suter [Mon, 27 Feb 2017 16:18:42 +0000 (17:18 +0100)]
less free in Dragonfly and Torus

2 years agocheck that max is not -1 befor using it as index
Frederic Suter [Mon, 27 Feb 2017 15:30:35 +0000 (16:30 +0100)]
check that max is not -1 befor using it as index

2 years agoDon't do the loop ourselves, let jenkins handle this part with Configuration Matrix
degomme [Mon, 27 Feb 2017 15:01:08 +0000 (16:01 +0100)]
Don't do the loop ourselves, let jenkins handle this part with Configuration Matrix
This should improve readability of the output

2 years agoAdd more infos to Flags output, and write them to a file.
degomme [Mon, 27 Feb 2017 13:59:29 +0000 (14:59 +0100)]
Add more infos to Flags output, and write them to a file.
This will be used to set the description of the build

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Mon, 27 Feb 2017 13:07:57 +0000 (14:07 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agoplease sonar
Frederic Suter [Mon, 27 Feb 2017 13:07:46 +0000 (14:07 +0100)]
please sonar

2 years agoname parameters in prototypes
Frederic Suter [Mon, 27 Feb 2017 12:48:09 +0000 (13:48 +0100)]
name parameters in prototypes

2 years agono == for doubles
Frederic Suter [Mon, 27 Feb 2017 10:04:53 +0000 (11:04 +0100)]
no == for doubles
TODO or FIXME, not both => TODO

2 years agoCleanup Sanitizer jenkins script to allow separate launch
degomme [Mon, 27 Feb 2017 09:30:27 +0000 (10:30 +0100)]
Cleanup Sanitizer jenkins script to allow separate launch

2 years agochecking if a unsigned int is >0 is a bad idea
Frederic Suter [Mon, 27 Feb 2017 09:06:10 +0000 (10:06 +0100)]
checking if a unsigned int is >0 is a bad idea
1-2 = 4294967295 is positive !! check 1 > 2 instead
revalidate tesh, previous one might have been silently false
should please sanitizer

2 years agodo not try to remove an out-of-index piece
Frederic Suter [Mon, 27 Feb 2017 08:47:50 +0000 (09:47 +0100)]
do not try to remove an out-of-index piece
should please sanitizer

2 years agoplug a bunch of memleaks
Martin Quinson [Sat, 25 Feb 2017 14:55:26 +0000 (15:55 +0100)]
plug a bunch of memleaks

2 years agoplug a memleak
Martin Quinson [Sat, 25 Feb 2017 10:06:40 +0000 (11:06 +0100)]
plug a memleak

2 years agouseless cleanups to the Java bindings
Martin Quinson [Sat, 25 Feb 2017 09:53:35 +0000 (10:53 +0100)]
useless cleanups to the Java bindings

2 years agouseless cleanups to the Java bindings
Martin Quinson [Sat, 25 Feb 2017 07:47:15 +0000 (08:47 +0100)]
useless cleanups to the Java bindings

2 years agosonar: JNI mandates extra parameters (+cosmetics)
Martin Quinson [Sat, 25 Feb 2017 07:41:48 +0000 (08:41 +0100)]
sonar: JNI mandates extra parameters (+cosmetics)

2 years agoplease sonar
Martin Quinson [Fri, 24 Feb 2017 22:02:13 +0000 (23:02 +0100)]
please sonar

- Ignore RngStream, which we did not author
- Mark the whole java bindings as using the C linkage, because the JVM
  wants it so. So we cannot move these symbols in a namespace.
- Mark the whole XML cruft as using the C linkage, because FlexML
  wants it so.
- Mark the whole MSG API as using the C linkage, because it does.

2 years agofinish the transition from C structures to C++ objects
Martin Quinson [Fri, 24 Feb 2017 15:54:26 +0000 (16:54 +0100)]
finish the transition from C structures to C++ objects

Actually, we should stop using sg_platf_new_link and use
NetZone::createLink() but this later function is not ready yet.

And I don't want to revert the work of Fred so I'd better fix it.

2 years agowoops
Frederic Suter [Fri, 24 Feb 2017 14:21:18 +0000 (15:21 +0100)]
woops

2 years agoWIP stop using const char* in C++ layers
Frederic Suter [Fri, 24 Feb 2017 13:47:51 +0000 (14:47 +0100)]
WIP stop using const char* in C++ layers
(and ugly bprintf, const_cast and free that sonar doesn't like)

2 years agocosmetics in ChangeLog
Martin Quinson [Fri, 24 Feb 2017 10:55:21 +0000 (11:55 +0100)]
cosmetics in ChangeLog

2 years agoplease sonar
Martin Quinson [Fri, 24 Feb 2017 10:43:14 +0000 (11:43 +0100)]
please sonar

- Remove redundent declaration
- mark functions as extern "C" in their declaration, not only prototypes
- also stop having a C function and a C++ function sharing the same
  name with differing prototypes, even if sonar did not complain.

2 years agoplease sonar by moving includes at the top
Martin Quinson [Fri, 24 Feb 2017 08:42:01 +0000 (09:42 +0100)]
please sonar by moving includes at the top

2 years agoUpdate changelog
degomme [Fri, 24 Feb 2017 10:14:54 +0000 (11:14 +0100)]
Update changelog

2 years agoplug some other memleaks
Martin Quinson [Fri, 24 Feb 2017 07:09:05 +0000 (08:09 +0100)]
plug some other memleaks

2 years agoplug a bunch of memleaks
Martin Quinson [Fri, 24 Feb 2017 03:19:20 +0000 (04:19 +0100)]
plug a bunch of memleaks

2 years agocosmetics while digging for a memleak
Martin Quinson [Fri, 24 Feb 2017 02:38:21 +0000 (03:38 +0100)]
cosmetics while digging for a memleak

2 years agoplug a memleak. I should beter kill that XBT container instead
Martin Quinson [Fri, 24 Feb 2017 02:24:58 +0000 (03:24 +0100)]
plug a memleak. I should beter kill that XBT container instead

2 years agoplug an obvious memleak
Martin Quinson [Fri, 24 Feb 2017 02:03:20 +0000 (03:03 +0100)]
plug an obvious memleak

2 years agotesh: better handling of very large sorted output
Martin Quinson [Thu, 23 Feb 2017 23:50:16 +0000 (00:50 +0100)]
tesh: better handling of very large sorted output

- truncate the diff at 250 lines (in any cases)
- show the unsorted output if the diff was asked to be sorted

2 years agofix a suppurious failure on sanitizers
Martin Quinson [Thu, 23 Feb 2017 11:27:10 +0000 (12:27 +0100)]
fix a suppurious failure on sanitizers

2 years agosave some useless calls when there is no request to handle
Martin Quinson [Thu, 23 Feb 2017 11:21:29 +0000 (12:21 +0100)]
save some useless calls when there is no request to handle

2 years agocosmetics for sonar
Martin Quinson [Thu, 23 Feb 2017 11:16:26 +0000 (12:16 +0100)]
cosmetics for sonar

2 years agoAvoid unsafe things
degomme [Thu, 23 Feb 2017 11:03:39 +0000 (12:03 +0100)]
Avoid unsafe things

2 years agosave some more lines
degomme [Thu, 23 Feb 2017 09:34:06 +0000 (10:34 +0100)]
save some more lines

2 years agosmells and bugs
Frederic Suter [Thu, 23 Feb 2017 10:52:26 +0000 (11:52 +0100)]
smells and bugs

2 years agoAlso hash the first char, and please sonar by splitting increment to its own expr
Martin Quinson [Thu, 23 Feb 2017 10:40:41 +0000 (11:40 +0100)]
Also hash the first char, and please sonar by splitting increment to its own expr

2 years agochange a 'use after free' into a leak. For now
Martin Quinson [Thu, 23 Feb 2017 09:56:52 +0000 (10:56 +0100)]
change a 'use after free' into a leak. For now

2 years agocosmetics to please sonar
Martin Quinson [Thu, 23 Feb 2017 08:17:16 +0000 (09:17 +0100)]
cosmetics to please sonar

2 years agotry to get some coverage info for sonar on travis
Martin Quinson [Thu, 23 Feb 2017 00:00:23 +0000 (01:00 +0100)]
try to get some coverage info for sonar on travis

2 years agooops
Augustin Degomme [Thu, 23 Feb 2017 00:20:55 +0000 (01:20 +0100)]
oops

2 years agoAt this point of the program simgrid values are not to be trusted as we had a segfault.
degomme [Wed, 22 Feb 2017 19:05:47 +0000 (20:05 +0100)]
At this point of the program simgrid values are not to be trusted as we had a segfault.
This broke some tests, sometimes, as this was wrongly evaluated to true.

2 years agosomehow off_t is not enough there, and ends up being negative on some 32 bits systems
degomme [Wed, 22 Feb 2017 17:20:12 +0000 (18:20 +0100)]
somehow off_t is not enough there, and ends up being negative on some 32 bits systems

2 years agoThis should be an off_t (breaks some 32 bits systems)
degomme [Wed, 22 Feb 2017 14:24:38 +0000 (15:24 +0100)]
This should be an off_t (breaks some 32 bits systems)

2 years agoSpeedup mpich tests by trying less sizes.
degomme [Wed, 22 Feb 2017 14:05:44 +0000 (15:05 +0100)]
Speedup mpich tests by trying less sizes.

2 years agoTry to get rid of some old and weird way of doing things for fortran.
degomme [Wed, 22 Feb 2017 13:19:49 +0000 (14:19 +0100)]
Try to get rid of some old and weird way of doing things for fortran.
This was not working properly on osx...

2 years agoMacOS mappings were 1 byte off, preventing extension of them after one section
degomme [Wed, 22 Feb 2017 08:48:00 +0000 (09:48 +0100)]
MacOS mappings were 1 byte off, preventing extension of them after one section

2 years agocomment to explain what's going on
Martin Quinson [Wed, 22 Feb 2017 16:38:37 +0000 (17:38 +0100)]
comment to explain what's going on

2 years agodoh, I broke SMPI
Martin Quinson [Wed, 22 Feb 2017 15:26:30 +0000 (16:26 +0100)]
doh, I broke SMPI

2 years agoput MsgActorExt in a namespace
Martin Quinson [Sat, 18 Feb 2017 01:34:18 +0000 (02:34 +0100)]
put MsgActorExt in a namespace

2 years agoProtect access to remote process list of rma requests, to avoid race conditions..
degomme [Tue, 21 Feb 2017 22:18:13 +0000 (23:18 +0100)]
Protect access to remote process list of rma requests, to avoid race conditions..

2 years agodocument changes related to issue #138
Frederic Suter [Tue, 21 Feb 2017 08:14:16 +0000 (09:14 +0100)]
document changes related to issue #138

2 years agosonar: reinterpret_cast is OK in jMSG
Martin Quinson [Fri, 17 Feb 2017 21:40:21 +0000 (22:40 +0100)]
sonar: reinterpret_cast is OK in jMSG

2 years agoRevert "Let our examples compile with java 1.6, or almost"
Martin Quinson [Fri, 17 Feb 2017 21:35:34 +0000 (22:35 +0100)]
Revert "Let our examples compile with java 1.6, or almost"

Sonar is actually right to not like that brain overload. Plus, not all
examples were compiling with prehistorical Java anyway.

This reverts commit 848ff1f03de7a300b6314015e5fb4b28348490c4.

2 years agoplease sonar
Martin Quinson [Fri, 17 Feb 2017 21:33:21 +0000 (22:33 +0100)]
please sonar

- don't use void as parameter
- prefer if to ternary conditional

2 years agomark s4u::Link as public for windows (+cosmetics)
Martin Quinson [Fri, 17 Feb 2017 21:23:39 +0000 (22:23 +0100)]
mark s4u::Link as public for windows (+cosmetics)

2 years agostatic inline are the modern macros
Martin Quinson [Fri, 17 Feb 2017 20:12:05 +0000 (21:12 +0100)]
static inline are the modern macros

2 years agokill some other occurences of the fifo
Martin Quinson [Fri, 17 Feb 2017 16:50:49 +0000 (17:50 +0100)]
kill some other occurences of the fifo

2 years agouse the C++ API from the flatifier (as an example of that API)
Martin Quinson [Fri, 17 Feb 2017 16:41:05 +0000 (17:41 +0100)]
use the C++ API from the flatifier (as an example of that API)

2 years agodocument my last changes
Martin Quinson [Thu, 16 Feb 2017 22:26:20 +0000 (23:26 +0100)]
document my last changes

2 years agoforgot one include
Frederic Suter [Fri, 17 Feb 2017 15:45:24 +0000 (16:45 +0100)]
forgot one include

2 years agoxbt_fifo are no longer used, remove the code
Frederic Suter [Fri, 17 Feb 2017 15:16:08 +0000 (16:16 +0100)]
xbt_fifo are no longer used, remove the code

2 years agoFriday smell chase
Frederic Suter [Fri, 17 Feb 2017 14:08:24 +0000 (15:08 +0100)]
Friday smell chase

2 years agodo not test equality on double
Frederic Suter [Fri, 17 Feb 2017 09:12:53 +0000 (10:12 +0100)]
do not test equality on double

modify (and rename macro) to return -1 if the cost array is not defined
then we can test if the cost is 0 (nothing to do) or -1 (doesn't even
exist) and thus use <= instead of ==

2 years agodead store-- + rewrite of hash function
Frederic Suter [Fri, 17 Feb 2017 09:05:57 +0000 (10:05 +0100)]
dead store-- + rewrite of hash function

2 years agoplease sonar there
Frederic Suter [Fri, 17 Feb 2017 08:50:37 +0000 (09:50 +0100)]
please sonar there
c'mon we all know how to access elements in 2D arrays, no need for a
macro;)

2 years agotypo --
Augustin Degomme [Thu, 16 Feb 2017 23:21:10 +0000 (00:21 +0100)]
typo --

2 years agowarning --, and ignore map_populate when not implemented (linux only, but just adviso...
Augustin Degomme [Thu, 16 Feb 2017 23:15:52 +0000 (00:15 +0100)]
warning --, and ignore map_populate when not implemented (linux only, but just advisory, so let's follow https://gnu.org/software/hurd/glibc/mmap.html)

2 years agoImplement a new algorithm for SMPI_SHARED_MALLOC: global
Martin Quinson [Thu, 16 Feb 2017 21:42:29 +0000 (22:42 +0100)]
Implement a new algorithm for SMPI_SHARED_MALLOC: global

This maps any new block onto a single file in memory.
The results will be awfully disfigured but that is damn efficient in
memory.

2 years agosanitize the option smpi/shared-malloc, and improve its doc
Martin Quinson [Thu, 16 Feb 2017 16:11:52 +0000 (17:11 +0100)]
sanitize the option smpi/shared-malloc, and improve its doc

2 years agoxbt_fifo -> std::list
Frederic Suter [Thu, 16 Feb 2017 14:22:40 +0000 (15:22 +0100)]
xbt_fifo -> std::list

2 years agoempty statements...
Frederic Suter [Thu, 16 Feb 2017 13:03:00 +0000 (14:03 +0100)]
empty statements...

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Thu, 16 Feb 2017 11:34:57 +0000 (12:34 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agowoops
Frederic Suter [Thu, 16 Feb 2017 11:34:48 +0000 (12:34 +0100)]
woops

2 years agoAvoid doing things when there is nothing to do.
degomme [Thu, 16 Feb 2017 11:08:44 +0000 (12:08 +0100)]
Avoid doing things when there is nothing to do.

2 years agoAvoid potential invalid reads
degomme [Thu, 16 Feb 2017 10:58:13 +0000 (11:58 +0100)]
Avoid potential invalid reads

2 years agoHandle MPI_UNSIGNED_LONG_LONG in this algorithm
degomme [Thu, 16 Feb 2017 09:33:30 +0000 (10:33 +0100)]
Handle MPI_UNSIGNED_LONG_LONG in this algorithm

2 years agoAvoid overflowing in the MPI_Prod version by using larger datatype
degomme [Thu, 16 Feb 2017 09:33:08 +0000 (10:33 +0100)]
Avoid overflowing in the MPI_Prod version by using larger datatype

2 years agosign the parameters
Frederic Suter [Thu, 16 Feb 2017 10:49:36 +0000 (11:49 +0100)]
sign the parameters

2 years agoMerge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Frederic Suter [Thu, 16 Feb 2017 10:36:21 +0000 (11:36 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid

2 years agofifo2dynar
Frederic Suter [Thu, 16 Feb 2017 10:35:43 +0000 (11:35 +0100)]
fifo2dynar

2 years agoSometimes (most notably in StarPU-MPI case) the issuer of the comm is neither its...
Luka Stanisic [Wed, 8 Feb 2017 17:24:33 +0000 (18:24 +0100)]
Sometimes (most notably in StarPU-MPI case) the issuer of the comm is neither its src nor dst. I dont see how this can jeopardies the correct execution of SimGrid in other use cases, but I may be missing something.

This commit is indispensable for the correct termination of StarPU-MPI+SimGrid execution.

2 years agoa bit of cleaning
Frederic Suter [Thu, 16 Feb 2017 09:55:02 +0000 (10:55 +0100)]
a bit of cleaning

2 years agocosmetics
Frederic Suter [Thu, 16 Feb 2017 09:54:11 +0000 (10:54 +0100)]
cosmetics

2 years agoyet another bitwise pickyness + hey that's a foreach
Frederic Suter [Thu, 16 Feb 2017 08:59:48 +0000 (09:59 +0100)]
yet another bitwise pickyness + hey that's a foreach

2 years agomatch new with delete, not with free
Martin Quinson [Thu, 16 Feb 2017 08:55:44 +0000 (09:55 +0100)]
match new with delete, not with free

2 years agofix newly introduced bugs
Frederic Suter [Thu, 16 Feb 2017 08:14:14 +0000 (09:14 +0100)]
fix newly introduced bugs

2 years agosimplify code by using default values for the fields
Martin Quinson [Thu, 16 Feb 2017 01:53:45 +0000 (02:53 +0100)]
simplify code by using default values for the fields

2 years agoreindent
Martin Quinson [Thu, 16 Feb 2017 01:47:08 +0000 (02:47 +0100)]
reindent

2 years agofurther untangle the msg_process creation by using a default value for auto_restart
Martin Quinson [Thu, 16 Feb 2017 01:42:18 +0000 (02:42 +0100)]
further untangle the msg_process creation by using a default value for auto_restart