Christian Heinrich [Tue, 21 Jul 2015 10:20:22 +0000 (12:20 +0200)]
[SURF] Callbacks now use the INITIALIZER macros
- Callbacks such as s_sg_platf_link_cbarg_t were not at all
correctly initialized before; this was fixed!
- Several initializers used valid C99 initializer syntax; however,
this is not supported by C++. Hence, these initializers were changed
to simple lists. See also
https://gcc.gnu.org/onlinedocs/gcc/Designated-Inits.html
for more details on this issue.
Martin Quinson [Tue, 21 Jul 2015 10:04:53 +0000 (12:04 +0200)]
cosmetics in the check of library version vs. header version
Martin Quinson [Mon, 20 Jul 2015 23:27:19 +0000 (01:27 +0200)]
Remove a bunch of stuff from Host that was delegated to cpu anyway
- getSpeed()
- getAvailableSpeed()
- getCore()
- getCurrentPowerPeak();
- getPowerPeakAt(pstate_index);
- getNbPstates();
- {set,get}Pstate()
Martin Quinson [Mon, 20 Jul 2015 23:17:16 +0000 (01:17 +0200)]
further doxygen cleanups
Martin Quinson [Mon, 20 Jul 2015 23:06:25 +0000 (01:06 +0200)]
cosmetics and documentation around VM parameters
Martin Quinson [Mon, 20 Jul 2015 22:26:18 +0000 (00:26 +0200)]
kill dead prototypes
Martin Quinson [Mon, 20 Jul 2015 21:36:51 +0000 (23:36 +0200)]
more rewrite of the doxygen
Martin Quinson [Mon, 20 Jul 2015 21:13:01 +0000 (23:13 +0200)]
less void*, more sg_host_t
Martin Quinson [Mon, 20 Jul 2015 20:39:48 +0000 (22:39 +0200)]
remove Host::communicate() (use network instead)
This is my latest attempt to untangle things in surf: empty the Host
class and move everything to network/cpu/storage instead.
Martin Quinson [Mon, 20 Jul 2015 17:01:11 +0000 (19:01 +0200)]
code factorization
Martin Quinson [Mon, 20 Jul 2015 16:56:50 +0000 (18:56 +0200)]
kill an unused field (+cosmetics)
Martin Quinson [Mon, 20 Jul 2015 16:42:44 +0000 (18:42 +0200)]
actually kill models' names
Martin Quinson [Mon, 20 Jul 2015 16:14:21 +0000 (18:14 +0200)]
Kill models getName() call sites.
- Debug messages now depend on typeid(...).name(). Beware, the output
may be compiler dependent.
- It was mainly used to have a specific handling of NS3, which method
shareRessources() is not idempotent (ie, it actually moves the model
in the future with no possibility of rolling it back afterward).
So NS3->shareRessource() needs to be run after all other
shareRessources, and before the updateState()s.
Add a shareResourcesIsIdempotent() method to all models to detect
whether we need to activate that strange behavior without having to
rely on the model name.
- Removing models namesmakes Model pure virtual, possibly allowing
diamon-shaped inheritance in L07, where we need only one model for
CpuModel and NetworkModel.
Martin Quinson [Mon, 20 Jul 2015 15:17:48 +0000 (17:17 +0200)]
typo in comments
Martin Quinson [Mon, 20 Jul 2015 15:10:03 +0000 (17:10 +0200)]
kill an unused function
Martin Quinson [Mon, 20 Jul 2015 15:03:22 +0000 (17:03 +0200)]
kill the sorry remainings of a parallelization of SURF
Martin Quinson [Mon, 20 Jul 2015 14:51:50 +0000 (16:51 +0200)]
inline two static functions
Martin Quinson [Mon, 20 Jul 2015 13:51:54 +0000 (15:51 +0200)]
initialize fields at declaration
Martin Quinson [Mon, 20 Jul 2015 13:41:28 +0000 (15:41 +0200)]
Remove one of the many pimple: HostModel::p_cpuModel
Gosh this code is hairly. This one came from the VM war, apparently.
Martin Quinson [Mon, 20 Jul 2015 13:07:45 +0000 (15:07 +0200)]
kill unused function
Martin Quinson [Mon, 20 Jul 2015 13:05:27 +0000 (15:05 +0200)]
make a pimple smaller
Martin Quinson [Mon, 20 Jul 2015 12:27:45 +0000 (14:27 +0200)]
reduce a bit the amount of void* (in L07)
Martin Quinson [Mon, 20 Jul 2015 11:18:34 +0000 (13:18 +0200)]
rename HostL07Action to L07Action
L07 actions are indeed hybrid actions that are both CPU and network.
Martin Quinson [Mon, 20 Jul 2015 10:57:32 +0000 (12:57 +0200)]
code simplification
Also, add some comments
Martin Quinson [Mon, 20 Jul 2015 10:37:23 +0000 (12:37 +0200)]
kill dead code: Action::recycle never got implemented
Martin Quinson [Mon, 20 Jul 2015 10:32:19 +0000 (12:32 +0200)]
more cosmetics in doxygen comments
Martin Quinson [Mon, 20 Jul 2015 10:10:05 +0000 (12:10 +0200)]
cosmetics in the info message of L07 model starting up
Martin Quinson [Mon, 20 Jul 2015 09:53:13 +0000 (11:53 +0200)]
cosmetics in doxygen comments
And kill dead code and other useless cosmetics
Martin Quinson [Mon, 20 Jul 2015 08:31:24 +0000 (10:31 +0200)]
Get ride of ???Ptr types in C++: make pointers explicit
Making pointers explicit is not what we did in C so far, but what we
did was a bad idea. It's easier to code when you know what is a
pointer and what is not.
Also:
- Convert a bunch of C-typecast into static_cast
- Kill some dead or dupplicated code
Martin Quinson [Sun, 19 Jul 2015 16:14:46 +0000 (18:14 +0200)]
inline a static function called only once
Martin Quinson [Sun, 19 Jul 2015 15:33:34 +0000 (17:33 +0200)]
sort my TODO list :-(
plus cosmetics
Martin Quinson [Sun, 19 Jul 2015 15:20:33 +0000 (17:20 +0200)]
Kill link_lib and have C++ handle the hashmap of all links
I did not manage to get SD_application_reinit() working, so this
function is currently disabled. I think that it should be removed
anyway.
Martin Quinson [Sun, 19 Jul 2015 06:49:55 +0000 (08:49 +0200)]
kill some remainings of sd_link_t
Martin Quinson [Sun, 19 Jul 2015 00:14:49 +0000 (02:14 +0200)]
start refreshing and fixing the GTNetS interface (WIP)
Martin Quinson [Sat, 18 Jul 2015 21:35:18 +0000 (23:35 +0200)]
Overhaul of NS3 support in SimGrid
- NS3 is now found if installed from the default package (eg, Debian)
- If enable_ns3 and not found, cmake will fail
- 3 log channels were a bit too much
- Don't use NS3 log channels as it fails when running in Java somehow
- Drop support for ancient version of NS3 (3.10)
- Documentation refresh/rewrite
- Revalidate the tesh files (rounding errors on timings)
Martin Quinson [Sat, 18 Jul 2015 20:50:57 +0000 (22:50 +0200)]
fix doxygen
Martin Quinson [Sat, 18 Jul 2015 17:19:25 +0000 (19:19 +0200)]
sanitize C API of sg_link
Martin Quinson [Sat, 18 Jul 2015 17:13:29 +0000 (19:13 +0200)]
there is no need for a SimDag's link type
Martin Quinson [Sat, 18 Jul 2015 16:33:56 +0000 (18:33 +0200)]
kill an unused field
Martin Quinson [Sat, 18 Jul 2015 16:31:51 +0000 (18:31 +0200)]
kill LinkPtr type and make Link* explicit
Martin Quinson [Sat, 18 Jul 2015 16:13:30 +0000 (18:13 +0200)]
give Link a proper public interface
Martin Quinson [Sat, 18 Jul 2015 14:17:26 +0000 (16:17 +0200)]
Rename NetworkLink type to Link
Yup, that's an useless cosmetics.
Martin Quinson [Sat, 18 Jul 2015 13:56:02 +0000 (15:56 +0200)]
cosmetics in the comments
Martin Quinson [Sat, 18 Jul 2015 13:46:02 +0000 (15:46 +0200)]
kill unused type definitions
Martin Quinson [Sat, 18 Jul 2015 13:25:34 +0000 (15:25 +0200)]
make SD_HOST_LEVEL private
Martin Quinson [Sat, 18 Jul 2015 12:26:59 +0000 (14:26 +0200)]
New function: sg_hosts_as_dynar
This was implemented several times, in each layer
Martin Quinson [Sat, 18 Jul 2015 10:52:10 +0000 (12:52 +0200)]
cosmetics: make semantic more clear
Martin Quinson [Sat, 18 Jul 2015 10:42:26 +0000 (12:42 +0200)]
[SGpp] make ROUTING_HOST_LEVEL private. That was painful
Martin Quinson [Sat, 18 Jul 2015 10:37:42 +0000 (12:37 +0200)]
whatever, eclipse
Martin Quinson [Wed, 15 Jul 2015 08:30:02 +0000 (10:30 +0200)]
Use MSG_host_by_name() instead of MSG_get_host_by_name()
Both work, but the first is more coherent with the rest of the
interface.
Martin Quinson [Wed, 15 Jul 2015 08:05:38 +0000 (10:05 +0200)]
kill simcall_host_get_by_name: nobody uses it anyway
Martin Quinson [Wed, 15 Jul 2015 07:59:34 +0000 (09:59 +0200)]
[SGpp] make SURF_CPU_LEVEL private
Martin Quinson [Wed, 15 Jul 2015 00:59:47 +0000 (02:59 +0200)]
kill unimplemented (braindead) prototype
Martin Quinson [Wed, 15 Jul 2015 00:57:44 +0000 (02:57 +0200)]
[SGpp] make MSG_HOST_LEVEL private
Martin Quinson [Tue, 14 Jul 2015 23:58:35 +0000 (01:58 +0200)]
damn, missing files
Martin Quinson [Tue, 14 Jul 2015 23:58:09 +0000 (01:58 +0200)]
sanitize a bit the prototype of SIMIX_host_create()
Martin Quinson [Tue, 14 Jul 2015 23:46:11 +0000 (01:46 +0200)]
[SGpp] make SIMIX_HOST_LEVEL private
Martin Quinson [Tue, 14 Jul 2015 21:25:33 +0000 (23:25 +0200)]
Give simgrid's host its own header file and datatype
This is one step further toward SGpp, the C++ version of SimGrid.
Christian Heinrich [Fri, 17 Jul 2015 17:10:57 +0000 (19:10 +0200)]
[SURF] Cosmetics - indented assignments in surf_routing.cpp.
Christian Heinrich [Fri, 17 Jul 2015 17:10:21 +0000 (19:10 +0200)]
[Doc] Added Cluster description. Fixes #18
Gabriel Corona [Fri, 17 Jul 2015 09:36:15 +0000 (11:36 +0200)]
[mc] Remove content addressable page store
- It did not prove useful compared fo the default option.
- It is implemented as a compile-time option instead of a runtime one.
- Let's remove compile time options!
- It makes the code more complex.
Gabriel Corona [Fri, 17 Jul 2015 09:30:40 +0000 (11:30 +0200)]
[mc] Cleanup mc_hash
Gabriel Corona [Fri, 17 Jul 2015 08:58:25 +0000 (10:58 +0200)]
[doc] Add missing layers (MC, SIMIX, SIMDAG) in README.coding
Gabriel Corona [Fri, 17 Jul 2015 08:52:41 +0000 (10:52 +0200)]
[doc] Various fixes to README.git
- Some typographic fixes.
- Remove reference to sourceforge.
- Some other minor modifications.
Gabriel Corona [Fri, 17 Jul 2015 08:21:39 +0000 (10:21 +0200)]
[mc] Fix another signedness conversion issue with addresses on 32 bit
Gabriel Corona [Fri, 17 Jul 2015 07:35:51 +0000 (09:35 +0200)]
[mc] Remove useless cast
Christian Heinrich [Thu, 16 Jul 2015 19:03:56 +0000 (21:03 +0200)]
[Doc] Changed stylesheet to make headings look better
Christian Heinrich [Thu, 16 Jul 2015 14:25:01 +0000 (16:25 +0200)]
[Doc] Significant update to the storage documentation
Christian Heinrich [Thu, 16 Jul 2015 17:17:41 +0000 (19:17 +0200)]
[SURF] Cosmetics in surfxml_parse.c
Gabriel Corona [Thu, 16 Jul 2015 14:37:31 +0000 (16:37 +0200)]
[mc] Fix cross process memory access in 32 bit
Sign extension issue when converting offset bewteen signed and unsigned.
Gabriel Corona [Thu, 16 Jul 2015 11:32:04 +0000 (13:32 +0200)]
[mc] SimGrid 3.11 Chanelog/NEWS
degomme [Fri, 10 Jul 2015 14:40:10 +0000 (16:40 +0200)]
protect against MPI_BOTTOM
degomme [Fri, 10 Jul 2015 13:32:53 +0000 (15:32 +0200)]
Fix fortran mpi_type_struct and mpi_type_create_struct functions
Martin Quinson [Tue, 14 Jul 2015 22:08:57 +0000 (00:08 +0200)]
Revert "A candidate patch to Github issue #15. Three problems related to TRACE might be corrected by this patch."
This reverts commit
58a9c67822f4640964044ec485426d090bcb123f.
Martin Quinson [Tue, 14 Jul 2015 22:08:32 +0000 (00:08 +0200)]
Revert "rename SIMIX_process_get_maxpid() to SIMIX_process_get_nextpid() and make it public"
This reverts commit
9306ddd4b063b1934dbab43f967e380d0181e393.
Martin Quinson [Tue, 14 Jul 2015 20:04:03 +0000 (22:04 +0200)]
kill an unused definition
Martin Quinson [Tue, 14 Jul 2015 20:02:33 +0000 (22:02 +0200)]
rename SIMIX_process_get_maxpid() to SIMIX_process_get_nextpid() and make it public
This ends the Fix #15 after the recent commit of @mpoquet. Thanks for
that again.
Martin Quinson [Tue, 14 Jul 2015 19:55:18 +0000 (21:55 +0200)]
Merge branch 'master' of github.com:mquinson/simgrid
Martin Quinson [Tue, 14 Jul 2015 19:54:24 +0000 (21:54 +0200)]
Merge pull request #17 from mpoquet/master
Candidate patch to issue #15
Martin Quinson [Sun, 12 Jul 2015 21:54:43 +0000 (23:54 +0200)]
kill two empty files
Martin Quinson [Sun, 12 Jul 2015 21:54:27 +0000 (23:54 +0200)]
who needs such an empty constructor?
Martin Quinson [Sun, 12 Jul 2015 21:30:05 +0000 (23:30 +0200)]
less void*, new type: surf_host_t
Plus some small cosmetics
Martin Quinson [Fri, 10 Jul 2015 23:58:18 +0000 (01:58 +0200)]
Merge branch 'master' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid
Martin Quinson [Fri, 10 Jul 2015 23:55:25 +0000 (01:55 +0200)]
do not output a user unfriendly line
"Switching host model to compound since you changed the network and/or cpu model(s)"
- Hard to understand what's going on without reading the right part of
the source.
- I cannot think of a situation where this would do any good
- My evil plan is to change the xbt_lib of all hosts by a boost map of
Hosts, so the host model will be gone, and hosts will always
dispatch to their content. One this is done, objectifying everything
will be really easy.
Martin Quinson [Fri, 10 Jul 2015 23:40:24 +0000 (01:40 +0200)]
ignore more cruft
Martin Quinson [Fri, 10 Jul 2015 23:36:01 +0000 (01:36 +0200)]
MASSIVE cleanup: s/workstation/host/
- simdag is not affected for now (it's a user interface, it will be
rewamped soon)
- It means that the workstation/model option is changed into host/model
That's a really intrusive change, and I'm sorry for the ones having
unmerged branches (Sorry Christian, I thought of you when I was almost
done). But I feel that this is a good move for the project, important
to the objectification of the whole.
Lucas Schnorr [Fri, 10 Jul 2015 17:00:21 +0000 (14:00 -0300)]
using trace timestamp precision when printing to trace files
Lucas Schnorr [Fri, 10 Jul 2015 16:59:09 +0000 (13:59 -0300)]
new option to set trace timestamp precision
the number must be an integer, whose value is the
desired number of digits after the decimal point
for all timestamps registered when tracing
Gabriel Corona [Thu, 9 Jul 2015 13:52:48 +0000 (15:52 +0200)]
[mc] Cleanup the DSO ignore list and add some entries
Gabriel Corona [Thu, 9 Jul 2015 12:44:58 +0000 (14:44 +0200)]
[mc] Fix the DSO debug ignore list
This is nasty. We must get rid of this.
Gabriel Corona [Thu, 9 Jul 2015 11:40:12 +0000 (13:40 +0200)]
Fix the cmake for the host-on-off tests
Fix #7
Gabriel Corona [Thu, 9 Jul 2015 10:53:21 +0000 (12:53 +0200)]
Fix the host-on-off refcount test
Gabriel Corona [Mon, 6 Jul 2015 14:21:17 +0000 (16:21 +0200)]
host-on-off: add basic tests
Gabriel Corona [Fri, 3 Jul 2015 12:41:09 +0000 (14:41 +0200)]
[jenkins] Enable debug (Ã-g) in the builds
Gabriel Corona [Fri, 3 Jul 2015 12:28:25 +0000 (14:28 +0200)]
Add the missing breaks
Gabriel Corona [Fri, 3 Jul 2015 11:36:53 +0000 (13:36 +0200)]
Move the host-on-off fix above
If our host is OFF, we should always get a HOST_FAILURE.
This seems to fix issues where a host was turned off but its processes
were not woken up with HOST_FAILURE.
Gabriel Corona [Tue, 30 Jun 2015 11:36:49 +0000 (13:36 +0200)]
Clean the possibly dangling waiting_synchro pointer
Gabriel Corona [Mon, 29 Jun 2015 11:08:51 +0000 (13:08 +0200)]
Fix host-on-off
This is a temporary workaround. This should probably be fixed at a
lower layer instead.
The test stil has a refcount handling bug (two +1 but only one -1).
Arnaud Legrand [Wed, 1 Jul 2015 19:01:54 +0000 (21:01 +0200)]
Fix the documentation
Arnaud Legrand [Wed, 1 Jul 2015 07:05:09 +0000 (09:05 +0200)]
Comment on a function whose behavior may be misleading...