GitLab

because i'd like to use that name for the ocaml/lutin lib.

The C binding does not work yet (hence the test in examples/C and
examples/lustre are broken).

It was due to an uninitialized local var ; the bug was revealed by a change in
caml/mlvalues.h::CAMLparam() done in 4.02.0

nb : it also fixes (i don't really know why) a performance bug in the rabbit simu.

Update the lus2lic plugin

cf rdbg git 66d01f567940e06eab9a49604663d43c29fdfe01

Indeed, in source/Lutin/Makefile.lut4ocaml, I was including
rdbg-plugin when building the lut4ocaml.cma/cmxa lib. The consequence
was that the OcamlRM module was included twice, with its local
Hashtbl (used for registering/loading Dynlink stuff) duplicated!

Dynlink is to be used with great care !!!

Also reflect the changes in the rdbg-plugin
- in the OcamlRM module (s/add_/reg_/g)
- in the Event module (more field of src_info type)
cf rdbg git version 142783a77cad1a0f7ef91972b8376f0b0e44b878

also mv lut_evt stuff from LutinRun to rdbg (in LutinRdbg module).

(cf 55d3d77c470f416befd16a1b7d47f751eb2171fc)

remove ldbg tests (ldbg is dead, vive rdbg!)

Note that it triggered a bug in the use of Random.int: it cannot be
use on max_int, as its max admissible value is 2^30, and max_int=2^62
on 64 bits arch!!

Also, -rif now points to -quiet (so that it has the same meaning as for ecexe),
and the previous -rif is obtained via -o.

Note that rif_base has been renamed rifIO, and event to rdbgEvent.

The idea is that this repo will contain closed-source files, and that
rdbg will be open-source.

Moreover, to be able to use Lutin from rdbg, I needed to tweek
Makefiles, and rename a few modules that were clashing with ocaml
compiler-libs for instance.

Event, Data, Expr, Failure are now defined in rdbg and should be installed
in the working/<arch>/lib/ directory

Remove everything related to xlurette.

Indeed, there is no reason to have clashes between oracle outputs.

Also, expand var types in LustreRun (as done in SocExec).

(and do not go via ec anymore).

Split the Failure module so that it can be shared with the lus2lic
ldbg plugin without requiring the bdd lib.

+ mv the lus2licRun.ml/mli file to the Lurette git repo.

+ add a copy (in read-only mode) of the source files of the lus2lic compiler

+ Share the comion content of Makefile.lurettetop and Makefile.ldbg via a new
Makefile.comon that is included in both.

+ Merge code of modules that are now shared with the lus2lic repo:
 -  Verbose, Data, Rif_base

Also extend the Data module so that it can be shared with lus2lic.

(the butter, the money of the butter, and the *bip* of the Dairywoman).

 - Add a -dyn option that parses (with the last command) the end of the rif file
 - Add a -pdf option
 - Add the possibility to call the GnuplotRif main function from a ocaml program

Actually, I've simplified the code a lot while being more efficient.

1) Instead of generating explicitely a gnuplot data file (to remove
#outs/#step, etc), I use a sed/grep script file (cf util/read-rif.sh)
2) Instead of calibrating the display while generating this file in ocaml, I do
it directly via gnuplot
3) the dynamic part relies on the Unix last command

GBDDML: add forall and exist in the caml API

Actually, I've replaced a particularly horrible chunk
of code that was buggy (I knew it, but i am lazy...).

Make sure everything is reproductible during time traveling, or when
starting back from any stored event, by reseting the random seed
using the event number each time the event number changes.

Add time travelling support (à la appel/Tolmach) in Ldbg_utils.

Re-organise the Event.t structure

Add Call events.

Fix the step number for "run" lutin machines.

Add Events when an elected contraint fails (backtrack).

Attach the event number to the Deadlock exception in non debug mode
to ease the debugging when an unexpected deadlock is raised at
top-level.

Trace also the SUT if it is a Lutin program.

Attach the event number to the Deadlock exception in non debug mode
to ease the debugging when an unexpected deadlock is raised at
top-level.

Add hooks support for Ldbg_utils.next.

Add a simple Ldbg_utils.profiler.

Add a reset_rp command.

Also, do not call clear_step all the time but only before the step
(which change some non-reg test).

Add a -check option to Lutin and performs all the checks but do not
run.

A pass to remove performance penalty in lurettetop (now the panalty in less than 1%).
The idea was to compute source info lasily.

Add nonreg tests for ldbg + crazy-rabbit, as well as missing files (oops).

Add online help msgs for the Ldbg_utils modules.

Better source code information. In particular, I've attached a
'Syntaxe.val_exp option' to each lxm:Lexeme.t in the
CoIdent.scope_stack type. The rationnale is to bez able to display
the whole expression associated to a lxm in ldbg (versus only the
lxm).

Among other things:

- split ldbg into ldbg and ldbg_utils.
- refine the event content.
- add support for online help.

Now I can trace lutin step called from run statements!

I've duplicated a lot of code in LutExe ; i'll fix that later...

nb : the gnuplot window needs to be refresed manually ('a' keystroke).
I should find a way to refresh it automatically.

Add a  --ocamlc-version option to lurettetop and lutin.

The idea is to put all the .exp files into a `lurettetop/lutin --ocamlc-version`
dir so that I have an .exp file for each ocaml version.

If the `lurettetop --ocaml-version`/*.exp does not exist, I generate it.

Also, don't non-reg test lucky dirs anymore (because they are boring).

Aldo, don't rely on the HOSTTYPE env var, but use Sys.os_type instead.

This breaks non-reg tests in lutin/test_ok because the exc Polka.overflow
is raised whereas the previous version was silently doing something wrong.