On Sat, 24 Jul 2010, Richard W.M. Jones wrote:

> On Sat, Jul 24, 2010 at 08:32:06AM +0200, Julia Lawall wrote:
> > I had to patch pycaml/pycaml_ml.c as well.  But with this change, 
> > everything was fine for me, both on demos/simple.cocci, which doesn't even 
> > use python, and demos/regexp.c, which does use python.
> > 
> > Please let me knowhow it works out.
> 
> It compiles OK, but the test still fails in the same way:
> 
> http://koji.fedoraproject.org/koji/getfile?taskID=2347786&name=build.log
> 
> I still suspect the test failure may be completely unrelated, but (a)
> it only fails when I patch pycaml_ml.c, and (b) I don't actually have
> a system with Python 2.7 to test against so all I can do is submit
> jobs to our build system.  This makes it very hard to diagnose what is
> really going on ...

Is there any way that you can define the environment variable 
OCAMLRUNPARAM to b and then use spatch rather than spatch.opt?  Then you 
will get a backtrace from ocaml that might give more information about 
where is the problem.

julia



> 
> Rich.
> 
> -- 
> Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
> New in Fedora 11: Fedora Windows cross-compiler. Compile Windows
> programs, test, and build Windows installers. Over 70 libraries supprt'd
> http://fedoraproject.org/wiki/MinGW http://www.annexia.org/fedora_mingw
> 
_______________________________________________
Cocci mailing list
[email protected]
http://lists.diku.dk/mailman/listinfo/cocci
(Web access from inside DIKUs LAN only)

Reply via email to