Bug#404346: vino-server segmentation fault

[Guy Roussin]

Loïc Minier <lool at dooz.org> a écrit :

Hi Loïc,

> > Program received signal SIGSEGV, Segmentation fault.
> > [Switching to Thread -1224267552 (LWP 15831)]
> > vino_input_init (display=0x808f098) at vino-input.c:132
> > 132                   if (global_input_data.keycodes [keysym] != 0)
> > (gdb) bt
> > #0  vino_input_init (display=0x808f098) at vino-input.c:132
> > #1  0x080501e0 in main (argc=1, argv=0xbfba5f64) at vino-main.c:78
> 
>  Weird but interesting.  The crash happens very near this upstream
>  change:
>     <http://bugzilla.gnome.org/attachment.cgi?id=77801>
>  (from GNOME #369884 http://bugzilla.gnome.org/show_bug.cgi?id=369884)
> 
>  Could you try this upstream patch?  If it fixes the segfault, I can
>  pull it easily.

Yes i try the patch but it doen't fix the crash :-(

I try also this patch but again without success (same crash):
2005-11-16  Mark McLoughlin  <mark at skynet.ie>
	Fix crash with unicode mapped keysyms (bug #321516)
	Patch from Gary Coady <gary at lyranthe.org>
	* server/vino-input.c: update VINO_IS_LATIN1_KEYSYM
	to not match keysyms with byte 4 set.

>  If it doesn't and you have the time to, you can try
>  building the upstream CVS of vino to see whether your bug is already
>  fixed upstream (I don't experience it myself), and if it is extract a
>  patch: this might make it possible to push the change in Debian before
>  the release of etch.
I have time but i don't know the process to build upstream CVS of vino.
Perhaps you can give some hints. Then i give it a try ?

> PS: I'll be on holidays for a week starting tonight.
Thank you Loïc and good holidays :-)

--
Guy Roussin

___________________________________
http://www.teledetection.fr