Bug#407875: X server dies with backtrace and signal 11
On Mon, Mar 19, 2007 at 08:10:47PM +0100, Julien Cristau wrote:
> Hi Eric,
>
> On Mon, Mar 19, 2007 at 19:29:39 +0100, Eric wrote:
>
> >
> > Package: xserver-xorg
> > Version: 1:7.1.0-15
> >
> > My laptop (AMD Athlon Mobile 1200MHz running testing with Savage video
> > chipset) triggers a similar bug when I try to restart X-Window. I need
> > to reboot to get X working again.
> >
> > This bug seems to be the same as #60288 in Ubuntu. See
> > https://launchpad.net/ubuntu/+source/xorg-server/+bug/60288 .
> > A patch is provided there. It also might be already fixed upstream (in
> > Xorg 7.2 as stated in https://bugs.freedesktop.org/show_bug.cgi?id=8537 ).
> >
> Your log doesn't seem to have anything to do with any of those bugs, or,
> for that matter, Debian bug #407875. Please open a new report when you
> have a different problem (if the backtrace printed in the X log is
> significantly different, as is the case here, chances are you're hitting
> a different bug).
>
> > I applied the patch provided in Ubuntu BTS then rebuilt and installed
> > xserver-xorg which now seems to work. No more signal 11 when I restart
> > X.
>
> That's surprising, given that the backtrace you included below doesn't
> seem to be related to the code this patch touches.
Hello Julien,
from what I read at different places, I thought this bug were related
to a call to FontFileCompleteXLFD using, somewhere deep into its call
tree, an uninitialized, or not fully initialized, DRIScreenPrivPtr.
Even in my stacktrace the problem is not shown into the dri library
itself but, in effect, some changes to the content of the structure
pointed to by this DRIScreenPrivPtr could lead to memory corruption,
and such memory corruption could trigger a segfault later on.
Sorry if I did it wrong. I will open a new bug if you'd rather like.
Tell me.
> > So, when would you plan to upgrade to 7.2 in testing?
> >
> Never. Etch will release with xserver 1.1, lenny will have xserver 1.3
> when etch is released.
Ok, so could you apply the patch provided by Ubuntu please? It's rather
annoying to have to reboot after each logout to get X working again (and
I did not stressed X to discover all other usecases related to this
bug).
Best regards,
Eric.
--
Eric Delaunay | Le travail est trop sérieux pour le confier
delaunay@debian.org | à ceux qui veulent se tuer avec. Jissey.
Reply to: