[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: glibc: causes segfault in Xorg



On Mit, 2011-05-04 at 02:18 -0500, Jonathan Nieder wrote: 
> 
> Michel Dänzer wrote:
> > On Mit, 2011-05-04 at 00:10 -0500, Jonathan Nieder wrote: 
> >> Steve M. Robbins wrote:
> 
> >>> Program received signal SIGSEGV, Segmentation fault.
> >>> __memcpy_ssse3 () at ../sysdeps/x86_64/multiarch/memcpy-ssse3.S:153
> >>> 153     ../sysdeps/x86_64/multiarch/memcpy-ssse3.S: No such file or directory.
> >>>         in ../sysdeps/x86_64/multiarch/memcpy-ssse3.S
> >>> (gdb) bt
> >>> #0  __memcpy_ssse3 () at ../sysdeps/x86_64/multiarch/memcpy-ssse3.S:153
> >>> #1  0x00007ffff3858db1 in shadowUpdatePacked () from /usr/lib/xorg/modules/libshadow.so
> >>> #2  0x00007ffff385843f in ?? () from /usr/lib/xorg/modules/libshadow.so
> >>> #3  0x000000000043571d in BlockHandler ()
> >>> #4  0x000000000045dcda in WaitForSomething ()
> >>> #5  0x00000000004314b2 in ?? ()
> >>> #6  0x00000000004257de in _start ()
> [...]
> > The purpose of shadowUpdatePacked is to copy pixels from a shadow
> > framebuffer to the visible screen. It would be rather pointless for the
> > shadow framebuffer to be contained within the visible screen, so
> > shadowUpdatePacked should always be able to safely use memcpy as far as
> > overlapping areas is concerned.
> >
> > If shadowUpdatePacked is indeed calling memcpy for overlapping areas
> > here, that's probably a bug in the X driver being used.
> 
> Thanks, Michel.  Steve, could you install xserver-xorg-video-radeon-dbg
> and get a full backtrace (bt full), or even better, run xorg under
> valgrind and see what it says?

Actually, looking at the Xorg.0.log file, there's no mention of the
radeon driver at all, it's using the fbdev driver:

[ 3604.046] (II) FBDEV(0): hardware: radeondrmfb (video memory: 9000kB) 
[  3604.046] (II) FBDEV(0): checking modes against framebuffer device...
[  3604.046] (II) FBDEV(0): checking modes against monitor...
[  3604.046] (--) FBDEV(0): Virtual size is 3200x1200 (pitch 3200)

And the numbers don't seem to add up, 3200x1200 would require almost 15M
of VRAM.

Steve, can you provide the output of fbset -i and the
"/etc/X11/xorg.conf file as well? Any reason for not using the radeon
driver?


-- 
Earthling Michel Dänzer           |                http://www.vmware.com
Libre software enthusiast         |          Debian, X and DRI developer


Reply to: