Hi Florian,
since it seems you can reproduce this easily it would be nice to get
some more info from gdb, see below.
On Wed, Jun 11, 2008 at 20:16:36 +0200, Florian Kulzer wrote:
> $ gdb /usr/lib/openoffice/program/soffice.bin $(pidof soffice.bin)
>
> [...]
>
> (gdb) continue
> Continuing.
> [Thread 0x41ef0950 (LWP 15165) exited]
> [Thread 0x42ef2950 (LWP 15168) exited]
> [Thread 0x408dd950 (LWP 15166) exited]
>
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 0x7fba2089e780 (LWP 15163)]
> _XFreeExtData (extension=0x32500000010) at ../../src/FreeEData.c:39
> 39 ../../src/FreeEData.c: No such file or directory.
> in ../../src/FreeEData.c
34 int
35 _XFreeExtData (XExtData *extension)
36 {
37 XExtData *temp;
38 while (extension) {
39 if (extension->free_private)
40 (*extension->free_private)(extension);
41 else Xfree ((char *)extension->private_data);
42 temp = extension->next;
43 Xfree ((char *)extension);
44 extension = temp;
45 }
46 return 0;
47 }
Looks like the extension structure is fucked up (actually the extension
pointer itself looks strange, so maybe it's the display structure that's
corrupted somehow). Can you type 'p *extension' at the gdb prompt at
this point?
> Current language: auto; currently c
> (gdb) backtrace
> #0 _XFreeExtData (extension=0x32500000010) at ../../src/FreeEData.c:39
> #1 0x00007fba1cc9f2a4 in _XFreeDisplayStructure (dpy=0x6a71e0) at
> ../../src/OpenDis.c:860
856 if (dpy->pixmap_format) {
857 register int i;
858
859 for (i = 0; i < dpy->nformats; i++)
860 _XFreeExtData (dpy->pixmap_format[i].ext_data);
861 Xfree ((char *)dpy->pixmap_format);
862 }
Would be nice to print *dpy here.
> #2 0x00007fba1cc8ab72 in XCloseDisplay (dpy=0x6a71e0) at
> ../../src/ClDisplay.c:82
Also, as a data point maybe you can downgrade libx11-6 to a previous
version and see if the bug still triggers.
Thanks,
Julien
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
