-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Me again.
It seems that in my case this is a different bug (according to strace).
It is fglrx fault after all. I tried various X.org configurations, tried
downgrading and upgrading various packages, tested with various versions
of OpenOffice.org, and the only thing that helped, was using free
drivers instead of proprietary ones.
If I did manage to start OpenOffice.org (which seemed to be totally
random), it usually resulted in X.org crash, or kernel oops. According
to google, some Ubuntu users are also affected by this.
The weird thing is that fglrx drivers worked before, but I could not
determine what upgrade caused the failure, and, quite frankly, I killed
enough brain cells not to care anymore. No more fglrx for me.
I am attaching some information I gathered running 2.2.0-7, and fglrx,
in case you are interested.
As I have already mentioned, this most probably is a different bug
(coincidence?), so I cannot confirm the crash reported by Jean-Matthieu.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
iD8DBQFGRdjnztOe9mov/y4RAvaKAJ9gf6gFoR0Gju1rr3nWQ12aYUph6ACgvF5v
rEmErQIXbPSyqFKMtV1DqUo=
=xmp6
-----END PGP SIGNATURE-----
xrdb[6687] trap divide error rip:401997 rsp:7fffe54ff848 error:0
xrdb[6690] trap divide error rip:401997 rsp:7fff91d640a8 error:0
xrdb[6693] trap divide error rip:401997 rsp:7fff65dc3108 error:0
soffice.bin[6738] trap divide error rip:2aaf18cb53b4 rsp:7fff9ba62c10 error:0
soffice.bin[6796] trap divide error rip:2b9b600d43b4 rsp:7fff54641430 error:0
gdb /usr/lib/openoffice/program/soffice.bin
GNU gdb 6.6-debian
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu"...
Using host libthread_db library "/lib/libthread_db.so.1".
(gdb) run
Starting program: /usr/lib/openoffice/program/soffice.bin
[Thread debugging using libthread_db enabled]
[New Thread 47198352553696 (LWP 6882)]
warning: Lowest section in /usr/lib/libicudata.so.36 is .hash at
0000000000000120
Program received signal SIGFPE, Arithmetic exception.
[Switching to Thread 47198352553696 (LWP 6882)]
0x00002aed3a3733b4 in SalDisplay::Init () from
/usr/lib/openoffice/program/libvclplug_gen680lx.so
(gdb) bt
#0 0x00002aed3a3733b4 in SalDisplay::Init () from
/usr/lib/openoffice/program/libvclplug_gen680lx.so
#1 0x00002aed392b5dfc in GtkXLib::Init () from
/usr/lib/openoffice/program/libvclplug_gtk680lx.so
#2 0x00002aed392b67b5 in create_SalInstance () from
/usr/lib/openoffice/program/libvclplug_gtk680lx.so
#3 0x00002aed30c3e588 in tryInstance () from
/usr/lib/openoffice/program/libvcl680lx.so
#4 0x00002aed30c3e86e in CreateSalInstance () from
/usr/lib/openoffice/program/libvcl680lx.so
#5 0x00002aed309f4dde in InitVCL () from
/usr/lib/openoffice/program/libvcl680lx.so
#6 0x00002aed309f51ca in ImplSVMain () from
/usr/lib/openoffice/program/libvcl680lx.so
#7 0x00002aed309f53c5 in SVMain () from
/usr/lib/openoffice/program/libvcl680lx.so
#8 0x000000000041c11f in main ()
strace -v /usr/lib/openoffice/program/soffice.bin
[...]
poll([{fd=6, events=POLLIN, revents=POLLIN}], 1, -1) = 1
read(6, "\1\2W\0(\0\0\0\3\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 32) = 32
read(6, "^\0\0\0\0\2\2\0^\0\0\0\1\2\0\0]\0\0\0\2\1\1\0\1\4\f\0\2"..., 160) = 160
write(6, "\226\3\2\0\0\0\240\2", 8) = 8
read(6, 0x7fff2e57d5b0, 32) = -1 EAGAIN (Resource temporarily
unavailable)
poll([{fd=6, events=POLLIN, revents=POLLIN}], 1, -1) = 1
read(6, "\1\3X\0\2\0\0\0\4\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0_\265V\0"..., 32) = 32
read(6, "\1`\2b\3\0\6g", 8) = 8
uname({sysname="Linux", nodename="compaq", release="2.6.21.1.666.0",
version="#1 SMP PREEMPT Fri May 11 22:15:04 EEST 2007", machine="x86_64"}) = 0
open("/home/linas/.Xdefaults-compaq", O_RDONLY) = -1 ENOENT (No such file or
directory)
--- SIGFPE (Floating point exception) @ 0 (0) ---
+++ killed by SIGFPE +++
Process 6872 detached
