On Thu, Aug 29, 2002 at 01:40:25PM +0300, Martti Kuparinen wrote:
> As you can see #7 is unresolved and at a completely different
> address than the rest of the stuff...

The server caught a fatal signal 11 there, caught it and logs it.
I'm not sure if 0xbfbfdfdc realy is the signal trampoline or if gdb just
borked decoding the trapframe, but this does not matter.

> #5  0x811ce45 in FatalError (
>     f=0x843d120 "Caught signal %d.  Server aborting\n") at utils.c:1399
> #6  0x80ba094 in xf86SigHandler (signo=11) at xf86Events.c:1094
> #7  0xbfbfdfdc in ?? ()
> #8  0x804da54 in NVDACRestore (pScrn=0x8520000, vgaReg=0x8612024,
>     nvReg=0x85208c0, primary=1) at nv_dac.c:203

Here, in #8 lies the problem. Look at what place the signal happens and why.
You might try running the whole X server (from another machine via network)
in gdb. This way gdb catches the signal before xf86SigHandler is invoked.

Martin