On Sun, Jul 17, 2005 at 10:10:07AM -0400, Michael wrote:
> Hmm, that's odd. Did the Xserver leave a coredump? Did you have anything
> like -pthread in your $CFLAGS? The crash must have been something odd
> since XFree usually traps SIGSEGV and friends so it can exit cleanly. Do
> you have Option "NoTrapSignals" in your XF86Config?

Nope, no core dump that I could find. I didn't build with any custom CFLAGS
and NoTrapSignals isn't set (I'm still using a fairly minimal XF86Config -
the one generated by XFree86 -configure with one or two minor changes).

> You can always restart X from a remote login.

D'oh! Didn't think of that - was so surprised when X crashed like that that
I didn't think straight :)

> Please do this:
> - run XFree86 in gdb from a remote login
> - start an xterm ( from another remote login... ) and from that one run
> firefox - see if you can reproduce the crash and send me the stack trace
> ( use the bt command in gdb ) 

OK, will do so shortly

> btw. do you get these crashes when running a binary snapshot from
> releng?

Haven't tried yet - I normally cross-build my own _BETA and -current
releases. If this problem continues, I'll try a releng snapshot.

-mj
-- 
Michael-John Turner | http://mj.turner.org.za/
mj@turner.org.za    | Open Source in WC ZA - http://www.clug.org.za/