[PATCH] Don't call FatalError from the signal handler

[Edgar Toernig]

Peter Hutterer wrote:
>
> > > The attached patch sets up a siglongjmp environment and - in case of a 
> > > fatal error during a signal - jumps back to the set up state before 
> > > bringing down the server.
> > 
> > This won't help at all - the lock is still held.  A longjmp doesn't
> > magically unlocks all locks and brings all data structures back into
> > a consistent state.  It's really only a jump and the rules which
> > functions are save to be called still apply.
> 
> Thanks. Do you have any suggestions to avoid this problem?

Fix the bugs causing the segv ;-)

I guess, the whole "bringing down the server" is about restoring the
video state.  An "emergency_shutdown()" callback in the video driver
that does only the minimally necessary (stop DMA and restore video
mode) without using any problematic functions (like malloc/free,
anything from stdio, etc) would IMHO be the most promising solution.

Another more messy way: call the regular FatalError in the hope that
it succeeds but before doing so set an alarm timer to i.e. 2 seconds
and _exit() if it fires off.  [Maybe the right place to add this time-
out is in FatalError itself...]

Ciao, ET.