From: Marcus Sundberg <mackan@stacken.kth.se>
  To  : ggi-develop@eskimo.com
  Date: Fri, 30 Apr 1999 00:20:09 +0000

Re: LibGGI buglets

Marcus Sundberg wrote:
> 
> Andrew Apted wrote:
> > I've tracked this one down.  It happens because input/linux_kbd/input.c
> > calls ggRegisterCleanup(), which calls register_sighandler(), which
> > calls signal() on the SWITCHSIG setup by display/linvtswitch/visual.c,
> > and this causes the behaviour to become SA_RESETHAND, thus linvtswitch's
> > signal handler is set to SIG_DFL after the VT_PROCESS release signal
> > comes in, thus causing the acquire signal to kill the program.
> >
> > AAAAAAARRRRRRRRRGGGGGGGHHHHHHHHH  !
> >
> > The solution is not to use signal() but always use sigaction().  Marcus,
> > could you please fix that ?  Thanks heaps !
> 
> Ah, gotto love those little differences between libcand glibc...
> Thanks for finding the problem, will fix.

Ok, should be fixed now. Don't forget to reconfigure so HAVE_SIGACTION
gets defined.

//Marcus
-- 
-------------------------------+------------------------------------
        Marcus Sundberg        | http://www.stacken.kth.se/~mackan/
 Royal Institute of Technology |       Phone: +46 707 295404
       Stockholm, Sweden       |   E-Mail: mackan@stacken.kth.se