From: Branko Lankester (lankeste@fwi.uva.nl)
Date: 05/19/92


From: lankeste@fwi.uva.nl (Branko Lankester)
Subject: Re: gdb postmortem debugging...
Date: Tue, 19 May 1992 10:34:11 GMT

torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds) writes:

>Yes, this will fix the problem, but it's not completely safe: if the
>error occurs in the kernel, it will probably lock up. Most of the

Ok, here is a `save' patch :)

*** traps.c.orig Fri Apr 24 22:24:36 1992
--- traps.c Tue May 19 12:30:00 1992
***************
*** 83,89 ****
        for(i=0;i<10;i++)
                printk("%02x ",0xff & get_seg_byte(esp[1],(i+(char *)esp[0])));
        printk("\n\r");
! do_exit(11); /* play segment exception */
  }
  
  void do_double_fault(long esp, long error_code)
--- 83,92 ----
        for(i=0;i<10;i++)
                printk("%02x ",0xff & get_seg_byte(esp[1],(i+(char *)esp[0])));
        printk("\n\r");
! if ((0xffff & esp[1]) == 0xf)
! send_sig(SIGSEGV, current, 0);
! else
! do_exit(11); /* play segment exception */
  }
  
  void do_double_fault(long esp, long error_code)