Re: A patch for the file kernel/fork.c

[Andrew Morton <akpm@osdl.org>]

Alexander Nyberg <alexn@dsv.su.se> wrote:
>
> +	/* We're taking recursive faults here in do_exit. Safest 
>  +	 * is to just leave this task alone and wait for reboot. */

I find this comment-block style a bit hard to maintain, and am anal about
consistency.

>  +	if (unlikely(tsk->flags & PF_EXITING)) {
>  +		printk(KERN_ALERT "\nFixing recursive fault but reboot is needed!\n");
>  +		set_current_state(TASK_UNINTERRUPTIBLE);
>  +		schedule();
>  +	}
>  +

In the printk string, a \n will terminate the current facility level, so
your KERN_ALERT there is a no-op.  I simply removed it, which might cause
messy output sometimes but that seems better than always adding a newline.


--- 25/kernel/exit.c~avoid-recursive-oopses	2005-05-06 18:03:45.000000000 -0700
+++ 25-akpm/kernel/exit.c	2005-05-06 18:06:01.000000000 -0700
@@ -795,6 +795,17 @@ fastcall NORET_TYPE void do_exit(long co
 		ptrace_notify((PTRACE_EVENT_EXIT << 8) | SIGTRAP);
 	}
 
+	/*
+	 * We're taking recursive faults here in do_exit. Safest is to just
+	 * leave this task alone and wait for reboot.
+	 */
+	if (unlikely(tsk->flags & PF_EXITING)) {
+		printk(KERN_ALERT
+			"Fixing recursive fault but reboot is needed!\n");
+		set_current_state(TASK_UNINTERRUPTIBLE);
+		schedule();
+	}
+
 	tsk->flags |= PF_EXITING;
 
 	/*
_