Re: crash in entry.S restore_all, 2.6.12-rc2, x86, PAGEALLOC

[Ingo Molnar]

* Mikael Pettersson <mikpe@xxxxxxxxx> wrote:

>  > -	jz restore_all
>  > +	jz restore_nocheck
>  >  	testl $IF_MASK,EFLAGS(%esp)     # interrupts off (exception path) ?
>  > -	jz restore_all
>  > +	jz restore_nocheck
>  >  	call preempt_schedule_irq
>  >  	jmp need_resched
>  >  #endif
> 
> Is this sufficient or do we also need the 
> s/restore_all/restore_nocheck/ at around line 553 which was in the 
> first posted patch?

all 3 restore_all->restore_nocheck changes are needed to make the bug 
not trigger. I dont think it's a real fix because whatever is being 
preempted involuntarily, it ought to have a proper kernel stack to begin 
with. I'll do some more debugging.

	Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/