Re: 2.6.8.1-mm1

[Zwane Mwaikambo]

On Tue, 17 Aug 2004, Nathan Lynch wrote:

> i386 seems to want something like this to avoid crashing in
> find_busiest_group... looks like there's a short window in fixup_irqs
> where interrupts are enabled while we're taking a cpu down.

> Index: 2.6.8.1-mm1/arch/i386/kernel/apic.c
> ===================================================================
> --- 2.6.8.1-mm1.orig/arch/i386/kernel/apic.c
> +++ 2.6.8.1-mm1/arch/i386/kernel/apic.c
> @@ -1138,7 +1138,8 @@
>  	 * interrupt lock, which is the WrongThing (tm) to do.
>  	 */
>  	irq_enter();
> -	smp_local_timer_interrupt(&regs);
> +	if (!cpu_is_offline(cpu))
> +		smp_local_timer_interrupt(&regs);
>  	irq_exit();

Hmm i really am trying to avoid that, thats why we also mask the APIC
timer after fixup_irqs, can we try tackling that instead?
-
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/