On Wed, 2007-01-31 at 10:59 +0100, Ingo Molnar wrote:
> * Daniel Walker <[EMAIL PROTECTED]> wrote:
>
> > This is something Thomas already dropped, [...]
>
> (i think you forgot to Cc: Thomas here, nor is this something that
> Thomas' change dropped.)
Yes your right .
> > [...] and I'm just sticking with that .. If you register your
> > clocksource _twice_ your kernel will likely not work correctly (and
> > might crash).
>
> this is a quite bad change. John's original clocksource code protects
> against double registry:
>
> if (is_registered_source(c)) {
> printk("register_clocksource: Cannot register %s. "
> "Already registered!", c->name);
> ret = -EBUSY;
>
> and Thomas' change keeps that property, but doesnt printk. Your change
> makes double registry possible, potentially crashing the kernel later
> on! (And this isnt theoretical, double registry did happen in practice
> when i debugged suspend problems on my SMP laptop.)
My original patch set makes the duplicate register checking better
(which I dropped) .. I'll be happy to reintroduce that part of it..
Daniel
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
