Re: The O(1) scheduler breaks UML

[Jeff Dike <jdike@karaya.com>]

davidel@xmailserver.org said:
> Yes, this should work :
>     if (likely(prev != next)) {
>         rq->nr_switches++;
>         rq->curr = next;
>         next->cpu = prev->cpu;
>         spin_unlock_irq(&rq->lock);
>         context_switch(prev, next);
>     } else
>         spin_unlock_irq(&rq->lock);
> and there's no need for barrier() and rq reload in this way.

Yup, UML works much better with that.

			Jeff