Re: [RFC patch 08/18] cnt32_to_63 should use smp_rmb()
From: David Howells
Date: Fri Nov 07 2008 - 18:56:54 EST
Nicolas Pitre <nico@xxxxxxx> wrote:
> I explained several times already why I disagree. Preemption is not a
> problem unless you're preempted away for long enough, or IOW if your
> counter is too fast.
That's my point. Say a nice -19 process gets preempted... what guarantees
that it will resume within, the time it takes the counter to wrap? Even if
the preempting process goes back to sleep, in that time a bunch of other
processes could have woken up and could starve it for a long period of time.
David
--
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/