> > > - we want _long_ timeslices to maximize speed and to minimize
> > > impact on cache/memory/etc. (say 1 minute)
> >
> > doing a full cacheflush takes some ~2 millisecs on most x86 boxes. ie the
> > default 200 msecs timeslice is OK. Also even though they are SCHED_IDLE,
> > it does not mean rc5 clients want to time out on network connections
> > just because they happen to have a 1 minute timeslice each ...
>
> I hadn't considered this... OTOH, it might be desirable for
> large simulations to have 'slices' of 15 minutes so that only
> one of these will be in memory at once (when memory pressure
> is high).
IMO you are mixing two things: priorities and lengths of
timeslices. These are separate: you want low-priority short-timeslice
for screensavers so that multiple screensavers move smoothly; you want
low-priority long-timeslice for simulations. You might even want
high-priority long-timeslice for important simulations :-). Do not mix
that up.
Pavel
-- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-).- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html