Re: [uPATCH] SMP scheduling fix (?)

Roger Espel Llima (espel@llaic.u-clermont1.fr)
Wed, 20 Jan 1999 14:26:49 +0100


Stephen Tweedie wrote:
> D'oh. Sleeping processes are irrelevant, of course. :-O
>
> The sched-idle scheduler must never stall a kernel-mode runnable process
> indefinitely. As long as it only stalls preempted user-mode processes,
> all is well.

Maybe I'm talking out of my ass here, but to prevent "nice -20"
processes from hogging a resource indefinitely, wouldn't it be enough to
temporarily bump them up (i.e treat them like "nice -19") when they are
actually holding one?

When they aren't holding any lock or waiting on anything (which should
be most of the time anyway on background computation tasks), they can be
scheduled only if there's nothing else to run.

-- 
Roger Espel Llima, espel@llaic.u-clermont1.fr
http://www.eleves.ens.fr:8080/home/espel/index.html

- 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.tux.org/lkml/