Re: [uPATCH] SMP scheduling fix (?)

Rik van Riel (riel@humbolt.geo.uu.nl)
Mon, 18 Jan 1999 20:08:59 +0100 (CET)


On Fri, 15 Jan 1999, Pavel Machek wrote:

> > I'd _really_ like to have a nice value that says "don't run unless you
> > are twiddling your thumbs". Ie a nice 20 perhaps, that says I don't
> > want
>
> I'd like that, too. But it is currently impossible with linux: it
> opens DoS attack.
>
> make your nice 20 task hold some lock, and then run 'normal' task so
> that nice 20 task will never ever get cpu again. BOom, you are holding
> lock and are not going to release it.
>
> Is there solution to this? Maybe processes running in kernel get
> automagical priority boost?

An soon as a process gives up the CPU (and reschedules)
it's in kernel mode. We can, however, use the arch-specific
TSS structure to see if the task is in kernel mode.

Since we only have to do that when rescheduling SCHED_IDLE
processes, the costs will actually be bearable for the kernel
and Linus can be convinced that it's worth the trouble...

Rik -- If a Microsoft product fails, who do you sue?
+-------------------------------------------------------------------+
| Linux memory management tour guide. riel@nl.linux.org |
| Scouting Vries cubscout leader. http://www.nl.linux.org/~riel |
+-------------------------------------------------------------------+

-
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/