> >>>>> "Rik" == Rik van Riel <riel@nl.linux.org> writes:
> > I think this might be worthwhile because the large
> > recalculation holds the tasklist_lock for a _long_
> > time, blocking the other CPU's access to the runqueue.
>
> The use of separate runqueues per processor is obvious enough that
> it must have been mentioned already, but I can't seem to find a discussion
> of it in the archive. What's wrong with such an option?
It's bloody impossible to get load balancing across CPUs
done well cheaply. Besides, if you've got 16 CPU's, walking
the task list is going to be the least of your problems...
Rik -- Open Source: you deserve to be in control of your data.
+-------------------------------------------------------------------+
| Le Reseau netwerksystemen BV: http://www.reseau.nl/ |
| Linux Memory Management site: http://humbolt.geo.uu.nl/Linux-MM/ |
| Nederlandse Linux documentatie: http://www.nl.linux.org/ |
+-------------------------------------------------------------------+
-
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/