On Mon, 9 Apr 2001, Alan Cox wrote:
> > > Its worth doing even on the ancient x86 boards with the PIT.
> >
> > Note that programming the PIT is sloooooooow and doing it on every timer
> > add_timer/del_timer would be a pain.
>
> You only have to do it occasionally.
>
> When you add a timer newer than the current one
> (arguably newer by at least 1/2*HZ sec)
That's only if we want to do no better than the current system. We'd want
a new variable called timer_margin or something, which would be dependent
on interrupt source and processor, and could be tuned up or down via
sysctl.
> When you finish running the timers at an interval and the new interval is
> significantly larger than the current one.
Make that larger or smaller. If we come out of a quiescent state (1 hz
interrupts, say) and start getting 10ms timers, we want to respond to them
right away.
> Remember each tick we poke the PIT anyway
We could also have a HZ_max tunable above which we would not try to
reprogram the interval. On older systems, this could be set at
100-200HZ...
-- "Love the dolphins," she advised him. "Write by W.A.S.T.E.."- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Apr 15 2001 - 21:00:16 EST