Re: [RFC V2 PATCH 1/5] cfq-iosched: adapt slice to number of processes doing I/O

From: Jeff Moyer
Date: Mon Oct 26 2009 - 11:10:35 EST


Corrado Zoccolo <czoccolo@xxxxxxxxx> writes:

> Hi Jeff,
> On Wed, Oct 21, 2009 at 6:36 PM, Jeff Moyer <jmoyer@xxxxxxxxxx> wrote:
>> Corrado Zoccolo <czoccolo@xxxxxxxxx> writes:
>>
>> OK. ÂThe idea was to try to make the thing a bit more digestable.
>> Honestly, I had to re-write it to figure out what it was doing. ÂCould
>> you rework it so the logic is more obvious to others and still correct?
>
> What about:
> /* interested queues (we consider only the ones with the same priority class) */
> unsigned iq = cfq_get_avg_queues(cfqd, cfq_class_rt(cfqq));
> unsigned sync_slice = cfqd->cfq_slice[1];
> unsigned expected_latency = sync_slice * iq;
> if (expected_latency > target_latency) {
> unsigned base_low_slice = 2 * cfqd->cfq_slice_idle;
> /* scale low_slice according to IO priority and sync vs async */
> unsigned low_slice = min(slice, base_low_slice * slice / sync_slice);
> /* the adapted slice value is scaled to fit all iqs into the target latency */
> slice = max(slice * cfq_target_latency / expected_latency, low_slice);
> }

Yeah, that looks a ton better. Thanks for doing that!

Cheers,
Jeff
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/