Re: [PATCH 2/5] cfq-iosched: Implment IOPS mode for group scheduling

[Vivek Goyal <vgoyal@redhat.com>]

On Tue, Jul 27, 2010 at 01:47:39PM +0800, Gui Jianfeng wrote:

[..]
> > @@ -913,19 +927,21 @@ static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg,
> >  				struct cfq_queue *cfqq)
> >  {
> >  	struct cfq_rb_root *st = &cfqd->grp_service_tree;
> > -	unsigned int used_sl, charge_sl;
> > +	unsigned int used_sl, charge;
> >  	int nr_sync = cfqg->nr_cfqq - cfqg_busy_async_queues(cfqd, cfqg)
> >  			- cfqg->service_tree_idle.count;
> >  
> >  	BUG_ON(nr_sync < 0);
> > -	used_sl = charge_sl = cfq_cfqq_slice_usage(cfqq);
> > +	used_sl = charge = cfq_cfqq_slice_usage(cfqq);
> >  
> > -	if (!cfq_cfqq_sync(cfqq) && !nr_sync)
> > -		charge_sl = cfqq->allocated_slice;
> > +	if (iops_mode(cfqd))
> > +		charge = cfqq->slice_dispatch;
> 
> Hi Vivek,
> 
> At this time, requests may still stay in dispatch list, shall we add a new variable
> in cfqq to keep track of the number of requests that go into driver, and charging
> this number?
> 

Hi Gui,

How does that help. Even if request is in dispatch list, sooner or later
it will be dispatched. As long as we can make sure that requests in 
dispatch list are in proportion to group weights, things should be just
fine.

Thanks
Vivek