[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 7 Jun 2016 00:52:27 +0200
From: Florian Westphal <fw@...len.de>
To: Eric Dumazet <eric.dumazet@...il.com>
Cc: Florian Westphal <fw@...len.de>, netdev@...r.kernel.org
Subject: Re: [PATCH -next] cbq: remove only caller of qdisc->drop()
Eric Dumazet <eric.dumazet@...il.com> wrote:
> > static void cbq_ovl_drop(struct cbq_class *cl)
> > {
> > - if (cl->q->ops->drop)
> > - if (cl->q->ops->drop(cl->q))
> > - cl->qdisc->q.qlen--;
> > + struct sk_buff *skb = cl->q->ops->dequeue(cl->q);
> > +
> > + if (skb) {
> > + cl->deficit -= qdisc_pkt_len(skb);
> > + cl->qdisc->q.qlen--;
> > + qdisc_drop(skb, cl->qdisc);
> > + }
> > +
> > cl->xstats.overactions++;
> > cbq_ovl_classic(cl);
> > }
>
> A drop() is not equivalent to a dequeue() followed by qdisc_drop() for
> statistics.
>
> dequeue() will update stats of _sent_ packets/bytes, while drop() should
> not.
Well, I could send patch to just remove cbq_ovl_drop completely,
you can't configure this facility with iproute2.
You are right of course, but is it really worth to have this?
Not calling cl->q->ops->drop() in cbq would allow removal of ~300 LOC
in qdiscs...
Powered by blists - more mailing lists