Re: netem and hierarchical ingress traffic shaping

[Eric Dumazet <eric.dumazet@gmail.com>]

Le vendredi 23 décembre 2011 à 11:07 -0800, Stephen Hemminger a écrit :

> skb_cb is the dumping ground of the networking layer.
> The assumption was that the qdisc could use the skb_cb
> for it's own scratchpad. Netem is using it for tagging
> packets in the queue. 
> 
> So basically, netem, choke, and sfb are incompatible with
> each other. This is not that bad, why not add a flag to qdisc
> ops to indicate which qdisc are using cb and block user from
> trying to do something bogus.

This is not how I planned to solve the problem.

I think we need an internal tfifo for netem use.

Then be able to add another qdisc on top of netem.


tfifo as a first stage, hardcoded (only limit is tunable)
[ optional 2nd stage, any qdisc ]

netem_queue()
{
	queue packet to tfifo
	(eventually at head, of reordering)
}

netem_dequeue()
{
	if (other_qdisc) {
		for_each_packet_from_tfifo_time_ready() {
			other_qdisc->enqueue(skb);
		}
		try_to_dequeue_one_packet_from(other_qdisc);
	} else {
		dequeue_one_packet_from_tfifo_time_ready();
	}
}