Re: [PATCH] netem: account for packets in delayed queue in qlen

[jamal <hadi@cyberus.ca>]

On Fri, 2005-22-04 at 01:39 +0200, Patrick McHardy wrote:
> Stephen Hemminger wrote:
> > I'm thinking of changing enqueue (and maybe later dequeue) API to decouple
> > the qlen assumption.
> > 
> > Either:
> > 	rc = qdisc->enqueue(skb, qdisc, &my->qlen)
> > or add NET_XMIT_DUPPED
> > 	rc = qdisc->enqueue(skb, qdisc);
> > 	if (rc < NET_XMIT_SUCCESS) {
> > 		++my->dropped;
> > 	} else {
> > 		my->qlen++;
> > 		if (rc == NET_XMIT_DUPPED)
> > 			my->qlen++;
> > }
> 
> To be frank, I don't like either one. Both have the same problem
> wrt. HFSC, the first solution requires changes all over the place,
> the second one is unflexible and also requires lots of changes. I
> don't see what could benefit from this API change besides netem,
> so I'd vote to go with my proposed solution: store the parent
> pointers in struct Qdisc, add code to walk up the tree and adjust
> the qlen to netem, and fix up HFSC.

Duplication of packets would be trivial to do with mirred mirror option
attached to the qdisc. Something along the lines of:

on device ethx \
match all packets (or whatever filter you want) \
action mirred mirror to device ethx

To duplicate packet more than once:
on device ethx \
match all packets (or whatever filter you want) \
action mirred mirror to device ethx \
action mirred mirror to device ethx

repeat for as many dups as you want

more funky?
randomly duplicate one or two packets

on device ethx \
match all packets (or whatever filter you want) \
action gact random stop here \
action mirred mirror to device ethx \
action gact random stop here \
action mirred mirror to device ethx

cheers,
jamal