Re: [PATCH net-next 3/6] flow_dissector: Add hash_extra field to flow_keys struct

[Florian Westphal <fw@strlen.de>]

Tom Herbert <therbert@google.com> wrote:
> On Sat, Feb 28, 2015 at 12:31 PM, Florian Westphal <fw@strlen.de> wrote:
> > I've been toying around with reducing skb->cb[] to 44 bytes,
> > Seems Tom could integrate following patch from my test branch:
> >
> > http://git.breakpoint.cc/cgit/fw/net-next.git/commit/?h=skb_cb_44_01&id=29d711e1a71244b71940c2d1e346500bef4d6670
> >
> > It makes sfq use a smaller flow key state.
> 
> Alternatively, I think we might be able to eliminate the use of
> flow_keys and flow_dissect from the qdisc code altogether. It looks
> like this is only being used to determine a hash over the addresses,
> ports, and protocol so I am thinking that we can just call
> skb_get_hash for that. Will try to post some patches soon.

The problem with this is that you'll lose the secret input to jhash
in sfq_hash().

assuming you have packets p1 and p2 (from different flows)
with skb_get_hash(p1) == skb_get_hash(p2) those flows share same
queue/bin forever as the hash pertubation will no longer work.

For sfq, hash collisions may exist as well but they'll be resolved
after some time when q->perturbation (its part of hash input) is reseeded.