On 17 October 2015 at 13:14, Florian Westphal <fw@xxxxxxxxx> wrote:
> [ CC netdev since patch #2 isn't nf-specific. Dave, if you want
> I can resubmit that one after the next nf-pull request; let me know if
> you would prefer that ].
>
> Openvswitch seems broken wrt. to defragmentation, it doesn't call
> nf_ct_frag6_consume_orig to free the original fragments.
This will need to be fixed for 'net' as well, do you have a path in
mind for that?
> Moreover, openvswitch design seems to require that it can reuse current
> skb rather than work with a new skb pointer (it uses skb_morph for this).
>
> Instead of OVS-side fix this series tries to alter netfilter ipv6 defrag
> accordingly.
>
> 1. nf_ct_frag6_consume_orig is removed, since
> commit 6aafeef03b9d9ecf ("netfilter: push reasm skb through instead of
> original frag skbs") nothing needs the original fragments so there is no
> reason why we need to clone+store original skb -- just stash original
> skbs in the frag_list.
>
> 2. Use skb_morph to make the last skb processed (not necessarily last
> fragment) the reassembled one.
>
> 3. remove the no-longer needed recursion into nf_iterate, we can now just
> return ACCEPT/STOLEN as needed instead of NF_HOOK_THRESH()+NF_STOLEN.
>
> Tested with flood-ping6+ fault-injection framework 'failslab' type.
Overall, I like it in that it seems to make the IPv6 code more
consistent with IPv4, and it's a net negative LoC, simplifying the
code.
Patch 3 when taken independently from patch 4 hides user-visible error
codes on the OVS side. The OVS conntrack action hides -EINPROGRESS
from userspace, treating it as a successful execution. All other
errors are returned up. With that patch, all errors will be hidden. I
see that it's fixed in Patch 4, so maybe it's not a biggie but those
two patches should be tightly coupled.
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
