On Wed, Nov 15, 2017 at 12:30:11PM +0100, Florian Westphal wrote:
> Sebastian Gottschall <s.gottschall@xxxxxxxxxx> wrote:
> > your patch is not clean
>
> Please stop top-posting.
>
> > patching file net/netfilter/nf_nat_core.c
> > Hunk #7 succeeded at 550 (offset 6 lines).
> > Hunk #8 FAILED at 672.
> > Hunk #9 succeeded at 819 (offset 10 lines).
> > Hunk #10 succeeded at 852 (offset 10 lines).
> > Hunk #11 succeeded at 872 (offset 10 lines).
> >
> > nf_nat_cleanup_conntrack is wrong
> >
> > your patch
> >
> > @@ -672,9 +645,11 @@ EXPORT_SYMBOL_GPL(nf_nat_l3proto_unregister);
> > /* No one using conntrack by the time this called. */
> > static void nf_nat_cleanup_conntrack(struct nf_conn *ct)
> > {
> > - if (ct->status & IPS_SRC_NAT_DONE)
> > - rhltable_remove(&nf_nat_bysource_table, &ct->nat_bysource,
> > - nf_nat_bysource_params);
> > + if (ct->status & IPS_SRC_NAT_DONE) {
> > + spin_lock_bh(&nf_nat_lock);
> > + hlist_del_rcu(&ct->nat_bysource);
> > + spin_unlock_bh(&nf_nat_lock);
> > + }
> > }
>
> It assumes
> git cherry-pick 6e699867f84c0f358fed233fe6162173aca28e04
>
> to get the IPS_SRC_NAT_DONE test.
Can you all submit a "clean" patch series against the next 4.9-stable
release I do in an hour?
thanks,
greg k-h