On Sat, Jul 09, 2016 at 01:30:38AM +0200, Florian Westphal wrote:
> Aaron Conole <acon...@bytheb.org> wrote:
> > --- a/net/netfilter/core.c
> > +++ b/net/netfilter/core
> [..]
> > +#define nf_entry_dereference(e) \
> > +   rcu_dereference_protected(e, lockdep_is_held(&nf_hook_mutex))
> >  
> > -static struct list_head *nf_find_hook_list(struct net *net,
> > -                                      const struct nf_hook_ops *reg)
> > +static struct nf_hook_entry *nf_find_hook_list(struct net *net,
> > +                                          const struct nf_hook_ops *reg)
> >  {
> > -   struct list_head *hook_list = NULL;
> > +   struct nf_hook_entry *hook_list = NULL;
> >  
> >     if (reg->pf != NFPROTO_NETDEV)
> > -           hook_list = &net->nf.hooks[reg->pf][reg->hooknum];
> > +           hook_list = rcu_dereference(net->nf.hooks[reg->pf]
> > +                                       [reg->hooknum]);
> >     else if (reg->hooknum == NF_NETDEV_INGRESS) {
> >  #ifdef CONFIG_NETFILTER_INGRESS
> >             if (reg->dev && dev_net(reg->dev) == net)
> > -                   hook_list = &reg->dev->nf_hooks_ingress;
> > +                   hook_list =
> > +                           rcu_dereference(reg->dev->nf_hooks_ingress);
> 
> Both of these should use nf_entry_dereference() to avoid the lockdep
> splat reported by kbuild robot:
> 
> net/netfilter/core.c:75 suspicious rcu_dereference_check() usage!
> 2 locks held by swapper/1:
> #0:  (rtnl_mutex){+.+.+.}, at: [<ffffffff81c2e567>] rtnl_lock+0x17/0x20
> #1:  (nf_hook_mutex){+.+...}, at: [<ffffffff81c58fcb>] 
> nf_register_net_hook+0xcb/0x240

Aaron, please, send a v2.

I have a patchset that changes the footprint of the hook function as
it was discussed during the last Netfilter Workshop that clashes with
this.

Thanks!

Reply via email to