On Mon, Jul 21, 2008 at 06:47:22PM +0200, Jan Engelhardt wrote:
quoted text
> On Monday 2008-07-21 18:29, Alexey Dobriyan wrote:
> >-			      int (*okfn)(struct sk_buff *))
> >+static unsigned int arpt_in_hook(unsigned int hook,
> >+				 struct sk_buff *skb,
> >+				 const struct net_device *in,
> >+				 const struct net_device *out,
> >+				 int (*okfn)(struct sk_buff *))
> > {
> >-	return arpt_do_table(skb, hook, in, out, init_net.ipv4.arptable_filter);
> >+	return arpt_do_table(skb, hook, in, out,
> >+			     dev_net(in)->ipv4.arptable_filter);
> >+}
> 
> What I dislike with this netns stuff is that you need to touch
> the ->ipv4. structure if you want to add new tables and there is
> no more way to dynamically add tables from 3rd party modules unless
> the admin loading them is fine with having them register against
> &init_net only.

Not exactly sure, if it's sufficient for you, you can use
net_assign_generic() infrastructure from module and not touch "struct net"
at all.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html