From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pablo Neira Ayuso Subject: Re: [PATCH nft] src: allow to specify the base chain type Date: Tue, 27 Aug 2013 00:37:32 +0200 Message-ID: <20130826223732.GA4031@localhost> References: <1377185191-9961-1-git-send-email-pablo@netfilter.org> <52171199.5090700@linux.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netfilter-devel@vger.kernel.org To: Tomasz Bursztyka Return-path: Received: from mail.us.es ([193.147.175.20]:51836 "EHLO mail.us.es" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752373Ab3HZWhw (ORCPT ); Mon, 26 Aug 2013 18:37:52 -0400 Content-Disposition: inline In-Reply-To: <52171199.5090700@linux.intel.com> Sender: netfilter-devel-owner@vger.kernel.org List-ID: On Fri, Aug 23, 2013 at 10:39:05AM +0300, Tomasz Bursztyka wrote: [...] > > add table mangle > > add chain mangle OUTPUT { type route hook NF_INET_LOCAL_OUT 0; } > > Instead of NF_INET_LOCAL_OUT could we get OUT? (not literally ;) ) makes sense. > IN, OUT, PRE-ROUTING, FORWARD, POST-ROUTING etc... I prefer: input, output, prerouting, forward, postrouting. > And depending on chain's family, nft would use the right value there > (IN: is NF_INET_LOCAL_IN for ipv4, NF_ARP_IN for arp, etc...) > > It would also make command line easier and more readable. > > One more debatable: > What about adding prio keyword? So it would be { type hook > prio } > Or keeping both possibility or prio ? it also makes sense to me, but that should come in a separated patch. > I can prepare a patch if you want, Go ahead. Thanks.