From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [RFC PATCH nf-next 0/1] payload mangling support Date: Mon, 9 Nov 2015 16:00:07 +0000 Message-ID: <20151109160007.GL8098@macbook.localdomain> References: <1447078234-21628-1-git-send-email-kaber@trash.net> <20151109153029.GA7759@breakpoint.cc> <20151109154127.GK8098@macbook.localdomain> <20151109155014.GB7759@breakpoint.cc> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: pablo@netfilter.org, netfilter-devel@vger.kernel.org To: Florian Westphal Return-path: Received: from 161-169.trash.net ([213.144.137.169]:49253 "EHLO stinky.trash.net" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1751825AbbKIQAL (ORCPT ); Mon, 9 Nov 2015 11:00:11 -0500 Content-Disposition: inline In-Reply-To: <20151109155014.GB7759@breakpoint.cc> Sender: netfilter-devel-owner@vger.kernel.org List-ID: On 09.11, Florian Westphal wrote: > Patrick McHardy wrote: > > On 09.11, Florian Westphal wrote: > > > Patrick McHardy wrote: > > > > So the main questions is basically, do we want to support stateless NAT? > > > > The downside is that we have to add protocol specific checksumming functions. > > > > > > For now i'm only be interested in mangling MAC addresses for nft bridge > > > to provide ebtables snat/dnat/redirect equivalents. > > > > I have not tested that, it probably needs some minor adjustments to skip > > checksum updates for NFT_PAYLOAD_LL_HEADER, but other than that, this will > > obviously work. > > For redirect we also need to alter skb->pkttype to PACKET_HOST. > > I can take care of this later if needed, I'd just need advice on > how to add this (meta statement?). Good point. I guess that would be the proper solution, I don't think we want to try to figure out whether an address is local in the payload module.