Return-path: Received: from mail-wm0-f42.google.com ([74.125.82.42]:36240 "EHLO mail-wm0-f42.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752278AbcDSLaX convert rfc822-to-8bit (ORCPT ); Tue, 19 Apr 2016 07:30:23 -0400 Received: by mail-wm0-f42.google.com with SMTP id v188so157922456wme.1 for ; Tue, 19 Apr 2016 04:30:22 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <1461063747.2766.22.camel@sipsolutions.net> References: <1455710131-9967-1-git-send-email-grzegorz.bajorski@tieto.com> <56C4A5ED.4060802@openwrt.org> <1461057103.2766.15.camel@sipsolutions.net> <1461063747.2766.22.camel@sipsolutions.net> Date: Tue, 19 Apr 2016 13:30:21 +0200 Message-ID: (sfid-20160419_133030_880053_D339EEC4) Subject: Re: [RFC] mac80211: add extap functionality From: Michal Kazior To: Johannes Berg Cc: Felix Fietkau , Grzegorz Bajorski , linux-wireless Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 19 April 2016 at 13:02, Johannes Berg wrote: >> > How much of that could be done with nftables btw? > >> I'm not sure if I follow. Do you mean what I've been able to do with >> relayd until now? Without link-local ipv6 routing DHCPv6 is broken >> (could probably addressed with DHCPv6 Relay to a certain degree) and >> RS/RA may not work (if it propagates fe80:: routes). Also apps that >> rely on fe80:: socket binding/addressing will fail. > > Ok, so that makes sense I guess - but you were speaking of some packet > mangling etc. and I was wondering if the nftables virtual machine could > actually do something like that. By mangling I meant packet payload needs to be modified in various ways (typically just ether_dest/src, but ARP/NS/NA/DHCP needs extra care). I don't think you can force a link-local packet to switch interfaces through nftables anyway (or can you?). Even if you could I guess you could argue it's a bug? Michał