From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?UTF-8?B?c3TDqXBoYW5lIGJyeWFudA==?= Subject: queuing to userspace for the bridge family, using nftables Date: Sat, 11 Oct 2014 09:00:44 +0200 Message-ID: <5438D59C.1030706@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Cc: kaber@trash.net, pablo@netfilter.org, marc@petit-huguenin.org To: netfilter-devel@vger.kernel.org Return-path: Received: from mail-wi0-f174.google.com ([209.85.212.174]:38792 "EHLO mail-wi0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750973AbaJKHAx (ORCPT ); Sat, 11 Oct 2014 03:00:53 -0400 Received: by mail-wi0-f174.google.com with SMTP id cc10so3807161wib.1 for ; Sat, 11 Oct 2014 00:00:52 -0700 (PDT) Sender: netfilter-devel-owner@vger.kernel.org List-ID: Hello, We are currently working on a project which may need the use of nftables to queue AF_BRIDGE packets to the userspace and back. Apparently this is already supported in the nftables code itself, but not in the bridge netfilter code (the corresponding nf_afinfo is missing, at least). I would like to know if there are plans to support this? If yes, in what timeframe? If no, would you be interested if we were to submit a patch to that effect? Also, a. what should be included in the payload send to userspace? (starting from the 802.3 LLC/protocol header, or also including up to the MAC addresses ?) b. alternatively, shouldn't the dest MAC address be added to the netlink attributes sent to userspace? regards stephane bryant