From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: relationship between ingress policer and netfilter hooks? Date: Wed, 03 Jun 2009 09:18:55 +0200 Message-ID: <4A2623DF.5040800@trash.net> References: <4A25457F.2070000@nortel.com> <4A255BAC.8070406@trash.net> <4A256E55.9090309@nortel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: Chris Friesen Return-path: Received: from stinky.trash.net ([213.144.137.162]:59166 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751713AbZFCHS4 (ORCPT ); Wed, 3 Jun 2009 03:18:56 -0400 In-Reply-To: <4A256E55.9090309@nortel.com> Sender: netdev-owner@vger.kernel.org List-ID: Chris Friesen wrote: > Patrick McHardy wrote: >> Chris Friesen wrote: >> >>> Hi all, >>> >>> Can someone tell me where the hook is for incoming packets to be sent >>> through the ingress policer? >>> >>> In particular, I'm trying to find out how the ingress traffic shaping >>> hook relates to the various netfilter hooks so that I can set them up to >>> behave nicely together. > >> They're not related at all. The ingress qdisc is hooked at >> net/dev/core.c:handle_ing(). > > Thanks. Based on that the shaping is done before any of the netfilter > hooks, which is what I was trying to figure out. Correct. > On a related note...the deliver_skb() call right before the call to > handle_ing()--is that to allow packet sockets to receive everything that > arrives on the wire even if some of it may be dropped later? Yes, its the final delivery for ptype_all handlers.