From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH 2/2] packet: Add fanout support. Date: Tue, 05 Jul 2011 09:08:09 -0700 (PDT) Message-ID: <20110705.090809.1320878912859271524.davem@davemloft.net> References: <4E12B5A6.2020802@inliniac.net> <1309849214.2720.45.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: eric.dumazet@gmail.com, victor@inliniac.net, netdev@vger.kernel.org To: Chetan.Loke@netscout.com Return-path: Received: from shards.monkeyblade.net ([198.137.202.13]:57591 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754146Ab1GEQIY (ORCPT ); Tue, 5 Jul 2011 12:08:24 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: From: "Loke, Chetan" Date: Tue, 5 Jul 2011 12:03:29 -0400 > Can we not do something like: > > a = src_ip_addr; > b = dst_ip_addr; > > if (ip_is_fragment(ip_hdr(skb))) > c = ip_hdr->id; > else > c = src_port | dest_port ; /* port_32 etc - Similar to what we have today A UDP flow can be composed of fragmented and non-fragmented parts, we want all of the packets from that flow to land on the same hash. Your scheme does not provide that essential property.