From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-next 2/2] flow_dissector: add tipc support Date: Mon, 26 Jan 2015 17:04:41 -0800 (PST) Message-ID: <20150126.170441.724106999176865818.davem@davemloft.net> References: <1421943032-29924-1-git-send-email-erik.hugne@ericsson.com> <1421943032-29924-2-git-send-email-erik.hugne@ericsson.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: richard.alpe@ericsson.com, netdev@vger.kernel.org, jon.maloy@ericsson.com, ying.xue@windriver.com, tipc-discussion@lists.sourceforge.net To: erik.hugne@ericsson.com Return-path: Received: from shards.monkeyblade.net ([149.20.54.216]:46436 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757396AbbA0BEm (ORCPT ); Mon, 26 Jan 2015 20:04:42 -0500 In-Reply-To: <1421943032-29924-2-git-send-email-erik.hugne@ericsson.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Date: Thu, 22 Jan 2015 17:10:32 +0100 > From: Erik Hugne > > The flows are hashed on the sending node address, which allows us > to spread out the TIPC link processing to RPS enabled cores. There > is no point to include the destination address in the hash as that > will always be the same for all inbound links. We have experimented > with a 3-tuple hash over [srcnode, sport, dport], but this showed to > give slightly lower performance because of increased lock contention > when the same link was handled by multiple cores. > > Signed-off-by: Ying Xue > Signed-off-by: Erik Hugne > Reviewed-by: Jon Maloy Applied.