From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michal Kubecek Subject: Re: [RFC PATCH 1/1] net/ipv4: Enable flow-based ECMP Date: Tue, 28 Jul 2015 09:20:52 +0200 Message-ID: <20150728072052.GA12945@unicorn.suse.cz> References: <55B6E8AC.4060600@alliedtelesis.co.nz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "netdev@vger.kernel.org" , "jmorris@namei.org" To: Richard Laing Return-path: Received: from mx2.suse.de ([195.135.220.15]:46032 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751775AbbG1HUz (ORCPT ); Tue, 28 Jul 2015 03:20:55 -0400 Content-Disposition: inline In-Reply-To: <55B6E8AC.4060600@alliedtelesis.co.nz> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Jul 28, 2015 at 02:27:57AM +0000, Richard Laing wrote: > From: Richard Laing > > Enable flow-based ECMP. > > Currently if equal-cost multipath is enabled the kernel chooses between > equal cost paths for each matching packet, essentially packets are > round-robined between the routes. This means that packets from a single > flow can traverse different routes. If one of the routes experiences > congestion this can result in delayed or out of order packets arriving > at the destination. > > This patch allows packets to be routed based on their > flow - packets in the same flow will always use the same route. This > prevents out of order packets. There are other issues with round-robin > based ECMP routing related to variable path MTU handling and debugging. > See RFC2991 for more details on the problems associated with packet > based ECMP routing. > > This patch relies on the skb hash value to select between routes. The > selection uses a hash-threshold algorithm (see RFC2992). > > Signed-off-by: Richard Laing The patch looks corrupted (long lines split, tabs converted to (four?) spaces etc. Michal Kubecek