From: Michal Kubecek Subject: Re: [PATCH v3 23/30] ipv6: defrag: drop non-last frags smaller than min mtu Date: Fri, 11 Jan 2019 13:21:16 +0100 Message-ID: <20190111122116.GF14607@unicorn.suse.cz> References: <20180913145902.17531-1-sthemmin@microsoft.com> <20180913145902.17531-24-sthemmin@microsoft.com> <20190110222231.w7fyfgaepbhuh2fi@breakpoint.cc> <69e28b74-fe8b-67dc-f290-44c2c0370c85@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Florian Westphal , Tom Herbert , Linux Kernel Network Developers , "David S. Miller" , Eric Dumazet , Peter Oskolkov , Timothy Winters To: Eric Dumazet Return-path: Received: from mx2.suse.de ([195.135.220.15]:48294 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1730724AbfAKMVT (ORCPT ); Fri, 11 Jan 2019 07:21:19 -0500 Content-Disposition: inline In-Reply-To: <69e28b74-fe8b-67dc-f290-44c2c0370c85@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, Jan 11, 2019 at 02:57:39AM -0800, Eric Dumazet wrote: > On 01/10/2019 02:22 PM, Florian Westphal wrote: > > Tom Herbert wrote: > >> I couldn't find any mention of the advisory in the commit logs or > >> netdev discussion, and apparently there's no protocol requirement that > >> intermediate fragements need to be at least minimal MTU. Maybe this > >> patch should be reverted? > > > > Currently ipv6 reasm doesn't use rbtree infrastructure, so it would > > have to be converted first. > > > Section 4.5 of RFC 8200 allows for sending any fragment for > fragments as long they add up to the original packet. > > > I do not believe we need an rbtree to implement this idea. IMHO Florian meant that allowing arbitrarily small fragments would harm resistance against FragmentSmack type attacks so that we might need rbtree based queues to be reasonably safe. Michal Kubecek