From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vlad Yasevich Subject: Re: [RFC] sctp/tcp: Question -- ICMPv4 length check (not) redundant? Date: Wed, 30 Jul 2008 08:49:38 -0400 Message-ID: <48906362.2090902@hp.com> References: <488E0192.9030406@hp.com> <20080728174432.GA15892@gerrit.erg.abdn.ac.uk> <488E0B73.8070701@hp.com> <20080730.031935.61479223.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: gerrit@erg.abdn.ac.uk, netdev@vger.kernel.org To: David Miller Return-path: Received: from g1t0028.austin.hp.com ([15.216.28.35]:47459 "EHLO g1t0028.austin.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754380AbYG3Mto (ORCPT ); Wed, 30 Jul 2008 08:49:44 -0400 In-Reply-To: <20080730.031935.61479223.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: David Miller wrote: > From: Vlad Yasevich > Date: Mon, 28 Jul 2008 14:09:55 -0400 > >> The IPv6 checks are much different. The MUST requirement is to provide as much >> data as possible upto IPv6 min mtu. So, the IPv6 icmp code should probably look >> to see if min(payload_len, min_mtu) is provided. > > "Be liberal in what you accept..." -Jon Postel Yeah, that check is really too strict. > > Also, do you know any case where 576 bytes won't provide enough > quoted packet bytes? :-) > Yes, I've seen a case where a full 1280 was needed. It involved a very convoluted packet that had extension headers and then went through 2 encapsulations. This was an extreme test scenario and not a real world example. -vlad