From owner-freebsd-net@FreeBSD.ORG Wed Jan 7 14:56:15 2004 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E9E6A16A4CE; Wed, 7 Jan 2004 14:56:15 -0800 (PST) Received: from khavrinen.lcs.mit.edu (khavrinen.lcs.mit.edu [18.24.4.193]) by mx1.FreeBSD.org (Postfix) with ESMTP id B933743D55; Wed, 7 Jan 2004 14:56:14 -0800 (PST) (envelope-from wollman@khavrinen.lcs.mit.edu) Received: from khavrinen.lcs.mit.edu (localhost.nic.fr [IPv6:::1]) by khavrinen.lcs.mit.edu (8.12.9/8.12.9) with ESMTP id i07MuDDa066943 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK CN=khavrinen.lcs.mit.edu issuer=SSL+20Client+20CA); Wed, 7 Jan 2004 17:56:13 -0500 (EST) (envelope-from wollman@khavrinen.lcs.mit.edu) Received: (from wollman@localhost) by khavrinen.lcs.mit.edu (8.12.9/8.12.9/Submit) id i07MuCnH066940; Wed, 7 Jan 2004 17:56:12 -0500 (EST) (envelope-from wollman) Date: Wed, 7 Jan 2004 17:56:12 -0500 (EST) From: Garrett Wollman Message-Id: <200401072256.i07MuCnH066940@khavrinen.lcs.mit.edu> To: Andre Oppermann In-Reply-To: <3FFC8CBE.13B527A3@freebsd.org> References: <200401072023.UAA18922@starburst.demon.co.uk> <3FFC8CBE.13B527A3@freebsd.org> X-Spam-Score: -19.8 () IN_REP_TO,QUOTED_EMAIL_TEXT,REFERENCES,REPLY_WITH_QUOTES X-Scanned-By: MIMEDefang 2.37 cc: freebsd-net@FreeBSD.ORG Subject: Re: kern/60889 - zero IP id change issues in 5.2RC2 X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2004 22:56:16 -0000 < said: > 1. Do you think it is neccessary to do a htons() on the randomized > ip_id too? I'd say yes if there is a case where it has to > monotonically increase afterwards. Does it? IP IDs are nonces. The only requirement is that they not be reused for a packet to the same destination IP address before reassembly has timed out. In practice, this is impossible to guarantee, so the usual practice is to try to ensure a complete cycle (of 16-bit numbers). The order doesn't matter. I prefer not byte-swapping the address, but with a modern processor that overhead can probably be pipelined out anyway. -GAWollman