From owner-freebsd-net Thu Jun 25 15:21:49 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA09993 for freebsd-net-outgoing; Thu, 25 Jun 1998 15:21:49 -0700 (PDT) (envelope-from owner-freebsd-net@FreeBSD.ORG) Received: from almond.elite.net (root@almond.elite.net [205.199.220.5]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA09882; Thu, 25 Jun 1998 15:21:01 -0700 (PDT) (envelope-from nate@almond.elite.net) Received: (from nate@localhost) by almond.elite.net (8.8.3/ELITE) id PAA28609; Thu, 25 Jun 1998 15:20:43 -0700 (PDT) From: Nate Lawson Message-Id: <199806252220.PAA28609@almond.elite.net> Subject: Re: Apparent bug in sendto() with raw sockets To: fenner@parc.xerox.com (Bill Fenner) Date: Thu, 25 Jun 1998 15:20:43 -0700 (PDT) Cc: nate@elite.net, fenner@parc.xerox.com, julian@whistle.com, freebsd-bugs@FreeBSD.ORG, freebsd-net@FreeBSD.ORG In-Reply-To: <98Jun25.150659pdt.177515@crevenia.parc.xerox.com> from "Bill Fenner" at Jun 25, 98 03:06:52 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In a previous message, Bill Fenner said: >In message <199806252129.OAA24992@almond.elite.net> Nate Lawson wrote: >>Since traceroute ships as a standard utility with FreeBSD, couldn't the >>change to network byte order be done simultaneously, minimizing headaches? > >The standard utilities could be changed easily. I'm worried about >compatibility with externally-written programs; e.g. the binary-only >"pathchar" from LBL would become useless on an OS with this change. >Externally-written programs that are written to the well known BSD raw >socket interface would all have to be patched to work with an updated >system, as well. (For example, LBL's traceroute program.) A friend of mine, jpm@elite.net, suggested that it might be possible to handle both orderings (like endian-switching on an Alpha). The only thing I see this would break is raw packets over 32768 bytes long (i.e. if ip_len > 32768, swap bytes and use the result). Might a sysctl variable also be an option? I know that 2.0.5R behaved the way that OpenBSD and Linux behave. Were there any complaints or problems with it back then? -Nate To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message