From owner-freebsd-net@FreeBSD.ORG Sat Aug 6 15:34:29 2005 Return-Path: X-Original-To: freebsd-net@freebsd.org 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 F2CF516A41F for ; Sat, 6 Aug 2005 15:34:28 +0000 (GMT) (envelope-from gbergling@0xfce3.net) Received: from generic.0xfce3.net (generic.0xfce3.net [62.75.158.227]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9786F43D5C for ; Sat, 6 Aug 2005 15:34:27 +0000 (GMT) (envelope-from gbergling@0xfce3.net) Received: from port-212-202-34-189.dynamic.qsc.de ([212.202.34.189] helo=[192.168.1.7]) by vs158227.vserver.de with esmtpa (Exim 4.50) id 1E1Qgm-0006vx-90; Sat, 06 Aug 2005 17:34:20 +0200 Message-ID: <42F4D862.6040002@0xfce3.net> Date: Sat, 06 Aug 2005 17:33:54 +0200 From: Gordon Bergling User-Agent: Mozilla Thunderbird 1.0.6 (Macintosh/20050716) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Bruce M Simpson References: <42F4A261.9060207@0xfce3.net> <20050806121504.GC740@empiric.icir.org> In-Reply-To: <20050806121504.GC740@empiric.icir.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-net@freebsd.org Subject: Re: [RFC] Port of DfBSDs new IP Checksum Code X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 06 Aug 2005 15:34:29 -0000 Hi Bruce, Bruce M Simpson wrote: >On Sat, Aug 06, 2005 at 01:43:29PM +0200, Gordon Bergling wrote: > > >>I have ported the new IP Checksum Code from DragonflyBSD to FreeBSD. >>I have a patch available at >>http://generic.0xfce3.net/freebsd/ipchecksum/20050805-ipchecksum.patch >> >> > >There are inline i386 instructions in a machine-independent header; is this >intentional? Also the asm_ones32() code seems to exist only for i386. > > I can't say something about the internals of the rework of checksum code. I have the commit messages on the hand, but I can't find something related. >Without support for other platforms, and some cleanup, I don't think that >the patch can be committed as is, sadly. > > As far as I understand the commit messages [1 - 3] correctly the new code should be machine independent. What cleanup do you mean? >I do like the general idea of rewriting in_cksum() to take advantage of >branch prediction logic in modern microprocessors, however. > > I am willing to spend more time on this and, if anyone could point me to the right direction regarding the next steps to get this into the tree. regards, Gordon [1] http://leaf.dragonflybsd.org/mailarchive/commits/2004-02/msg00187.html [2] http://leaf.dragonflybsd.org/mailarchive/commits/2004-02/msg00198.html [3] http://leaf.dragonflybsd.org/mailarchive/commits/2004-02/msg00200.html