From owner-freebsd-current Mon Jan 20 1:19: 3 2003 Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 86CAC37B401; Mon, 20 Jan 2003 01:19:00 -0800 (PST) Received: from cheer.mahoroba.org (flets19-241.kamome.or.jp [218.45.19.241]) by mx1.FreeBSD.org (Postfix) with ESMTP id E9E5143F18; Mon, 20 Jan 2003 01:18:52 -0800 (PST) (envelope-from ume@mahoroba.org) Received: from localhost (IDENT:Uu30KmDdsELc59arwgaUpF3PNz7V8l1yca5dS7JWTWT7JeH639ZQSTwn9Tsl6ozi@localhost [IPv6:::1]) (user=ume mech=CRAM-MD5 bits=0) by cheer.mahoroba.org (8.12.7/8.12.7) with ESMTP/inet6 id h0K9IfRv092328 (version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NO); Mon, 20 Jan 2003 18:18:41 +0900 (JST) (envelope-from ume@mahoroba.org) Date: Mon, 20 Jan 2003 18:18:41 +0900 Message-ID: From: Hajimu UMEMOTO To: David Malone Cc: current@FreeBSD.org, emulation@FreeBSD.org Subject: Re: [CFR] IPv6 support for Linux sym In-Reply-To: <20030119195957.GA54865@walton.maths.tcd.ie> References: <20030119195957.GA54865@walton.maths.tcd.ie> User-Agent: xcite1.38> Wanderlust/2.10.0 (Venus) SEMI/1.14.4 (Hosorogi) FLIM/1.14.4 (=?ISO-8859-4?Q?Kashiharajing=FE-mae?=) APEL/10.4 Emacs/21.2 (i386--freebsd) MULE/5.0 (=?ISO-2022-JP?B?GyRCOC1MWhsoQg==?=) X-Operating-System: FreeBSD 4.7-STABLE MIME-Version: 1.0 (generated by SEMI 1.14.4 - "Hosorogi") Content-Type: text/plain; charset=US-ASCII X-Virus-Scanned: by AMaViS-perl11-milter (http://amavis.org/) X-Spam-Status: No, hits=-1.3 required=5.0 tests=IN_REP_TO,REFERENCES,SPAM_PHRASE_01_02,USER_AGENT version=2.43 Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Hi, >>>>> On Sun, 19 Jan 2003 19:59:57 +0000 >>>>> David Malone said: dwmalone> I have a patch which is quite similar to this, but also reduces the dwmalone> use of the stackgap in the Linux networking code and breaks out dwmalone> things like socket, bind and listen into an implimentation and a dwmalone> syscall wrapper (extending Ian's work in this area). I took these dwmalone> patches far enough to get the linux version of telnet to speak IPv6. Your approach of having syscall wrapper reduces the overhead of copyout(), and seems good. I thought that doing copyout() is slightly lengthy. However, only connect() and bind() have an wrapper. We need more wrappers for converting `to' field of sendto() and `msg_name' field of sendmsg(). These are not simple than the wrapper for connect() and bind(). I have no idea how these should be wrapped at the moment. dwmalone> 1) I hadn't figured out the best way to deal with syscalls dwmalone> which copyout a sockaddr - I suspect we need a per-sysvec dwmalone> sockaddr copyout function to do this cleanly, but I'm open dwmalone> to suggestion. Maybe yes. dwmalone> 2) The patches wouldn't compile on the alpha 'cos of some dwmalone> ifdefs in the linux emulation code. I just hadn't got around dwmalone> to resolving this. Oops, thank you for pointing this out. It seems that other than linux_connect() are ifdef'ed by __alpha__. Is it right? Is there any problem removing #ifdef __alpha__ and compiling the codes in on FreeBSD/Alpha? dwmalone> Maybe we could see if we could combine these patches? My patch is dwmalone> available at: dwmalone> http://www.maths.tcd.ie/~dwmalone/linux-socket I think so, too. Sincerely, -- Hajimu UMEMOTO @ Internet Mutual Aid Society Yokohama, Japan ume@mahoroba.org ume@bisd.hitachi.co.jp ume@{,jp.}FreeBSD.org http://www.imasy.org/~ume/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message