From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262731AbVAKAxV (ORCPT ); Mon, 10 Jan 2005 19:53:21 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262617AbVAJVc5 (ORCPT ); Mon, 10 Jan 2005 16:32:57 -0500 Received: from mail.tmr.com ([216.238.38.203]:32274 "EHLO gatekeeper.tmr.com") by vger.kernel.org with ESMTP id S262602AbVAJVQz (ORCPT ); Mon, 10 Jan 2005 16:16:55 -0500 To: linux-kernel@vger.kernel.org Path: not-for-mail From: Bill Davidsen Newsgroups: mail.linux-kernel Subject: Whatever happened to the real mode driver idea? Date: Mon, 10 Jan 2005 16:19:31 -0500 Organization: TMR Associates, Inc Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Trace: gatekeeper.tmr.com 1105391205 12457 192.168.12.100 (10 Jan 2005 21:06:45 GMT) X-Complaints-To: abuse@tmr.com User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5) Gecko/20041217 X-Accept-Language: en-us, en Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org There has been some discussion of some "universal real mode" driver to allow hardware without an open source driver to operate with Linux. It's been about a year since this came up, and as I recall the last discussion ended with people opposed to using that as a reverse engineering tool, due to possible legal issues. But I have a technical question, wouldn't a real mode use of BIOS code be more secure that the current situation of tainted drivers running in the kernel? Please take political discussion elsewhere, I'm curious about the safty issue, not the legal or advocacy issues. Being an old MULTICS guy, my first though when protection rings came out was that the drivers should be out of ring zero. -- -bill davidsen (davidsen@tmr.com) "The secret to procrastination is to put things off until the last possible moment - but no longer" -me