From owner-freebsd-security@FreeBSD.ORG Fri Aug 11 10:28:44 2006 Return-Path: X-Original-To: freebsd-security@freebsd.org Delivered-To: freebsd-security@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id EAC4816A4DA for ; Fri, 11 Aug 2006 10:28:44 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8C83943D55 for ; Fri, 11 Aug 2006 10:28:44 +0000 (GMT) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id 3491546C1D; Fri, 11 Aug 2006 06:28:43 -0400 (EDT) Date: Fri, 11 Aug 2006 11:28:43 +0100 (BST) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: Poul-Henning Kamp In-Reply-To: <38802.1155288265@critter.freebsd.dk> Message-ID: <20060811112511.T45647@fledge.watson.org> References: <38802.1155288265@critter.freebsd.dk> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-security@freebsd.org, =?ISO-8859-1?Q?Jos=E9_M=2E_Fandi=F1o?= Subject: Re: atheros chips dangerous? X-BeenThere: freebsd-security@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Security issues \[members-only posting\]" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Aug 2006 10:28:45 -0000 On Fri, 11 Aug 2006, Poul-Henning Kamp wrote: >> In my opinion the difference is that with NDA you place trust in a few >> persons (the ones with the code), whilst with open source drivers the code >> can be reviewed by all people with enough knowledge about the subject and >> since peer review is an important concept in FOSS quality (and security) it >> would be desirable to have free code. > > While that is certainly true, I also feel that the fact that Atheros has > actively tried to work with the FOSS people to get a good driver should be > credited to them. > > Other vendors have been totally impossible to work with. Something worth observing here is that many modern device drivers, especially more complex cards with significant offload of functionality to the card, have closed source components -- the firmware for the device. The HAL is a tiny wrapper around programming of a few very specific elements of the hardware behavior to do with software radio power/frequency, etc. Compared to the size of the closed source chunk in the firmware of many device drivers (ipw, many RAID controllers, etc, for example), it is miniscule, and is reviewed and maintained by an open source person. You could argue that this is significantly more forthcoming than many other vendors, for whom firmware binaries are entirely closed source. Robert N M Watson Computer Laboratory University of Cambridge