From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752599Ab1HXXEc (ORCPT ); Wed, 24 Aug 2011 19:04:32 -0400 Received: from terminus.zytor.com ([198.137.202.10]:54334 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751821Ab1HXXEa (ORCPT ); Wed, 24 Aug 2011 19:04:30 -0400 Message-ID: <4E55836D.8030209@zytor.com> Date: Wed, 24 Aug 2011 16:04:13 -0700 From: "H. Peter Anvin" User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:6.0) Gecko/20110816 Thunderbird/6.0 MIME-Version: 1.0 To: Vivien Didelot CC: linux-kernel@vger.kernel.org, x86@kernel.org, Thomas Gleixner , Ingo Molnar Subject: Re: [PATCH 1/4] x86: base support for the TS-5500 platform References: <1314217407-31249-1-git-send-email-vivien.didelot@savoirfairelinux.com> <1314217407-31249-2-git-send-email-vivien.didelot@savoirfairelinux.com> <4E55670F.90102@zytor.com> <20110824180837.6841f3d8@v0nbox> <4E5576CF.2050905@zytor.com> <20110824183814.64abc94c@v0nbox> In-Reply-To: <20110824183814.64abc94c@v0nbox> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/24/2011 03:38 PM, Vivien Didelot wrote: > > I agree with you. The platform does not provide many identification > points. We have specific hardware such as the CPU, Ethernet, Cardbus > and USB controllers; there is no DMI available; no PCI bridge subvendor > info, only the BIOS provides a 16-bit identification call... > We discussed that point in https://lkml.org/lkml/2011/5/4/437 without > success. > > What would be the proper way in your opinion? > > Vivien. First of all, given that these clowns seem to be selling a product specifically targeting Linux support, they really need to be given a clue transfusion. It is *NOT* acceptable to not provide DMI, and *NOT* acceptable not to set the subvendor information. We should arguably push for them to fix that in their BIOSes, but that is as usual too little, too late. Probably the best option we have is to look for some kind of magic string in their BIOS -- i.e. the classic way to detect things before DMI existed. If not, you're going to have an explicit switch on the kernel command line, but that is an absolute last resort. -hpa