From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753908AbZDMQ3W (ORCPT ); Mon, 13 Apr 2009 12:29:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753105AbZDMQ3E (ORCPT ); Mon, 13 Apr 2009 12:29:04 -0400 Received: from terminus.zytor.com ([198.137.202.10]:49466 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752902AbZDMQ3D (ORCPT ); Mon, 13 Apr 2009 12:29:03 -0400 Message-ID: <49E367E8.7080202@zytor.com> Date: Mon, 13 Apr 2009 09:27:20 -0700 From: "H. Peter Anvin" User-Agent: Thunderbird 2.0.0.14 (X11/20080501) MIME-Version: 1.0 To: Ingo Molnar CC: Avi Kivity , Linus Torvalds , Pavel Machek , mingo@redhat.com, linux-kernel@vger.kernel.org, tglx@linutronix.de, hpa@linux.intel.com, rjw@sisk.pl, linux-tip-commits@vger.kernel.org Subject: Re: [tip:x86/setup] x86, setup: "glove box" BIOS calls -- infrastructure References: <20090410103934.GA21506@elte.hu> <20090410104648.GA31516@elf.ucw.cz> <20090410112546.GD21506@elte.hu> <20090410113824.GA18823@elf.ucw.cz> <49E0C1AB.2050608@redhat.com> <49E17A6E.5000104@zytor.com> <20090412163356.GA2392@elte.hu> <49E2398A.3050405@redhat.com> <20090413041625.GF11652@elte.hu> <20090413042459.GA6479@elte.hu> In-Reply-To: <20090413042459.GA6479@elte.hu> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Ingo Molnar wrote: >> Yes, we could do memory checks, and ... hey, we already do that: >> >> bb577f9: x86: add periodic corruption check >> 5394f80: x86: check for and defend against BIOS memory corruption >> >> ... and i seem to be the one who implemented it! ;-) > > s/implemented/merged+fixed :-) Actually, what would probably be more productive than trying to track corruption would be to drop the low 1 MB of memory before suspend to RAM - make sure that it is as close to completely unused as possible. All *known* cases of low memory corruption are either boot time or due to s2ram. I don't know how realistic it is to make the low 1 MB completely unused over the s2ram cycle. The trivial way of doing it is to simply not use it -- it's only some 600K after all; a more sophisticated way would be to explicitly constrain it to transient uses that would be dead at s2ram. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.