From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S935015AbXGTHhf (ORCPT ); Fri, 20 Jul 2007 03:37:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757965AbXGTHh2 (ORCPT ); Fri, 20 Jul 2007 03:37:28 -0400 Received: from one.firstfloor.org ([213.235.205.2]:51388 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757201AbXGTHh1 (ORCPT ); Fri, 20 Jul 2007 03:37:27 -0400 Date: Fri, 20 Jul 2007 09:37:26 +0200 From: Andi Kleen To: Zachary Amsden Cc: Mathieu Desnoyers , Andi Kleen , jbeulich@novell.com, "S. P. Prasanna" , linux-kernel@vger.kernel.org, patches@x86-64.org, Jeremy Fitzhardinge Subject: Re: new text patching for review Message-ID: <20070720073726.GB19833@one.firstfloor.org> References: <200707191105.44056.ak@suse.de> <20070719133852.GA5490@Krystal> <200707191546.08919.ak@suse.de> <20070719173502.GB12955@Krystal> <20070719234912.GB30383@Krystal> <46A00CC2.6010407@vmware.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <46A00CC2.6010407@vmware.com> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Jul 19, 2007 at 06:15:46PM -0700, Zachary Amsden wrote: > Mathieu Desnoyers wrote: > >Yes, kprobes is case 1: atomic update. And we don't even have to bother > >about Intel's erratum. This one is ok. That's mainly the > >alternatives/paravirt code I worry about. > > > > Paravirt and alternatives should all be ok because they are done before > SMP bringup and with NMIs disabled. NMI watchdog is not setup until > smp_prepare_cpus/check_nmi_watchdog, which happens way later, not during > parse_args/setup_nmi_watchdog, which just decides which type of watchdog > to setup. There could be other NMIs too - e.g. caused by IO devices or NMI buttons - but they're relatively unlikely. > > I originally considered the NMI problem for paravirt-ops patching done > during module load, and found that I would need to modify > stop_machine_run to have an architecture specific callout to mask and > unmask NMIs. When your virtual machine never injects NMIs or MCEs you're ok. -Andi