Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755755Ab1EPUdx (ORCPT ); Mon, 16 May 2011 16:33:53 -0400 Received: from mx1.redhat.com ([209.132.183.28]:59140 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755575Ab1EPUdw (ORCPT ); Mon, 16 May 2011 16:33:52 -0400 Subject: Re: [PATCH v2] x86, vt-d: enable x2apic opt out From: Alex Williamson To: "Woodhouse, David" Cc: "Song, Youquan" , "linux-kernel@vger.kernel.org" , "akpm@linux-foundation.org" , "mingo@elte.hu" , "tglx@linutronix.de" , "hpa@zytor.com" , "hpa@linux.intel.com" , "Kay, Allen M" , "Siddha, Suresh B" , "Liu, Kent" , Youquan Song In-Reply-To: <1305126431.30435.163.camel@i7.infradead.org> References: <1302764783-24491-1-git-send-email-youquan.song@intel.com> <1305126431.30435.163.camel@i7.infradead.org> Content-Type: text/plain; charset="UTF-8" Date: Mon, 16 May 2011 14:32:57 -0600 Message-ID: <1305577977.3146.112.camel@x201> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1562 Lines: 32 On Wed, 2011-05-11 at 16:07 +0100, Woodhouse, David wrote: > On Thu, 2011-04-14 at 08:06 +0100, Song, Youquan wrote: > > New version of VT-d2 specification (http://download.intel.com/technology > > /computing/vptech/Intel(r)_VT_for_Direct_IO.pdf) includes a new feature that > > provide firmware a way to request system software to opt out of enable x2APIC > > mode. DMAR ACPI table newly define flags.1 bit: x2APIC_OPT_OUT which is set to > > request System software opt out xAPIC mode if flags.0 bit:INTR_REMAP is also > > set. > > > > This patch enable the feature. Also re-define x2apic_supported() to address > > platform x2apic support needs 1)processor has x2apic capability 2)interrupt > > remapping support 3)firmware does not request opt-out. > > Given that x2apic is *required* to be safe from irq injection tricks, > why would we ever want to manually disable it? > > Is this just a workaround for a crappy BIOS? What is the *actual* reason > for wanting to disable x2apic? Just a guess, but the OEM probably hasn't updated their SMI handlers to understand x2apic yet and won't before the product ships because some other OS doesn't bother to use x2apic. We can still enable interrupt remapping w/o x2apic though, so I'm curious what other irq injection tricks you're referring to. Thanks, Alex -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/