From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marc Zyngier Subject: Re: [PATCH 5/5] arm64: KVM: vgic-v2: Enable GICV access from HYP if access from guest is unsafe Date: Fri, 19 Aug 2016 15:54:21 +0100 Message-ID: <20160819155421.44355df9@arm.com> References: <1471610295-1456-1-git-send-email-marc.zyngier@arm.com> <1471610295-1456-6-git-send-email-marc.zyngier@arm.com> <57B7042D.1020702@arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Christoffer Dall , kvm-devel , arm-mail-list , "kvmarm@lists.cs.columbia.edu" To: Peter Maydell Return-path: Received: from foss.arm.com ([217.140.101.70]:36776 "EHLO foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754828AbcHSOyZ (ORCPT ); Fri, 19 Aug 2016 10:54:25 -0400 In-Reply-To: Sender: kvm-owner@vger.kernel.org List-ID: On Fri, 19 Aug 2016 14:31:12 +0100 Peter Maydell wrote: > On 19 August 2016 at 14:05, Marc Zyngier wrote: > > On 19/08/16 13:53, Peter Maydell wrote: > >> Is it worth specifically saying "performance will be worse", or do we > >> expect this to only happen on systems where the h/w can't permit direct > >> access (as opposed to those with bad dt info) ? > > > > We cannot distinguish between the two, unfortunately. Even worse, ACPI > > only gives us a base address, and not the size of the region. So even if > > the HW was perfectly compliant with SBSA, we have to assume the worse case. > > Right, but if we expect this is mostly going to be "you just have > to live with it on this hardware" there's less point in printing > an alarming message, whereas if there's a significant subset of > "dt is just wrong" cases then the alarm might help in getting them > fixed, maybe... That'd require some more infrastructure from the kernel's GIC driver (which now provides the various base addresses), but I guess that we can have a look as well. Thanks, M. -- Jazz is not dead. It just smells funny.