From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH] KVM: x86: Add three MSRs to the list of ignored MSRs Date: Thu, 14 Apr 2016 16:42:55 +0200 Message-ID: <570FAC6F.10704@redhat.com> References: <1460626109-24343-1-git-send-email-shuai.ruan@linux.intel.com> <20160414133332.GA3350@potion.brq.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: kvm@vger.kernel.org, allen.m.kay@intel.com To: =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= , Shuai Ruan Return-path: Received: from mx1.redhat.com ([209.132.183.28]:53764 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932235AbcDNOnD (ORCPT ); Thu, 14 Apr 2016 10:43:03 -0400 In-Reply-To: <20160414133332.GA3350@potion.brq.redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 14/04/2016 15:33, Radim Kr=C4=8Dm=C3=A1=C5=99 wrote: > The main bug in KVM is that it allows the guest to think that it runs= on > CPU that isn't emulated: any CPU that has MSR_PLATFORM_INFO shouldn'= t > be exposed in KVM. That's all of them. f/m/s values and the model name are a useful debugging tool. Considering that MSR_PLATFORM_INFO is hardly ever used= , it doesn't seem to be a great compromise. The actual question is why the hell a graphics driver cares about it. Paolo > Incorrect use of CPU also is a userspace problem, so > KVM could just pass this MSR for emulation to userspace. > (Passing to userspace won't directly fix your problem, though)