From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [patch 0/4] mmu audit update Date: Sun, 07 Jun 2009 10:14:03 +0300 Message-ID: <4A2B68BB.4070507@redhat.com> References: <20090602213655.640083007@localhost.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: sheng@linux.intel.com, kvm@vger.kernel.org To: Marcelo Tosatti Return-path: Received: from mx2.redhat.com ([66.187.237.31]:37428 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754141AbZFGHOE (ORCPT ); Sun, 7 Jun 2009 03:14:04 -0400 In-Reply-To: <20090602213655.640083007@localhost.localdomain> Sender: kvm-owner@vger.kernel.org List-ID: Marcelo Tosatti wrote: > Some updates to the MMU audit code. > > The third patch is "guessy" because I could not find the notrap spte > documentation, all I can see is the page-fault error code mask and match > fields in the VMCS, but can't see the link of that to sptes. Can someone > point it out please? > When bypass_guest_pf is set, we tell vmx not to trap if the fault is due to page-not-present. So if we know gpte.p == 0, we set spte.p = 0 and allow not-present page faults to go directly to the guest without trapping. Of course, we still need to trap cases where gpte.p = 1 but we haven't mapped the page yet. So we set a reserved bit in the spte and trap on that. -- Do not meddle in the internals of kernels, for they are subtle and quick to panic.