From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755901AbZBZORp (ORCPT ); Thu, 26 Feb 2009 09:17:45 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756054AbZBZOPt (ORCPT ); Thu, 26 Feb 2009 09:15:49 -0500 Received: from mx2.redhat.com ([66.187.237.31]:43053 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754150AbZBZOPk (ORCPT ); Thu, 26 Feb 2009 09:15:40 -0500 From: Avi Kivity To: kvm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Subject: [PATCH 21/40] KVM: MMU: Segregate mmu pages created with different cr4.pge settings Date: Thu, 26 Feb 2009 16:15:11 +0200 Message-Id: <1235657730-27683-22-git-send-email-avi@redhat.com> In-Reply-To: <1235657730-27683-1-git-send-email-avi@redhat.com> References: <1235657730-27683-1-git-send-email-avi@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Don't allow a vcpu with cr4.pge cleared to use a shadow page created with cr4.pge set; this might cause a cr3 switch not to sync ptes that have the global bit set (the global bit has no effect if !cr4.pge). This can only occur on smp with different cr4.pge settings for different vcpus (since a cr4 change will resync the shadow ptes), but there's no cost to being correct here. Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/x86.c | 1 + 2 files changed, 2 insertions(+), 0 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 28f875f..c2a01d0 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -183,6 +183,7 @@ union kvm_mmu_page_role { unsigned metaphysical:1; unsigned access:3; unsigned invalid:1; + unsigned cr4_pge:1; }; }; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 2477e87..873602b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -364,6 +364,7 @@ void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) } kvm_x86_ops->set_cr4(vcpu, cr4); vcpu->arch.cr4 = cr4; + vcpu->arch.mmu.base_role.cr4_pge = !!(cr4 & X86_CR4_PGE); kvm_mmu_sync_global(vcpu); kvm_mmu_reset_context(vcpu); } -- 1.6.0.6