From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joerg Roedel Subject: Re: [PATCH 16/22] KVM: MMU: Track page fault data in struct vcpu Date: Tue, 27 Apr 2010 15:57:00 +0200 Message-ID: <20100427135700.GK11097@amd.com> References: <1272364712-17425-1-git-send-email-joerg.roedel@amd.com> <1272364712-17425-17-git-send-email-joerg.roedel@amd.com> <4BD6DF7C.1090203@redhat.com> <20100427132801.GI11097@amd.com> <4BD6E8A6.5040008@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Cc: Marcelo Tosatti , kvm@vger.kernel.org, linux-kernel@vger.kernel.org To: Avi Kivity Return-path: Received: from va3ehsobe002.messaging.microsoft.com ([216.32.180.12]:16342 "EHLO VA3EHSOBE002.bigfish.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753416Ab0D0N5Q (ORCPT ); Tue, 27 Apr 2010 09:57:16 -0400 Content-Disposition: inline In-Reply-To: <4BD6E8A6.5040008@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On Tue, Apr 27, 2010 at 04:37:42PM +0300, Avi Kivity wrote: > On 04/27/2010 04:28 PM, Joerg Roedel wrote: > > > >>>diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h > >>>index d9dfc8c..8426870 100644 > >>>--- a/arch/x86/include/asm/kvm_host.h > >>>+++ b/arch/x86/include/asm/kvm_host.h > >>>@@ -298,6 +298,9 @@ struct kvm_vcpu_arch { > >>> /* Used for two dimensional paging emulation */ > >>> struct kvm_mmu nested_mmu; > >>> > >>>+ unsigned long fault_address; > >>Probably a problem on i386. How does npt handle faults when the > >>guest is using pae paging and the host (in our case the guest...) > >>isn't? I see it uses exit_info_2 for the address, which is a u64. > >This shouldn't be an issue. If we run on 32bit host with nested paging > >the guest can't have more than 4gb of addressable memory because of the > >page table limitations (nested page table is always in host format). > > But the nested guest can use pae paging and generate a #NPF with > exit_info_2 > 4GB. So we need to keep the full fault address; if we > truncate, the guest might actually resolve the fault and let the > nested guest continue. This could only be a malicious guest because it can't have memory above 4gb. But a guest could certainly setup its page tables to point there, thats true. So I change it to u64. Joerg