From patchwork Mon May 6 07:04:26 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Nakajima, Jun" X-Patchwork-Id: 2522891 Return-Path: X-Original-To: patchwork-kvm@patchwork.kernel.org Delivered-To: patchwork-process-083081@patchwork1.kernel.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by patchwork1.kernel.org (Postfix) with ESMTP id 93E373FD85 for ; Mon, 6 May 2013 07:04:56 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752955Ab3EFHEw (ORCPT ); Mon, 6 May 2013 03:04:52 -0400 Received: from mail-pb0-f49.google.com ([209.85.160.49]:35636 "EHLO mail-pb0-f49.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752934Ab3EFHEw (ORCPT ); Mon, 6 May 2013 03:04:52 -0400 Received: by mail-pb0-f49.google.com with SMTP id rp8so1851175pbb.36 for ; Mon, 06 May 2013 00:04:51 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:from:to:subject:date:message-id:x-mailer:in-reply-to :references:x-gm-message-state; bh=bcT060trJ2EyAUUrcZj8zGOqxWHpLp2Xv7IwnJwXLr4=; b=VNU8xjXPcJaRCTgNjXY00Gr5uJOfM/GIQUGxKASfStgKPQEGAPEpUcvMPS5DobOHXK iaX2m+2SQUawM+6pXElGbj0WWV+MlWhypCmkbjPB6KcmRFbFec8S+6msfk3djlW0aF59 grfCyM97Xnej1084CVs4Aao5kkh39j87a7KeobofX3B4teMa4vhcLR2cSuSx350GRGhE +huM2RM5E+UiVhGtehqAcMlnYMVqEXWlnbvelEmESNsIuuBy8m1CEnWm2khQdT/R3L2s LtZUEOghtb65NytmqV033srY1U458/4fHlG54aJJVBLRYJB1r9F5EbAPJpkFq8HInjsm dMjw== X-Received: by 10.66.89.65 with SMTP id bm1mr11974752pab.179.1367823891515; Mon, 06 May 2013 00:04:51 -0700 (PDT) Received: from localhost (c-98-207-34-191.hsd1.ca.comcast.net. [98.207.34.191]) by mx.google.com with ESMTPSA id li15sm24588989pab.2.2013.05.06.00.04.49 for (version=TLSv1.2 cipher=RC4-SHA bits=128/128); Mon, 06 May 2013 00:04:50 -0700 (PDT) From: Jun Nakajima To: kvm@vger.kernel.org Subject: [PATCH v2 07/13] nEPT: Fix wrong test in kvm_set_cr3 Date: Mon, 6 May 2013 00:04:26 -0700 Message-Id: <1367823872-25895-7-git-send-email-jun.nakajima@intel.com> X-Mailer: git-send-email 1.8.2.1.610.g562af5b In-Reply-To: <1367823872-25895-6-git-send-email-jun.nakajima@intel.com> References: <1367823872-25895-1-git-send-email-jun.nakajima@intel.com> <1367823872-25895-2-git-send-email-jun.nakajima@intel.com> <1367823872-25895-3-git-send-email-jun.nakajima@intel.com> <1367823872-25895-4-git-send-email-jun.nakajima@intel.com> <1367823872-25895-5-git-send-email-jun.nakajima@intel.com> <1367823872-25895-6-git-send-email-jun.nakajima@intel.com> X-Gm-Message-State: ALoCoQkUvjDrGtIf1Za9dxs2DVMEpVnWVHfygmchh975F+aF9OHllXr9PQ+s1AH4dwKoMS9ClGlb Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org kvm_set_cr3() attempts to check if the new cr3 is a valid guest physical address. The problem is that with nested EPT, cr3 is an *L2* physical address, not an L1 physical address as this test expects. As the comment above this test explains, it isn't necessary, and doesn't correspond to anything a real processor would do. So this patch removes it. Note that this wrong test could have also theoretically caused problems in nested NPT, not just in nested EPT. However, in practice, the problem was avoided: nested_svm_vmexit()/vmrun() do not call kvm_set_cr3 in the nested NPT case, and instead set the vmcb (and arch.cr3) directly, thus circumventing the problem. Additional potential calls to the buggy function are avoided in that we don't trap cr3 modifications when nested NPT is enabled. However, because in nested VMX we did want to use kvm_set_cr3() (as requested in Avi Kivity's review of the original nested VMX patches), we can't avoid this problem and need to fix it. Signed-off-by: Nadav Har'El Signed-off-by: Jun Nakajima Signed-off-by: Xinhao Xu --- arch/x86/kvm/x86.c | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e172132..c34590d 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -659,17 +659,6 @@ int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) */ } - /* - * Does the new cr3 value map to physical memory? (Note, we - * catch an invalid cr3 even in real-mode, because it would - * cause trouble later on when we turn on paging anyway.) - * - * A real CPU would silently accept an invalid cr3 and would - * attempt to use it - with largely undefined (and often hard - * to debug) behavior on the guest side. - */ - if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT))) - return 1; vcpu->arch.cr3 = cr3; __set_bit(VCPU_EXREG_CR3, (ulong *)&vcpu->arch.regs_avail); vcpu->arch.mmu.new_cr3(vcpu);