From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <stable-owner@vger.kernel.org>
Received: from legacy.ddn.com ([64.47.133.206]:35553 "EHLO legacy.ddn.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1752142AbcKVVAI (ORCPT <rfc822;stable@vger.kernel.org>);
        Tue, 22 Nov 2016 16:00:08 -0500
Date: Tue, 22 Nov 2016 13:55:31 -0700
From: Greg Edwards <gedwards@ddn.com>
To: Paolo Bonzini <pbonzini@redhat.com>
CC: <stable@vger.kernel.org>, <kai.huang@linux.intel.com>,
        <pfeiner@google.com>
Subject: Re: [PATCH] kvm: vmx: ensure VMCS is current while enabling PML
Message-ID: <20161122205531.GB6919@psuche>
References: <20161122154352.25784-2-pbonzini@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <20161122154352.25784-2-pbonzini@redhat.com>
Sender: stable-owner@vger.kernel.org
List-ID: <stable.vger.kernel.org>

On Tue, Nov 22, 2016 at 04:43:52PM +0100, Paolo Bonzini wrote:
> From: Peter Feiner <pfeiner@google.com>
>
> [ upstream commit 4e59516a12a6ef6dcb660cb3a3f70c64bd60cfec, for 4.1.y only ]
>
> Between loading the new VMCS and enabling PML, the CPU was unpinned.
> If the vCPU thread were migrated to another CPU in the interim (e.g.,
> due to preemption or sleeping alloc_page), then the VMWRITEs to enable
> PML would target the wrong VMCS -- or no VMCS at all:
>
>   [ 2087.266950] vmwrite error: reg 200e value 3fe1d52000 (err -506126336)
>   [ 2087.267062] vmwrite error: reg 812 value 1ff (err 511)
>   [ 2087.267125] vmwrite error: reg 401e value 12229c00 (err 304258048)
>
> This patch ensures that the VMCS remains current while enabling PML by
> doing the VMWRITEs while the CPU is pinned. Allocation of the PML buffer
> is hoisted out of the critical section.
>
> Signed-off-by: Peter Feiner <pfeiner@google.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Tested-by: Greg Edwards <gedwards@ddn.com>