From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760183AbZD1Kg2 (ORCPT ); Tue, 28 Apr 2009 06:36:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1759845AbZD1KgL (ORCPT ); Tue, 28 Apr 2009 06:36:11 -0400 Received: from outbound-dub.frontbridge.com ([213.199.154.16]:23403 "EHLO IE1EHSOBE005.bigfish.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759685AbZD1KgJ convert rfc822-to-8bit (ORCPT ); Tue, 28 Apr 2009 06:36:09 -0400 X-Greylist: delayed 903 seconds by postgrey-1.27 at vger.kernel.org; Tue, 28 Apr 2009 06:36:09 EDT X-BigFish: VPS-25(zz1432R98dR1805M936fKzz1202hzzz32i6bh43j61h) X-Spam-TCS-SCL: 0:0 X-WSS-ID: 0KIT3ET-02-P8U-01 Date: Tue, 28 Apr 2009 12:20:54 +0200 From: Joerg Roedel To: Avi Kivity CC: Marcelo Tosatti , , Subject: Re: [PATCH 0/7] KVM support for 1GB pages v2 Message-ID: <20090428102054.GX17438@amd.com> References: <1240574327-21575-1-git-send-email-joerg.roedel@amd.com> <49F6D767.9050801@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline In-Reply-To: <49F6D767.9050801@redhat.com> User-Agent: Mutt/1.5.18 (2008-05-17) Content-Transfer-Encoding: 8BIT X-OriginalArrivalTime: 28 Apr 2009 10:20:55.0134 (UTC) FILETIME=[03C64FE0:01C9C7EB] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Apr 28, 2009 at 01:16:07PM +0300, Avi Kivity wrote: > Joerg Roedel wrote: >> Hi, >> >> this is the second version of the patches implementing support for 1GB >> pages in KVM. There are too many changes to the first version to mention >> them all here. The core support was rewritten to work with mapping >> levels instead of page sizes. This was the main part of the change. >> The patches still only support KVM with nested paging enabled > > What is missing to get shadow paging working with this? I don't think > it's particularly important (it will be pretty hard to find a GB page > without any shadowed pages in it), but I don't like arbitrary > constraints. 1GB pages with shadow paging does not work yet. But I will post the patch for it when it works stable for me. It makes sense at least if the guest uses 1gb pages in hugetlbfs. > walk_addr() needs to change to support GB pages, did I miss it in the > patchset? Ah true. I missed that in this patchset. I will add it. Joerg -- | Advanced Micro Devices GmbH Operating | Karl-Hammerschmidt-Str. 34, 85609 Dornach bei München System | Research | Geschäftsführer: Jochen Polster, Thomas M. McCoy, Giuliano Meroni Center | Sitz: Dornach, Gemeinde Aschheim, Landkreis München | Registergericht München, HRB Nr. 43632