From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759124AbZGGQxW (ORCPT ); Tue, 7 Jul 2009 12:53:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755973AbZGGQxN (ORCPT ); Tue, 7 Jul 2009 12:53:13 -0400 Received: from sg2ehsobe003.messaging.microsoft.com ([207.46.51.77]:12554 "EHLO SG2EHSOBE003.bigfish.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1755225AbZGGQxM convert rfc822-to-8bit (ORCPT ); Tue, 7 Jul 2009 12:53:12 -0400 X-SpamScore: -34 X-BigFish: VPS-34(zz146fK1432R98dN936eM1805M9371Pzz1202hzzz32i6bh43j61h) X-Spam-TCS-SCL: 0:0 X-WSS-ID: 0KMF87X-03-N71-01 Date: Tue, 7 Jul 2009 18:52:48 +0200 From: Joerg Roedel To: Avi Kivity CC: Marcelo Tosatti , kvm@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 0/6 v4] KVM support for 1GB pages Message-ID: <20090707165248.GR5380@amd.com> References: <1246984867-15952-1-git-send-email-joerg.roedel@amd.com> <4A537C12.2040501@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline In-Reply-To: <4A537C12.2040501@redhat.com> User-Agent: Mutt/1.5.20 (2009-06-14) Content-Transfer-Encoding: 8BIT X-OriginalArrivalTime: 07 Jul 2009 16:52:48.0450 (UTC) FILETIME=[5BB81A20:01C9FF23] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Jul 07, 2009 at 07:47:14PM +0300, Avi Kivity wrote: > On 07/07/2009 07:41 PM, Joerg Roedel wrote: > >Hi, > > > >this is the forth version of the patches for KVM to support 1GB pages. > >This code still only can make use of 1GB pages with nested paging > >enabled. I will give the shadow paging code another debug round soon. > >Please comment or consider to apply these patches. > > > > I would really like to have shadow 1GB work, or at least gain some > understanding of why it doesn't. > > Can you detail the changes compared to the previous round? Until now I addressed the review comments from the previous round. The biggest change is patch #4 which was completly rewritten. I understand that it would be nicer to have shadow paging working too. But since most users will use this feature only on processors with nested paging (only processors with nested paging support gbpages) I don't think it should be a show-stopper. YMMV. Anyway, I will further try to get shadow paging support running and stable. Joerg -- | Advanced Micro Devices GmbH Operating | Karl-Hammerschmidt-Str. 34, 85609 Dornach bei München System | Research | Geschäftsführer: Thomas M. McCoy, Giuliano Meroni Center | Sitz: Dornach, Gemeinde Aschheim, Landkreis München | Registergericht München, HRB Nr. 43632