From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1764523AbXKTXgl (ORCPT ); Tue, 20 Nov 2007 18:36:41 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758246AbXKTXgR (ORCPT ); Tue, 20 Nov 2007 18:36:17 -0500 Received: from ns2.suse.de ([195.135.220.15]:39472 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754966AbXKTXgQ (ORCPT ); Tue, 20 Nov 2007 18:36:16 -0500 Date: Tue, 20 Nov 2007 15:14:53 -0800 From: Greg KH To: Jan Beulich Cc: gary.hade@us.ibm.com, lkml Subject: Re: Avoid creating P2P prefetch window for expansion ROMs Message-ID: <20071120231453.GA20462@suse.de> References: <4742C9D7.76E4.0078.0@novell.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4742C9D7.76E4.0078.0@novell.com> User-Agent: Mutt/1.5.16 (2007-06-09) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Nov 20, 2007 at 03:49:43AM -0700, Jan Beulich wrote: > This patch (in its incarnation in our SLE10SP2 tree) is causing resource > allocation failures on one of my machines. Does the latest 2.6.23 kernel also cause these same problems? > The condition for this is that besides ROMs behind a bridge not having > their base addresses assigned there's no extra space available in the > non-prefetch window to accommodate the ROMs' space. I therefore think > the change, while having a good reason, isn't complete so far, as it > would imply the need for checking whether space in the non-prefetch > window is sufficient and continuing to force the creation of a > prefetch window otherwise. Hm, I'll defer to Gary on this one :) thanks, greg k-h