From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758432AbYGNV22 (ORCPT ); Mon, 14 Jul 2008 17:28:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758257AbYGNV2Q (ORCPT ); Mon, 14 Jul 2008 17:28:16 -0400 Received: from outbound-mail-146.bluehost.com ([67.222.38.36]:38439 "HELO outbound-mail-146.bluehost.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1758243AbYGNV2P (ORCPT ); Mon, 14 Jul 2008 17:28:15 -0400 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=default; d=virtuousgeek.org; h=Received:From:To:Subject:Date:User-Agent:Cc:References:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Content-Disposition:Message-Id:X-Identified-User:DomainKey-Status; b=K1rmk7QN8jNOHzK9fHTy96lI4CEzaTBauQXWn5+fgBSZPpsLps8GHSpzFFGxvcwgOXqmLGFRwWZFfA0bXfgenXSek6vnPwHeBowRO4wKMWooYpN04GA6GP1P6HQTvcmK; From: Jesse Barnes To: Matthew Wilcox Subject: Re: x86: PCI multi domain/segment config access Date: Mon, 14 Jul 2008 14:28:10 -0700 User-Agent: KMail/1.9.9 Cc: John Keller , linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org References: <20080711211347.6461.68923.91734@attica.americas.sgi.com> <20080711212117.GW14894@parisc-linux.org> In-Reply-To: <20080711212117.GW14894@parisc-linux.org> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200807141428.11022.jbarnes@virtuousgeek.org> X-Identified-User: {642:box128.bluehost.com:virtuous:virtuousgeek.org} {sentby:smtp auth 75.111.27.49 authed with jbarnes@virtuousgeek.org} DomainKey-Status: no signature Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Friday, July 11, 2008 2:21 pm Matthew Wilcox wrote: > On Fri, Jul 11, 2008 at 04:13:47PM -0500, John Keller wrote: > > On x86 PCI Config space access to domains/segments > 0 appears to be > > broken. The noted patch related to PCI ext config space has all config > > accesses to offsets < 256 using conf1 if available, which is most likely > > the case. > > > > This would force usage of conf1 for non-extended config access (< 256) on > > PCI domains > 0. > > > > Am I missing something, or how would a conf1 access reference a domain > > other than zero? > > You're correct. Sorry about that, I wasn't thinking about domains at > the time I wrote the patch. Try this: > > [X86] Fix PCI config space for domains > 0 > > John Keller reports that PCI config space access is broken on machines > with more than one domain. conf1 accesses only work for domain 0. > > Signed-off-by: Matthew Wilcox Oops, yeah, good catch. Thanks for fixing this up, Matthew. Just applied it to my linux-next branch. Thanks, Jesse