From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S267195AbUG1PDb (ORCPT ); Wed, 28 Jul 2004 11:03:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S267197AbUG1PDb (ORCPT ); Wed, 28 Jul 2004 11:03:31 -0400 Received: from jade.spiritone.com ([216.99.193.136]:50391 "EHLO jade.spiritone.com") by vger.kernel.org with ESMTP id S267195AbUG1PD1 (ORCPT ); Wed, 28 Jul 2004 11:03:27 -0400 Date: Wed, 28 Jul 2004 08:01:20 -0700 From: "Martin J. Bligh" To: colpatch@us.ibm.com, Jesse Barnes cc: Christoph Hellwig , Jesse Barnes , Andi Kleen , LKML , LSE Tech Subject: Re: [Lse-tech] [RFC][PATCH] Change pcibus_to_cpumask() to pcibus_to_node() Message-ID: <82510000.1091026879@[10.10.2.4]> In-Reply-To: <1090953179.18747.19.camel@arrakis> References: <1090887007.16676.18.camel@arrakis> <20040727105145.A18533@infradead.org> <200407270822.43870.jbarnes@engr.sgi.com> <1090953179.18747.19.camel@arrakis> X-Mailer: Mulberry/2.2.1 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org >> I wonder though if we shouldn't add >> >> ... >> # ifdef CONFIG_NUMA >> int node; /* or nodemask_t if necessary */ >> # endif >> ... >> >> to struct pci_bus instead? That would make the existing code paths a little >> faster and avoid the need for a global array, which tends to lead to TLB >> misses. > > I like that idea! Stick a nodemask_t in struct pci_bus, initialize it > to NODE_MASK_ALL. If a particular arch wants to put something more > accurate in there, then great, if not, we're just in the same boat we're > in now. > > Anyone else have opinions one way or the other on Jesse's idea? Sounds great - if it's possible to add it to something more generic than PCI, that'd be even better, but pci would still be very useful. M.