From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S935835AbXFFV0Z (ORCPT ); Wed, 6 Jun 2007 17:26:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S935080AbXFFV0S (ORCPT ); Wed, 6 Jun 2007 17:26:18 -0400 Received: from mtagate5.uk.ibm.com ([195.212.29.138]:45909 "EHLO mtagate5.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934493AbXFFV0R (ORCPT ); Wed, 6 Jun 2007 17:26:17 -0400 Date: Thu, 7 Jun 2007 00:26:09 +0300 From: Muli Ben-Yehuda To: Jeff Garzik Cc: Andi Kleen , Andrew Morton , LKML Subject: Re: [PATCH] stop x86 ->sysdata abuse; introduce pci_sysdata Message-ID: <20070606212609.GS8541@rhun.ibm.com> References: <11798211663773-git-send-email-muli@il.ibm.com> <1179821166685-git-send-email-muli@il.ibm.com> <11798211661041-git-send-email-muli@il.ibm.com> <20070604113344.0e71ca8c.akpm@linux-foundation.org> <20070604190654.GA20299@havoc.gtf.org> <20070604191346.GL4556@rhun.ibm.com> <20070604210551.GH20299@havoc.gtf.org> <20070605102905.GA3305@rhun.haifa.ibm.com> <20070606210329.GQ8541@rhun.ibm.com> <20070606212151.GA3304@havoc.gtf.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070606212151.GA3304@havoc.gtf.org> User-Agent: Mutt/1.5.11 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Jun 06, 2007 at 05:21:51PM -0400, Jeff Garzik wrote: > > Ok, patch fixed, works for me with Calgary. Andi, it looks like you > > added the acpi.c NUMA bits originally, perhaps you could test and/or > > ack them? > > Just so there is no misunderstanding, I added the allocation bits. > > No idea who added the direct usage of ->sysdata for NUMA info. Andi Kleen did according to the git log. I'm trying to get access to a NUMA system, but in the meantime perhaps Andi can just eyeball it and ack. > > diff -r 05804111dbee arch/i386/pci/common.c > > --- a/arch/i386/pci/common.c Tue Jun 05 11:34:51 2007 +0300 > > +++ b/arch/i386/pci/common.c Wed Jun 06 22:36:53 2007 +0300 > > @@ -29,12 +29,14 @@ struct pci_raw_ops *raw_pci_ops; > > > > static int pci_read(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 *value) > > { > > - return raw_pci_ops->read(0, bus->number, devfn, where, size, value); > > + return raw_pci_ops->read(pci_domain_nr(bus), bus->number, > > + devfn, where, size, value); > > } > > > > static int pci_write(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 value) > > { > > - return raw_pci_ops->write(0, bus->number, devfn, where, size, value); > > + return raw_pci_ops->write(pci_domain_nr(bus), bus->number, > > + devfn, where, size, value); > > } > > > > struct pci_ops pci_root_ops = { > > Patch leakage? I thought I had removed this portion, as being > specific to PCI domains. Nope, it was there in your original patch and I kept it since it's obviously harmless (without CONFIG_PCI_DOMAINS pci_domain_nr() is defined to 0). > No objection, obviously, and it's obviously safe given that > CONFIG_PCI_DOMAINS is not defined on x86/x86-64. > > Nonetheless, it appears "off-topic" for NUMA/Calgary purposes. > > ACK everything, but IMO the pci_domain_nr() bits should be in a > separate patch. Fair enough, I'll drop it out of this one. Cheers, Muli