Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756043Ab2B0XAi (ORCPT ); Mon, 27 Feb 2012 18:00:38 -0500 Received: from mail-ww0-f44.google.com ([74.125.82.44]:34165 "EHLO mail-ww0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755174Ab2B0XAh convert rfc822-to-8bit (ORCPT ); Mon, 27 Feb 2012 18:00:37 -0500 Authentication-Results: mr.google.com; spf=pass (google.com: domain of bhelgaas@google.com designates 10.180.107.67 as permitted sender) smtp.mail=bhelgaas@google.com; dkim=pass header.i=bhelgaas@google.com MIME-Version: 1.0 In-Reply-To: References: <1330299202-3838-1-git-send-email-yinghai@kernel.org> <1330299202-3838-6-git-send-email-yinghai@kernel.org> From: Bjorn Helgaas Date: Mon, 27 Feb 2012 16:00:15 -0700 Message-ID: Subject: Re: [PATCH 5/8] PCI: add generic device into pci_host_bridge struct To: Yinghai Lu Cc: Jesse Barnes , linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT X-System-Of-Record: true Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1905 Lines: 49 On Mon, Feb 27, 2012 at 3:55 PM, Yinghai Lu wrote: > On Mon, Feb 27, 2012 at 2:24 PM, Bjorn Helgaas wrote: >>> >>> for every bus will have ->bridge, and that is device pointer. to ?pci >>> bridge dev. >>> >>> for root bus, there is not pci bridge for that. So create_root_bus >>> allocate local dummy one. >>> >>> that dummy device will be bus->dev's parent. >>> >>> Now pci_host_bridge is added, and it is allocated, so just put the >>> struct device into that pci_host_bridge. >>> >>> and use pci_host_bridge->dev as parent of root bus->dev's parent. >> >> What's the device passed in to pci_create_root_bus()? ?Why is that >> different than the root bus dev's parent? > > ia64 and x86 doesn't pass that. > > arch/ia64/pci/pci.c: ? ?pbus = pci_create_root_bus(NULL, bus, > &pci_root_ops, controller, > arch/powerpc/kernel/pci-common.c: ? ? ? bus = > pci_create_root_bus(hose->parent, hose->first_busno, > arch/sparc/kernel/pci.c: ? ? ? ?bus = pci_create_root_bus(parent, > pbm->pci_first_busno, pbm->pci_ops, > arch/x86/pci/acpi.c: ? ? ? ? ? ?bus = pci_create_root_bus(NULL, > busnum, &pci_root_ops, sd, > drivers/parisc/dino.c: ?dino_dev->hba.hba_bus = bus = > pci_create_root_bus(&dev->dev, > drivers/parisc/lba_pci.c: ? ? ? ? ? ? ? pci_create_root_bus(&dev->dev, > lba_dev->hba.bus_num.start, > > sparc, powerpc, parisc have their local ... > > so local hose etc's dev ---> allocated dev ---> root bus's dev > > this patch change to > > local hose etc's dev ---> allocated host bridge's dev ---> root bus's dev Why do we have a separate dev for the hose and the root bus's dev? The hose *is* a PCI host bridge. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/