From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH v2 14/29] sparc/PCI: convert to pci_create_root_bus() for correct root bus resources Date: Fri, 14 Oct 2011 00:46:54 -0400 (EDT) Message-ID: <20111014.004654.210648346142583443.davem@davemloft.net> References: <20111014042142.23504.70417.stgit@bhelgaas.mtv.corp.google.com> <20111014042818.23504.61318.stgit@bhelgaas.mtv.corp.google.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20111014042818.23504.61318.stgit@bhelgaas.mtv.corp.google.com> Sender: linux-pci-owner@vger.kernel.org To: bhelgaas@google.com Cc: linux-pci@vger.kernel.org, linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: linux-arch.vger.kernel.org From: Bjorn Helgaas Date: Thu, 13 Oct 2011 22:28:18 -0600 > Convert from pci_create_bus() to pci_create_root_bus(). This way the root > bus resources are correct immediately. This patch doesn't fix anything > because sparc fixed them before scanning the bus, but it makes it more > consistent with other architectures. > > I didn't work out where pcibios_fixup_bus() is called. I don't see a > use of pci_scan_bus(), pci_scan_bus_parented(), or pci_scan_child_bus(). > > CC: David S. Miller > Signed-off-by: Bjorn Helgaas The pcibios_fixup_bus() case here (sparc64) is an artifact from when we actually probed the PCI config space for device probing on sparc64, and pci_bus_probe() would invoke it. Unfortunately we can't delete this function now even though it's empty with your changes, because drivers/pci/probe.o is unconditionally linked into the kernel when CONFIG_PCI even when an architecture (such as sparc64) doesn't use the majority of the code in it. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from shards.monkeyblade.net ([198.137.202.13]:60088 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752000Ab1JNErL (ORCPT ); Fri, 14 Oct 2011 00:47:11 -0400 Date: Fri, 14 Oct 2011 00:46:54 -0400 (EDT) Message-ID: <20111014.004654.210648346142583443.davem@davemloft.net> Subject: Re: [PATCH v2 14/29] sparc/PCI: convert to pci_create_root_bus() for correct root bus resources From: David Miller In-Reply-To: <20111014042818.23504.61318.stgit@bhelgaas.mtv.corp.google.com> References: <20111014042142.23504.70417.stgit@bhelgaas.mtv.corp.google.com> <20111014042818.23504.61318.stgit@bhelgaas.mtv.corp.google.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org List-ID: To: bhelgaas@google.com Cc: linux-pci@vger.kernel.org, linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org Message-ID: <20111014044654.y2cO_amhIGANav6EG7ySeUov-Aetep00xr5d41pj8cU@z> From: Bjorn Helgaas Date: Thu, 13 Oct 2011 22:28:18 -0600 > Convert from pci_create_bus() to pci_create_root_bus(). This way the root > bus resources are correct immediately. This patch doesn't fix anything > because sparc fixed them before scanning the bus, but it makes it more > consistent with other architectures. > > I didn't work out where pcibios_fixup_bus() is called. I don't see a > use of pci_scan_bus(), pci_scan_bus_parented(), or pci_scan_child_bus(). > > CC: David S. Miller > Signed-off-by: Bjorn Helgaas The pcibios_fixup_bus() case here (sparc64) is an artifact from when we actually probed the PCI config space for device probing on sparc64, and pci_bus_probe() would invoke it. Unfortunately we can't delete this function now even though it's empty with your changes, because drivers/pci/probe.o is unconditionally linked into the kernel when CONFIG_PCI even when an architecture (such as sparc64) doesn't use the majority of the code in it.