From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from quartz.orcorp.ca ([184.70.90.242]:55242 "EHLO quartz.orcorp.ca" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1761031AbaCUQk1 (ORCPT ); Fri, 21 Mar 2014 12:40:27 -0400 Date: Fri, 21 Mar 2014 10:40:15 -0600 From: Jason Gunthorpe To: Arnd Bergmann Cc: linux-arm-kernel@lists.infradead.org, Phil.Edworthy@renesas.com, Sergei Shtylyov , linux-sh@vger.kernel.org, linux-pci@vger.kernel.org, Magnus Damm , Valentine Barshak , Simon Horman , Bjorn Helgaas , Ben Dooks Subject: Re: [PATCH v4 6/9] ARM: shmobile: Add PCIe device tree nodes for R8A7790 Message-ID: <20140321164015.GA3374@obsidianresearch.com> References: <1395397968-6242-1-git-send-email-phil.edworthy@renesas.com> <532C5ADE.4030708@cogentembedded.com> <6044873.2S2Xlhp0hC@wuerfel> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <6044873.2S2Xlhp0hC@wuerfel> Sender: linux-pci-owner@vger.kernel.org List-ID: On Fri, Mar 21, 2014 at 03:57:25PM +0100, Arnd Bergmann wrote: > On Friday 21 March 2014 14:53:48 Phil.Edworthy@renesas.com wrote: > > > > > > + reg = <0 0xfe000000 0 0x80000>; > > > > + #address-cells = >; > > > > + #size-cells = <2>; > > > > + device_type = "pci"; > > > > > > This legacy property is only meaningful for the true OF firmware and > > thus > > > have been phased out (it's only actively used for the memory nodes). > > Ok, thanks for pointing this out. > > > > I think the code still relies on it for PCI though. Right, it must be present for PCI nodes, it triggers special behavior in the address parsing routines. What Phil has looks correct to me. Regards, Jason