From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: [PATCH 1/2] pci: Add IORESOURCE_BIT entry for PCIe ECAM resources. Date: Mon, 02 Jun 2014 21:15:46 +0200 Message-ID: <4495090.ktpdXEyibD@wuerfel> References: <20140530233034.GH1677@bart.dudau.co.uk> <20140602162306.4AB0FC40476@trevor.secretlab.ca> <2F6515B1-48FE-4ED6-908E-CC1CAD7AF403@codeaurora.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <2F6515B1-48FE-4ED6-908E-CC1CAD7AF403@codeaurora.org> Sender: linux-kernel-owner@vger.kernel.org To: Kumar Gala Cc: Grant Likely , Liviu Dudau , Bjorn Helgaas , Rob Herring , Rob Herring , Pawel Moll , Mark Rutland , Ian Campbell , "devicetree@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "linux-pci@vger.kernel.org" , linux-arm-msm , "linux-arm-kernel@lists.infradead.org" , Kishon Vijay Abraham I List-Id: devicetree@vger.kernel.org On Monday 02 June 2014 13:09:08 Kumar Gala wrote: > > However, what do we do with the 2 cases that exist in upstream that > >> are using ranges for cfg space? > >=20 > > Ignore them in the core code? Make the specific host controller han= dle > > them I would think. >=20 > I just meant, should we =E2=80=98break=E2=80=99 their DTs and move th= em from using ranges to reg? dw-pcie is used on a lot of systems, I think we should make the common part of that driver always handle config space in a common way, and move out the part that parses the ranges property into the individual soc-specific glue drivers that want to keep optional backwards compatib= ility with existing dtbs. Which one is the other driver? Arnd