From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from terminus.zytor.com ([198.137.202.10]:54723 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933147Ab3HGWGg (ORCPT ); Wed, 7 Aug 2013 18:06:36 -0400 Message-ID: <5202C4C3.6050007@zytor.com> Date: Wed, 07 Aug 2013 15:05:55 -0700 From: "H. Peter Anvin" MIME-Version: 1.0 Subject: Re: [PATCH 00/26] STA2X11 devicetree support for amba/pci References: <52026AB5.80009@zytor.com> <20130807211235.GA17863@mail.gnudd.com> In-Reply-To: <20130807211235.GA17863@mail.gnudd.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Sender: devicetree-owner@vger.kernel.org To: Alessandro Rubini Cc: linux-kernel@vger.kernel.org, ciminaghi@gnudd.com, giancarlo.asnaghi@st.com, x86@kernel.org, mingo@redhat.com, linux@arm.linux.org.uk, tglx@linutronix.de, devicetree@vger.kernel.org List-ID: On 08/07/2013 02:12 PM, Alessandro Rubini wrote: > > I understand. My impression of devicetree is exactly like that, I must > say. What we have here is a very clean PCI enumeration of it all: we > only need to specify the mapping of GPIO pins (i.e. card-detect for > mmc) and DMA channels, as all the rest works by itself. > Could you elucidate this a bit more? Specifically, from the PCI host's perspective, how do you know what you are dealing with (presumably PCI VID:DID:SVID:SID?) How exactly does the mapping of PCI BARs and INT/MSI/MSI-X channels to the slave device play in? I think it is safe to say that the device models we have today simply don't account for this kind of usage, which, as you say, is a genuine problem, and I think we need to understand the problem space better in order to properly architect the solution, as I don't think we have one right now. -hpa