From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750905AbbCSPdT (ORCPT ); Thu, 19 Mar 2015 11:33:19 -0400 Received: from utopia.booyaka.com ([74.50.51.50]:34848 "EHLO utopia.booyaka.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751031AbbCSPdP (ORCPT ); Thu, 19 Mar 2015 11:33:15 -0400 Date: Thu, 19 Mar 2015 15:33:13 +0000 (UTC) From: Paul Walmsley To: Stephen Warren cc: linux-tegra@vger.kernel.org, linux-arm-kernel@vger.kernel.org, Mark Rutland , Alexandre Courbot , Pawel Moll , Ian Campbell , linux-kernel@vger.kernel.org, Eduardo Valentin , devicetree@vger.kernel.org, Rob Herring , Thierry Reding , Kumar Gala , Hiroshi DOYU Subject: Re: [PATCH 3/3] Documentation: DT bindings: Tegra AHB: note base address change In-Reply-To: <550859A0.9090500@wwwdotorg.org> Message-ID: References: <20150317083221.32662.14647.stgit@baseline> <20150317083221.32662.96822.stgit@baseline> <550859A0.9090500@wwwdotorg.org> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 17 Mar 2015, Stephen Warren wrote: > On 03/17/2015 02:32 AM, Paul Walmsley wrote: > > For Tegra132 and later chips, we can now use the correct hardware base > > address for the Tegra AHB IP block in the DT data. Update the DT binding > > documentation to reflect this change. > > > diff --git > > a/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-ahb.txt > > b/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-ahb.txt > > index 067c979..7692b4c 100644 > > --- a/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-ahb.txt > > +++ b/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-ahb.txt > > @@ -2,10 +2,15 @@ NVIDIA Tegra AHB > > > > Required properties: > > - compatible : For Tegra20, must contain "nvidia,tegra20-ahb". For > > - Tegra30, must contain "nvidia,tegra30-ahb". Otherwise, must contain > > - '"nvidia,-ahb", "nvidia,tegra30-ahb"' where is tegra124, > > - tegra132, or tegra210. > > -- reg : Should contain 1 register ranges(address and length) > > + Tegra30, must contain "nvidia,tegra30-ahb". For Tegra114 and Tegra124, > > must > > + contain '"nvidia,-ahb", "nvidia,tegra30-ahb"' where is > > tegra114 > > + or tegra124. For Tegra132, the compatible string must contain > > + "nvidia,tegra132-ahb". > > + > > +- reg : Should contain 1 register ranges(address and length). On Tegra20, > > + Tegra30, Tegra114, and Tegra124 chips, the low byte of the physical base > > + address of the IP block must end in 0x04. On DT files for later chips, > > the > > + actual hardware base address of the IP block should be used. > > A table-based approach rather than prose might make this more legible? > > - compatible: Must contain the following: > Tegra20: "nvidia,tegra20-ahb" > Tegra30: "nvidia,tegra30-ahb" > Tegra114: "nvidia,tegra114-ahb", "nvidia,tegra30-ahb" > Tegra124: "nvidia,tegra124-ahb", "nvidia,tegra30-ahb" > Tegra132: "nvidia,tegra132-ahb" > Tegra210: "nvidia,tegra210-ahb", "nvidia,tegra132-ahb" > > With any luck, we can extend that final item for future chips to be: > > Tegra210, TegraNNN: > "nvidia,tegra-ahb", "nvidia,tegra132-ahb" > > Perhaps we format the 114/124 entry that way too. I think I'm just going to drop this patch, since Russell prefers that the workaround is applied in the driver. With regards to using tables rather than narrative descriptions: perhaps consider a patch to Documentation/devicetree/bindings/submitting-patches.txt ? I don't know what the DT binding documentation maintainers' future plans are with regards to automated documentation reflow, etc., but submitting a patch there would stimulate at least some coordination on the issue. > As a historical note, I thought this issue might have been caused by the HW > lumping together a bunch of vaguely related features into the same address > window, with offset 0 being unrelated to AHB. This is certainly the > explanation for some other interesting reg value in DT on other HW modules in > Tegra. However as best I can tell that isn't the case for this HW module; we > just mapped the start of the reg value to the first defined register for some > reason rather than aligning it with the entry in the address map. So, this > series makes sense to me. > > The series, > Acked-by: Stephen Warren > > (it'd be nice if the DT doc was modified as I described above, but not > strictly necessary. I like Russell's ideas how to make this more > transparent/automatic so old DTs can be converted too). I'll add the ack on the first patch, but since the second one has been changed, I won't add one there yet. - Paul