From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753223AbbJSQdD (ORCPT ); Mon, 19 Oct 2015 12:33:03 -0400 Received: from avon.wwwdotorg.org ([70.85.31.133]:33799 "EHLO avon.wwwdotorg.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751273AbbJSQdA (ORCPT ); Mon, 19 Oct 2015 12:33:00 -0400 Subject: Re: [PATCH V3 1/2] Documentation: DT: Add binding documentation for NVIDIA ADMA To: Jon Hunter References: <1444980919-20331-1-git-send-email-jonathanh@nvidia.com> <1444980919-20331-2-git-send-email-jonathanh@nvidia.com> <56212139.4080101@wwwdotorg.org> <5624D28B.70803@nvidia.com> Cc: Laxman Dewangan , Vinod Koul , Thierry Reding , Alexandre Courbot , Rob Herring , Pawel Moll , Mark Rutland , Ian Campbell , Kumar Gala , dmaengine@vger.kernel.org, linux-tegra@vger.kernel.org, linux-kernel@vger.kernel.org From: Stephen Warren Message-ID: <56251B56.70007@wwwdotorg.org> Date: Mon, 19 Oct 2015 10:33:26 -0600 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.3.0 MIME-Version: 1.0 In-Reply-To: <5624D28B.70803@nvidia.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 10/19/2015 05:22 AM, Jon Hunter wrote: > > On 16/10/15 17:09, Stephen Warren wrote: >> On 10/16/2015 01:35 AM, Jon Hunter wrote: >>> Add device-tree binding documentation for the Tegra210 Audio DMA >>> controller. >> >>> diff --git a/Documentation/devicetree/bindings/dma/tegra210-adma.txt >>> b/Documentation/devicetree/bindings/dma/tegra210-adma.txt >> >>> +Required properties: >> >>> +- interrupt-parent: Phandle to the interrupt parent controller. >> >> Nit: Since that is more of a "system level"/standard property, it's >> typical not to document it. The property is not actually required if the >> inherited value is already correct. Still, it's obvious enough what this >> means, so I'd only suggest fixing this if you have to respin for some >> other reason. > > Ok. > >>> +- clocks: Must contain one entry for the ADMA module clock, "adma_ape". >>> +- clock-names: Must contain the entry "adma_ape". >> >> Which clock is this in the CAR? I don't see any adma_ape clock >> documented in the TRM. > > Darn. I thought I had checked this. Yes it should be the AHUB clock and > looking at the current t210 clock patches for mainline this is > TEGRA210_CLK_D_AUDIO. Ok will fix this. > >> Is there a dedicated reset signal for this module? If so, we should >> require a resets property. > > No there does not appear to be. Looking at the documentation the ADMA > would be reset by the main APE reset. Seems to be one reset that resets > most modules in the APE. > > That raises another issue, the ADMA is in the audio power partition and > currently our downstream driver assumes that this is on. I should at > least check this. IIRC the situation downstream w.r.t. the audio power partition is something like: The AGIC is in that power partition. The Linux kernel AGIC driver gets probed before the power domain driver for that domain. (I think) deferred probe doesn't work for the AGIC driver for some reason (perhaps this is just a bug or oversight in our downstream AGIC driver?). So, the AGIC driver can't turn on the audio power domain. So, there's a hack in the bootloader to turn the power domain on. So, everything in the kernel assumes that power domain is on at boot and so drivers don't have to explicitly control/request that power domain. We should fix this properly so that upstream doesn't make the assumption that the audio power domain is magically on before the kernel boots.