From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from lb3-smtp-cloud3.xs4all.net ([194.109.24.30]:52422 "EHLO lb3-smtp-cloud3.xs4all.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757645AbcAKJP1 (ORCPT ); Mon, 11 Jan 2016 04:15:27 -0500 Subject: Re: [RFC] ADV7604: VGA support To: Jean-Michel Hautbois References: <561B89FD.4010802@xs4all.nl> Cc: Linux Media Mailing List , Lars-Peter Clausen , Laurent Pinchart From: Hans Verkuil Message-ID: <569372A9.40504@xs4all.nl> Date: Mon, 11 Jan 2016 10:15:21 +0100 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Sender: linux-media-owner@vger.kernel.org List-ID: On 01/07/2016 06:20 PM, Jean-Michel Hautbois wrote: > Hi Hans, > > 2015-10-12 12:22 GMT+02:00 Hans Verkuil : >> On 10/04/2015 06:17 PM, Jean-Michel Hautbois wrote: >>> Hi, >>> >>> I had another look into the ADV7604 HW manual, and I understand that >>> in automatic mode, there is 4 AIN_SEL values possible, determining the >>> connection on AIN pins. >>> Now, having a look at the current ADV76xx files, I can see that two >>> pads are there : >>> ADV7604_PAD_VGA_RGB and ADV7604_PAD_VGA_COMP. >>> >>> According to the manual, my understanding is that we should have four >>> HDMI pads and four analog pads. The latter would be configured as RGB >>> or component, which allows four analog inputs as described in the HW >>> manual. >> >> When I wrote the driver we only needed one VGA input receiving either RGB >> or YCbCr. Hence there is only one analog input and two pads. I wouldn't have >> been able to test the additional analog inputs anyway. >> >> I chose to use pads to select between the two modes, but that's something >> that can be changed (it's not something you can autodetect, unfortunately). >> >> If you want to add support for all four analog inputs, then feel free to >> do so. > > OK, I don't have anything to test the additional inputs either... > Something else so, the adv7604_state struct constains two fields, > .ain_sel and .bus_order. > Should those be parsed from DT (I think so) ? If so, how should it be > named in the port{} node ? Do you need to change this from the current default? If not, then I would leave it as is. If you do have to change this, then you would indeed have to add it to the DT. ain_sel is Analog Devices specific, so just check that whatever you propose is OK for the adv7842 as well. Regarding bus order: this is probably something that can be more generic. I have a deja vu feeling that I saw a patch adding support for this recently, but I can't find it. Does it ring a bell for someone else? Regards, Hans