public inbox for linux-media@vger.kernel.org
 help / color / mirror / Atom feed
From: "Richard Röjfors" <richard.rojfors@pelagicore.com>
To: Rodolfo Giometti <giometti@enneenne.com>
Cc: Guennadi Liakhovetski <g.liakhovetski@gmx.de>,
	Linux Media Mailing List <linux-media@vger.kernel.org>,
	Mauro Carvalho Chehab <mchehab@redhat.com>
Subject: Re: adv7180 as SoC camera device
Date: Tue, 23 Feb 2010 00:19:13 +0100	[thread overview]
Message-ID: <4B8310F1.8070005@pelagicore.com> (raw)
In-Reply-To: <20100222160139.GL21778@enneenne.com>

On 2/22/10 5:01 PM, Rodolfo Giometti wrote:
> On Fri, Feb 19, 2010 at 08:36:38PM +0100, Guennadi Liakhovetski wrote:
>> On Fri, 19 Feb 2010, Rodolfo Giometti wrote:
>>
>>> Hello,
>>>
>>> on my pxa27x based board I have a adv7180 connected with the CIF
>>> interface. Due this fact I'm going to use the pxa_camera.c driver
>>> which in turn registers a soc_camera_host.
>>>
>>> In the latest kernel I found your driver for the ADV7180, but it
>>> registers the chip as a v4l sub device.
>>>
>>> I suppose these two interfaces are not compatible, aren't they?
>>
>> Congratulations! Thereby you're in a position to develop the first 
>> v4l2-subdev / soc-camera universal driver;) The answer to this your 
>> question is - they are... kinda. This means - yes, soc-camera is also 
>> using the v4l2-subdev API, but - with a couple of additions. Basically, 
>> there are two things you have to change in the adv7180 driver to make it 
>> compatible with soc-camera - (1) add bus-configuration methods, even if 
>> they don't do much (see .query_bus_param() and .set_bus_param() methods 
>> from struct soc_camera_ops), and (2) migrate the driver to the mediabus 
>> API. The latter one requires some care - in principle, mediabus should be 
>> the future API to negotiate parameters on the video bus between bridges 
>> (in your case PXA CIF) and clients, but for you this means you also have 
>> to migrate any other bridge drivers in the mainline to that API, and, if 
>> they also interface to some other subdevices - those too, and if those can 
>> also work with other bridges - those too...;) But, I think, that chain 
>> will terminate quite soon, in fact, I cannot find any users of that driver 
>> currently in the mainline, Richard?
>>
>>> In this situation, should I write a new driver for the
>>> soc_camera_device? Which is The-Right-Thing(TM) to do? :)
>>
>> Please, have a look and try to convert the driver as described above. All 
>> the APIs and a few examples are in the mainline, so, you should have 
>> enough copy-paste sources;) Ask on the list (with me on cc) if anything is 
>> still unclear.
> 
> Thanks for your quick answer! :)
> 
> What I still don't understand is if should I move the driver form
> v4l2-subdev to a soc_camera device or trying to support both API...
> 
> It seems to me that the driver is not used by any machines into
> mainline so if soc-camera is also using the v4l2-subdev API but with a
> couple of additions I suppose I can move it to soc_camera API...
> 
> Is that right?

We use it as a subdev to a driver not yet committed from us. So I think
you should extend it, not move it.

--Richard

  parent reply	other threads:[~2010-02-22 23:19 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-02-19 17:44 adv7180 as SoC camera device Rodolfo Giometti
2010-02-19 19:36 ` Guennadi Liakhovetski
2010-02-22 16:01   ` Rodolfo Giometti
2010-02-22 16:11     ` Hans Verkuil
2010-02-22 16:15       ` Rodolfo Giometti
2010-02-22 16:46         ` Guennadi Liakhovetski
2010-02-22 16:16     ` Guennadi Liakhovetski
2010-02-22 16:53       ` Jonathan Cameron
2010-02-22 23:19     ` Richard Röjfors [this message]
2010-03-30 14:06       ` Rodolfo Giometti
2010-03-30 14:40         ` Guennadi Liakhovetski
2010-03-30 15:42           ` Rodolfo Giometti
2010-04-02  7:06             ` Guennadi Liakhovetski
2010-04-01 11:48         ` Rodolfo Giometti
2010-04-02  7:09           ` Guennadi Liakhovetski
2010-04-02  7:44             ` Rodolfo Giometti

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4B8310F1.8070005@pelagicore.com \
    --to=richard.rojfors@pelagicore.com \
    --cc=g.liakhovetski@gmx.de \
    --cc=giometti@enneenne.com \
    --cc=linux-media@vger.kernel.org \
    --cc=mchehab@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox