linux-next.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jacopo Mondi <jacopo@jmondi.org>
To: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>,
	Sakari Ailus <sakari.ailus@linux.intel.com>,
	Linux Media Mailing List <linux-media@vger.kernel.org>,
	Jacopo Mondi <jacopo+renesas@jmondi.org>,
	Stephen Rothwell <sfr@canb.auug.org.au>,
	linux-next@vger.kernel.org
Subject: Re: [PATCH] media: i2c: fix max9271 build dependencies
Date: Mon, 8 Feb 2021 14:55:14 +0100	[thread overview]
Message-ID: <20210208135514.t2wymwvsehdl657m@uno.localdomain> (raw)
In-Reply-To: <20210208143150.03974bca@coco.lan>

Hi Mauro,

On Mon, Feb 08, 2021 at 02:31:50PM +0100, Mauro Carvalho Chehab wrote:
> Em Mon, 8 Feb 2021 14:11:02 +0100
> Mauro Carvalho Chehab <mchehab+huawei@kernel.org> escreveu:
>
> > Em Mon, 8 Feb 2021 12:41:42 +0100
> > Jacopo Mondi <jacopo@jmondi.org> escreveu:
> >
> > > > > If you do, instead:
> > > > >
> > > > >     if VIDEO_V4L2 && I2C
> > > > > 	config VIDEO_MAX9271_SERIALIZER
> > > > > 		tristate
> > > > >
> > > > > 	config VIDEO_RDACM20
> > > > > 		select VIDEO_MAX9271_SERIALIZER
> > > > > 		...
> > > > >
> > > > > 	config VIDEO_RDACM21
> > > > > 		select VIDEO_MAX9271_SERIALIZER
> > > > > 		...
> > > > >     endif
> > > > >
> > > > > Then you also won't need:
> > > > > 	depends on VIDEO_MAX9271_SERIALIZER || !VIDEO_MAX9271_SERIALIZER
> > > > >
> > > > > As select should do the right thing in this case, ensuring that MAX9271
> > > > > will be builtin either if RDACM20 or RDACM21 is builtin.
> > > >
> > > > I also vote for usage of "select".
> > > >
> > >
> > > I would prefer that too, I was concerned about possible un-met
> > > dependencies, as Sakari pointed out, but the current situation is no
> > > better, as the only Kconfig symbols where those can be listed are the
> > > camera modules one.
> >
> > Works for me. I'll make a patch for it.
>
> Hmm... after taking a deeper look at the rcma20 drivers, and on its
> Kconfig help text:
>
> 	config VIDEO_RDACM20
> 		tristate "IMI RDACM20 camera support"
> 		select V4L2_FWNODE
> 		select VIDEO_V4L2_SUBDEV_API
> 		select MEDIA_CONTROLLER
> 		help
> 		  This driver supports the IMI RDACM20 GMSL camera, used in
> 		  ADAS systems.
>
> 		  This camera should be used in conjunction with a GMSL
> 		  deserialiser such as the MAX9286.
>
> I'm starting to suspect that there's something very wrong here...
>
> The help text mentions the MAX9286 driver, which is a complete
> driver, and not MAX9271, which seems to implement a set of PHY functions
> needed by those drivers, and which lacks a proper I2C binding code on it.

What is it puzzling you here ? The fact max9286 is mentioned ?
Maybe it is not clear but the max9286 and max9271 are, respectively,
the deserializer and serializers chips that form a GMSL link.

Camera modules usually embed an image sensor (plus a variety of
ISP/uControllers for internal image processing) whose output is
directed to an embedded GMSL serializer (the max9271), which captures
the image output and serializes it on the GMSL link.

On the other side of the link a GMSLa deserializer (the max9286) is
required, to receive and interpret the GMSL signal and convert it back
to an image stream then transmitted though a MIPI CSI-2 interface to
the SoC.

Maybe the last statement is redundant and should not be placed in the
camera module Kconfig description, as system integrators are of course
aware that a deserializer is required on the other side of the link ?

>
> The I2C binding code is, instead, inside RDACM20 and RDACM21:
>
> 	static int rdacm21_initialize(struct rdacm21_device *dev)
> 	{
> 		int ret;
>
> 		/* Verify communication with the MAX9271: ping to wakeup. */
> 		dev->serializer.client->addr = MAX9271_DEFAULT_ADDR;
> 		i2c_smbus_read_byte(dev->serializer.client);
> 		usleep_range(3000, 5000);
>
> 		/* Enable reverse channel and disable the serial link. */
> 		ret = max9271_set_serial_link(&dev->serializer, false);
> 		if (ret)
> 			return ret;
>
> 		/* Configure I2C bus at 105Kbps speed and configure GMSL. */
> 		ret = max9271_configure_i2c(&dev->serializer,
> 					    MAX9271_I2CSLVSH_469NS_234NS |
> 					    MAX9271_I2CSLVTO_1024US |
> 					    MAX9271_I2CMSTBT_105KBPS);
>
> 		/* Several other max9271-specific init code */
>
> 		ret = ov490_initialize(dev);
> 		if (ret)
> 			return ret;
>
> And, at max9271 "driver", there's just a bunch of exported functions.
>

max9271 is a library module that provides functions for other drivers to use.
The MAX9271 chip alone has no actual use, as it is usually embedded in
a camera module with an image sensor (and other chips).

The binding documents of the MAX9286 and RDACM2x chips provides an
overview of the system architecture.

The max9286 bounces all the I2C messages it receives and which are
not directed to its address to the other side of the GMSL link, and
the configurations bits you have pasted here performs the intial setup
of the I2C-over-GMSL interface of the serializer.

What is the 'i2c binding code' that is bothering you ?

> This is all wrong.
>
> I'm seriously considering to move all those 3 drivers to staging,
> while they're not fixed to use a proper I2C binding mechanism.
>
> Thanks,
> Mauro

  parent reply	other threads:[~2021-02-08 13:55 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-08  0:32 linux-next: build warning after merge of the v4l-dvb tree Stephen Rothwell
2021-02-08  6:30 ` Mauro Carvalho Chehab
2021-02-08  8:33   ` Geert Uytterhoeven
2021-02-08  8:49     ` Mauro Carvalho Chehab
2021-02-08  8:52       ` Geert Uytterhoeven
2021-02-08  9:14         ` Mauro Carvalho Chehab
2021-02-08  6:53 ` [PATCH] media: i2c: fix max9271 build dependencies Mauro Carvalho Chehab
2021-02-08  7:27   ` Sakari Ailus
2021-02-08  8:36     ` Jacopo Mondi
2021-02-08  8:41       ` Sakari Ailus
2021-02-08  8:59         ` Jacopo Mondi
2021-02-08  9:03           ` Sakari Ailus
2021-02-08  9:08         ` Mauro Carvalho Chehab
2021-02-08  9:24           ` Sakari Ailus
2021-02-08 10:07             ` Mauro Carvalho Chehab
2021-02-08 11:32               ` Laurent Pinchart
2021-02-08 11:41                 ` Jacopo Mondi
2021-02-08 13:11                   ` Mauro Carvalho Chehab
2021-02-08 13:31                     ` Mauro Carvalho Chehab
2021-02-08 13:40                       ` Laurent Pinchart
2021-02-08 14:23                         ` Mauro Carvalho Chehab
2021-02-08 14:31                           ` Laurent Pinchart
2021-02-08 15:01                             ` Mauro Carvalho Chehab
2021-02-08 13:55                       ` Jacopo Mondi [this message]
2021-02-08 14:14                         ` Mauro Carvalho Chehab
2021-02-08 14:49                           ` Jacopo Mondi
2021-02-08 15:42                             ` Mauro Carvalho Chehab
2021-02-08 16:26                               ` Laurent Pinchart

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=20210208135514.t2wymwvsehdl657m@uno.localdomain \
    --to=jacopo@jmondi.org \
    --cc=jacopo+renesas@jmondi.org \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-next@vger.kernel.org \
    --cc=mchehab+huawei@kernel.org \
    --cc=sakari.ailus@linux.intel.com \
    --cc=sfr@canb.auug.org.au \
    /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;
as well as URLs for NNTP newsgroup(s).