Re: [PATCH] media: rcar-vin: add G/S_PARM ioctls

From: "Niklas Söderlund" <niklas.soderlund@ragnatech.se>
To: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
Cc: Mauro Carvalho Chehab <mchehab@kernel.org>,
	linux-media@vger.kernel.org, linux-renesas-soc@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	Vladimir Barinov <vladimir.barinov@cogentembedded.com>
Subject: Re: [PATCH] media: rcar-vin: add G/S_PARM ioctls
Date: Fri, 24 Sep 2021 11:54:16 +0200	[thread overview]
Message-ID: <YU2gSJKfsqP+gUci@oden.dyn.berto.se> (raw)
In-Reply-To: <20210924084115.2340-1-nikita.yoush@cogentembedded.com>

Hello Nikita and Vladimir,

Thanks for your patch.

On 2021-09-24 11:41:15 +0300, Nikita Yushchenko wrote:
> From: Vladimir Barinov <vladimir.barinov@cogentembedded.com>
> 
> This adds g/s_parm ioctls for parallel interface.

I would like to ask your use-case for this. I'm trying to make up the 
courage to move Gen2 inline with Gen3, that is move Gen2 to use the 
media graph interface to allow it more complex use-cases, including 
controlling parameters on the subdevice level.

So I'm curious if this solve a particular problem for you or if it's 
done "just" for completeness. If it solves a real problem then I think 
we should move a head with a v2 (with the small comment below fixed) if 
not I would like to try and reduce the non media graph usage of the API 
as much as possible.

> 
> Signed-off-by: Vladimir Barinov <vladimir.barinov@cogentembedded.com>
> Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
> ---
>  drivers/media/platform/rcar-vin/rcar-v4l2.c | 21 +++++++++++++++++++++
>  1 file changed, 21 insertions(+)
> 
> diff --git a/drivers/media/platform/rcar-vin/rcar-v4l2.c b/drivers/media/platform/rcar-vin/rcar-v4l2.c
> index bdeff51bf768..de9f8dd55a30 100644
> --- a/drivers/media/platform/rcar-vin/rcar-v4l2.c
> +++ b/drivers/media/platform/rcar-vin/rcar-v4l2.c
> @@ -527,6 +527,24 @@ static int rvin_s_selection(struct file *file, void *fh,
>  	return 0;
>  }
>  
> +static int rvin_g_parm(struct file *file, void *priv,
> +		       struct v4l2_streamparm *parm)
> +{
> +	struct rvin_dev *vin = video_drvdata(file);
> +	struct v4l2_subdev *sd = vin_to_source(vin);
> +
> +	return v4l2_g_parm_cap(video_devdata(file), sd, parm);

Please use &vin->vdev instead of video_devdata(file). 

> +}
> +
> +static int rvin_s_parm(struct file *file, void *priv,
> +		       struct v4l2_streamparm *parm)
> +{
> +	struct rvin_dev *vin = video_drvdata(file);
> +	struct v4l2_subdev *sd = vin_to_source(vin);
> +
> +	return v4l2_s_parm_cap(video_devdata(file), sd, parm);

Please use &vin->vdev instead of video_devdata(file). 

> +}
> +
>  static int rvin_g_pixelaspect(struct file *file, void *priv,
>  			      int type, struct v4l2_fract *f)
>  {
> @@ -743,6 +761,9 @@ static const struct v4l2_ioctl_ops rvin_ioctl_ops = {
>  	.vidioc_g_selection		= rvin_g_selection,
>  	.vidioc_s_selection		= rvin_s_selection,
>  
> +	.vidioc_g_parm			= rvin_g_parm,
> +	.vidioc_s_parm			= rvin_s_parm,
> +
>  	.vidioc_g_pixelaspect		= rvin_g_pixelaspect,
>  
>  	.vidioc_enum_input		= rvin_enum_input,
> -- 
> 2.30.2
> 

-- 
Regards,
Niklas Söderlund