Re: [PATCH 4/5] media: i2c: imx290: Add support to enumerate all frame sizes

From: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
To: karthik poduval <karthik.poduval@gmail.com>
Cc: Sakari Ailus <sakari.ailus@iki.fi>,
	mchehab@kernel.org, linux-media@vger.kernel.org,
	linux-kernel@vger.kernel.org, devicetree@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, c.barrett@framos.com,
	a.brela@framos.com, peter.griffin@linaro.org
Subject: Re: [PATCH 4/5] media: i2c: imx290: Add support to enumerate all frame sizes
Date: Mon, 16 Dec 2019 09:34:34 +0530	[thread overview]
Message-ID: <20191216040434.GA14778@Mani-XPS-13-9360> (raw)
In-Reply-To: <CAFP0Ok8Vqze8ZRyT1WvMXZeBLcx7oKcTO1Kad4kSFLbpHkok-A@mail.gmail.com>

On Sun, Dec 15, 2019 at 03:52:37PM -0800, karthik poduval wrote:
> What if someone adds RAW8 or RAW14 formats in future the enum frame sizes
> code doesn't have to be patched again if written using a loop on formats
> array.
> 

Please don't top post :)

IMX290 only supports RAW10 and RAW12 formats. And I don't think this driver
can handle any other CMOS sensors from Sony, so looping over imx290_formats
seems unnecessary to me.

Thanks,
Mani

> On Sun, Dec 15, 2019, 9:49 AM Manivannan Sadhasivam <
> manivannan.sadhasivam@linaro.org> wrote:
> 
> > Hi Sakari,
> >
> > On Tue, Dec 03, 2019 at 10:56:04AM +0200, Sakari Ailus wrote:
> > > On Sat, Nov 30, 2019 at 12:35:40AM +0530, Manivannan Sadhasivam wrote:
> > > > Add support to enumerate all frame sizes supported by IMX290. This is
> > > > required for using with userspace tools such as libcamera.
> > > >
> > > > Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org
> > >
> > > > ---
> > > >  drivers/media/i2c/imx290.c | 20 ++++++++++++++++++++
> > > >  1 file changed, 20 insertions(+)
> > > >
> > > > diff --git a/drivers/media/i2c/imx290.c b/drivers/media/i2c/imx290.c
> > > > index d5bb3a59ac46..f26c4a0ee0a0 100644
> > > > --- a/drivers/media/i2c/imx290.c
> > > > +++ b/drivers/media/i2c/imx290.c
> > > > @@ -468,6 +468,25 @@ static int imx290_enum_mbus_code(struct
> > v4l2_subdev *sd,
> > > >     return 0;
> > > >  }
> > > >
> > > > +static int imx290_enum_frame_size(struct v4l2_subdev *subdev,
> > > > +                             struct v4l2_subdev_pad_config *cfg,
> > > > +                             struct v4l2_subdev_frame_size_enum *fse)
> > > > +{
> > > > +   if ((fse->code != imx290_formats[0].code) &&
> > > > +       (fse->code != imx290_formats[1].code))
> > >
> > > Please use a loop over imx290_formats instead.
> > >
> >
> > May I know why? What benefit does it provide over current method?
> >
> > Thanks,
> > Mani
> >
> > > > +           return -EINVAL;
> > > > +
> > > > +   if (fse->index >= ARRAY_SIZE(imx290_modes))
> > > > +           return -EINVAL;
> > > > +
> > > > +   fse->min_width = imx290_modes[fse->index].width;
> > > > +   fse->max_width = imx290_modes[fse->index].width;
> > > > +   fse->min_height = imx290_modes[fse->index].height;
> > > > +   fse->max_height = imx290_modes[fse->index].height;
> > > > +
> > > > +   return 0;
> > > > +}
> > > > +
> > > >  static int imx290_get_fmt(struct v4l2_subdev *sd,
> > > >                       struct v4l2_subdev_pad_config *cfg,
> > > >                       struct v4l2_subdev_format *fmt)
> > > > @@ -820,6 +839,7 @@ static const struct v4l2_subdev_video_ops
> > imx290_video_ops = {
> > > >  static const struct v4l2_subdev_pad_ops imx290_pad_ops = {
> > > >     .init_cfg = imx290_entity_init_cfg,
> > > >     .enum_mbus_code = imx290_enum_mbus_code,
> > > > +   .enum_frame_size = imx290_enum_frame_size,
> > > >     .get_fmt = imx290_get_fmt,
> > > >     .set_fmt = imx290_set_fmt,
> > > >  };
> > >
> > > --
> > > Regards,
> > >
> > > Sakari Ailus
> >