Re: [PATCH v9 8/8] media: imx214: Add new control with V4L2_CID_UNIT_CELL_SIZE

From: Ricardo Ribalda Delgado
Date: Fri Oct 04 2019 - 06:35:18 EST


Hi Hans

On Fri, Oct 4, 2019 at 12:31 PM Hans Verkuil <hverkuil-cisco@xxxxxxxxx> wrote:
>
> On 10/1/19 1:24 PM, Ricardo Ribalda Delgado wrote:
> > According to the product brief, the unit cell size is 1120 nanometers^2.
> >
> > https://www.sony-semicon.co.jp/products_en/IS/sensor1/img/products/ProductBrief_IMX214_20150428.pdf
> >
> > Signed-off-by: Ricardo Ribalda Delgado <ribalda@xxxxxxxxxx>
> > ---
> > drivers/media/i2c/imx214.c | 9 +++++++++
> > 1 file changed, 9 insertions(+)
> >
> > diff --git a/drivers/media/i2c/imx214.c b/drivers/media/i2c/imx214.c
> > index 159a3a604f0e..adcaaa8c86d1 100644
> > --- a/drivers/media/i2c/imx214.c
> > +++ b/drivers/media/i2c/imx214.c
> > @@ -47,6 +47,7 @@ struct imx214 {
> > struct v4l2_ctrl *pixel_rate;
> > struct v4l2_ctrl *link_freq;
> > struct v4l2_ctrl *exposure;
> > + struct v4l2_ctrl *unit_size;
> >
> > struct regulator_bulk_data supplies[IMX214_NUM_SUPPLIES];
> >
> > @@ -948,6 +949,10 @@ static int imx214_probe(struct i2c_client *client)
> > static const s64 link_freq[] = {
> > IMX214_DEFAULT_LINK_FREQ,
> > };
> > + static const struct v4l2_area unit_size = {
> > + .width = 1120,
> > + .height = 1120,
> > + };
> > int ret;
> >
> > ret = imx214_parse_fwnode(dev);
> > @@ -1029,6 +1034,10 @@ static int imx214_probe(struct i2c_client *client)
> > V4L2_CID_EXPOSURE,
> > 0, 3184, 1, 0x0c70);
> >
> > + imx214->unit_size = v4l2_ctrl_new_std_compound(&imx214->ctrls,
> > + NULL,
> > + V4L2_CID_UNIT_CELL_SIZE,
> > + v4l2_ctrl_ptr_create((void *)&unit_size));
>
> The imx214 supports two modes: 4096x2304 and 1920x1080. I assume that the
> latter is using binning? So shouldn't the unit cell size be different in that
> case?

To be honest I do not know if it is cropping or binning. I can try to
turn on the sensor and figure out if the field of view changes.

>
> I'm not so sure the unit cell size should depend on binning. I'd rather have
> the binning information exposed somehow and let userspace do the calculation.

I think there are good arguments on both directions ;).

The issue with considering the binning is that if there a later step
on the pipeline that does the bining then it has to change also the
unit_cell_size control. I would say, for now lets keep them
independent.


>
> Regards,
>
> Hans
>
> > ret = imx214->ctrls.error;
> > if (ret) {
> > dev_err(&client->dev, "%s control init failed (%d)\n",
> >
>