[v1,1/3] dt-bindings: iio: adc: ti,tsc2046: add vref-supply property

Add property for the voltage reference supply.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
---
 Documentation/devicetree/bindings/iio/adc/ti,tsc2046.yaml | 3 +++
 1 file changed, 3 insertions(+)

On Tue, 30 Aug 2022 13:07:09 +0200
Oleksij Rempel <o.rempel@pengutronix.de> wrote:

> Add spi_device_id to silent following warning:
>  SPI driver tsc2046 has no spi_device_id for ti,tsc2046e-adc
> 
> Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
> ---
>  drivers/iio/adc/ti-tsc2046.c | 17 ++++++++++++++++-
>  1 file changed, 16 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/iio/adc/ti-tsc2046.c b/drivers/iio/adc/ti-tsc2046.c
> index bbc8b4137b0b1..b9a1fac659d46 100644
> --- a/drivers/iio/adc/ti-tsc2046.c
> +++ b/drivers/iio/adc/ti-tsc2046.c
> @@ -761,7 +761,15 @@ static int tsc2046_adc_probe(struct spi_device *spi)
>  		return -EINVAL;
>  	}
>  
> -	dcfg = device_get_match_data(dev);
> +	if (!dev_fwnode(dev)) {
> +		const struct spi_device_id *id;
> +
> +		id = spi_get_device_id(spi);
> +		dcfg = (const struct tsc2046_adc_dcfg *)id->driver_data;

Driver data not set below.

Otherwise this looks good to me.  An alternative more common form (I think...)
is call device_get_match_data() unconditionally and if that is null follow
the driver_data path. Either way is fine though.

Could you add to the patch description where
the warning is coming from?   Build time / runtime etc and what tool?

Thanks,

Jonathan


> +	} else {
> +		dcfg = device_get_match_data(dev);
> +	}
> +
>  	if (!dcfg)
>  		return -EINVAL;
>  
> @@ -878,11 +886,18 @@ static const struct of_device_id ads7950_of_table[] = {
>  };
>  MODULE_DEVICE_TABLE(of, ads7950_of_table);
>  
> +static const struct spi_device_id tsc2046_adc_spi_ids[] = {
> +	{ "tsc2046e-adc" },
> +	{ }
> +};
> +MODULE_DEVICE_TABLE(spi, tsc2046_adc_spi_ids);
> +
>  static struct spi_driver tsc2046_adc_driver = {
>  	.driver = {
>  		.name = "tsc2046",
>  		.of_match_table = ads7950_of_table,
>  	},
> +	.id_table = tsc2046_adc_spi_ids,
>  	.probe = tsc2046_adc_probe,
>  	.remove = tsc2046_adc_remove,
>  };

On 30/08/2022 14:07, Oleksij Rempel wrote:
> Add property for the voltage reference supply.
> 
> Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>


Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>


Best regards,
Krzysztof

On Tue, Aug 30, 2022 at 02:02:28PM +0100, Jonathan Cameron wrote:
> On Tue, 30 Aug 2022 13:07:09 +0200
> Oleksij Rempel <o.rempel@pengutronix.de> wrote:
> 
> > Add spi_device_id to silent following warning:
> >  SPI driver tsc2046 has no spi_device_id for ti,tsc2046e-adc
> > 
> > Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
> > ---
> >  drivers/iio/adc/ti-tsc2046.c | 17 ++++++++++++++++-
> >  1 file changed, 16 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/iio/adc/ti-tsc2046.c b/drivers/iio/adc/ti-tsc2046.c
> > index bbc8b4137b0b1..b9a1fac659d46 100644
> > --- a/drivers/iio/adc/ti-tsc2046.c
> > +++ b/drivers/iio/adc/ti-tsc2046.c
> > @@ -761,7 +761,15 @@ static int tsc2046_adc_probe(struct spi_device *spi)
> >  		return -EINVAL;
> >  	}
> >  
> > -	dcfg = device_get_match_data(dev);
> > +	if (!dev_fwnode(dev)) {
> > +		const struct spi_device_id *id;
> > +
> > +		id = spi_get_device_id(spi);
> > +		dcfg = (const struct tsc2046_adc_dcfg *)id->driver_data;
> 
> Driver data not set below.

..facepalm..

> Otherwise this looks good to me.  An alternative more common form (I think...)
> is call device_get_match_data() unconditionally and if that is null follow
> the driver_data path. Either way is fine though.
> 
> Could you add to the patch description where
> the warning is coming from?   Build time / runtime etc and what tool?

ack. It is runtime warning in the kernel log.

Regards,
Oleksij

On Tue, Aug 30, 2022 at 2:19 PM Oleksij Rempel <o.rempel@pengutronix.de> wrote:
>
> Add spi_device_id to silent following warning:
>  SPI driver tsc2046 has no spi_device_id for ti,tsc2046e-adc

Missed period.

...

> -       dcfg = device_get_match_data(dev);

Why remove this and duplicate the check below with the inverted conditional?

> +       if (!dev_fwnode(dev)) {
> +               const struct spi_device_id *id;
> +
> +               id = spi_get_device_id(spi);
> +               dcfg = (const struct tsc2046_adc_dcfg *)id->driver_data;
> +       } else {
> +               dcfg = device_get_match_data(dev);
> +       }

if (!dcfg) {
  ...try SPI ID...
}
if (!dfg)
  return -E...