Re: [PATCH v2 4/4] iio: adc: ad7768-1: add support for ADAQ776x-1 ADC Family

[Jonathan Cameron <jic23@kernel.org>]

On Tue, 26 Aug 2025 23:57:44 -0300
Jonathan Santos <jonath4nns@gmail.com> wrote:

> On 08/25, Jonathan Cameron wrote:
> > On Sun, 24 Aug 2025 01:10:13 -0300
> > Jonathan Santos <Jonathan.Santos@analog.com> wrote:
> >   
> > > Add support for ADAQ7767/68/69-1 series, which includes PGIA and
> > > Anti-aliasing filter (AAF) gains. Unlike the AD7768-1, they do not
> > > provide a VCM regulator interface.
> > > 
> > > The PGA gain is configured in run-time through the scale attribute,
> > > if supported by the device. PGA is enabled and controlled by the GPIO
> > > controller (GPIOs 0, 1 and 2) provided by the device with the SPI
> > > interface.
> > > 
> > > The AAF gain is defined by hardware connections and should be specified
> > > in device tree.
> > > 
> > > Signed-off-by: Jonathan Santos <Jonathan.Santos@analog.com>  
> > Hi Jonathan
> > 
> > Various minor comments inline.
> > 
> > Thanks,
> > 
> > Jonathan
> >   
> ...
> > >  	/* Register VCM output regulator */
> > > -	ret = ad7768_register_regulators(&spi->dev, st, indio_dev);
> > > -	if (ret)
> > > -		return ret;
> > > +	if (st->chip->has_vcm_regulator) {
> > > +		ret = ad7768_register_regulators(&spi->dev, st, indio_dev);
> > > +		if (ret)
> > > +			return ret;
> > > +	}
> > > +
> > > +	st->aaf_gain = AD7768_AAF_IN1;
> > > +	ret = device_property_read_u32(&spi->dev, "adi,gain-milli", &val);
> > > +	if (ret && st->chip->has_variable_aaf)
> > > +		dev_warn(&spi->dev, "AAF gain not specified, using default\n");
> > > +
> > > +	if (!ret && !st->chip->has_variable_aaf)
> > > +		dev_warn(&spi->dev, "AAF gain not supported for %s\n", st->chip->name);  
> > 
> > I'm confused.  If you hit this warn, you then go ahead and set it anyway. How does that
> > work?
> >   
> 
> I would be setting a variable that will not be used, so no functional impact.
> I could add another condition to avoid assigning it, or do you believe it
> is better to return error like before? 
Just don't set it to avoid confusion.  Alternative would be a comment that says that
it won't be used, which is probably more complex than the code.

> 
> > > +
> > > +	if (!ret) {  
> > use a local bool for ret here with a suitable name. Carrying on using ret is
> > fragile against later code changes where someone doesn't realise it is still in use.
> >   
> 
> Maybe it is better to handle this in a helper function and apply the
> suggestion.

If that works, sounds good!

J
> 
> > > +		/* If provided, validate and set the gain */
> > > +		switch (val) {
> > > +		case 1000:
> > > +			st->aaf_gain = AD7768_AAF_IN1;
> > > +			break;
> > > +		case 364:
> > > +			st->aaf_gain = AD7768_AAF_IN2;
> > > +			break;
> > > +		case 143:
> > > +			st->aaf_gain = AD7768_AAF_IN3;
> > > +			break;
> > > +		default:
> > > +			return dev_err_probe(&spi->dev, -EINVAL,
> > > +					     "Invalid firmware provided AAF gain\n");
> > > +		}
> > > +	}  
> 
> Regards,
> Jonathan S.
>