Re: [PATCH 2/2] iio: adc: ad4080: add support for AD4884

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Incorrect IIO scale calculation results in a factor of 1000 error for exposed voltage scales.
--

commit ec9f31937338c4b8462c0623182623b90d6e39ad
Author: Antoniu Miclaus <antoniu.miclaus@analog.com>

iio: adc: ad4080: add support for AD4884
    
Add support for the AD4884, a dual-channel, 16-bit, 40 MSPS SAR ADC.

> diff --git a/drivers/iio/adc/ad4080.c b/drivers/iio/adc/ad4080.c
> index 265d85ac171a9..764d49eca9e0e 100644
> --- a/drivers/iio/adc/ad4080.c
> +++ b/drivers/iio/adc/ad4080.c
[ ... ]
> @@ -541,6 +542,11 @@ static const struct iio_chan_spec ad4880_channels[] = {
>  	AD4880_CHANNEL_DEFINE(20, 32, 1),
>  };
>  
> +static const struct iio_chan_spec ad4884_channels[] = {
> +	AD4880_CHANNEL_DEFINE(16, 16, 0),
> +	AD4880_CHANNEL_DEFINE(16, 16, 1),
> +};
> +

[Severity: High]
This isn't a bug introduced by this patch, but I noticed that when the channel
exposes the scale attribute, the calculation in ad4080_get_scale() seems
mathematically incorrect:

drivers/iio/adc/ad4080.c:ad4080_get_scale() {
	unsigned int tmp;

	tmp = (st->info->scale_table[0][0] * 1000000ULL) >>
		    st->info->channels[0].scan_type.realbits;
	*val = tmp / 1000000;
	*val2 = tmp % 1000000;

	return IIO_VAL_INT_PLUS_NANO;
}

Since the fraction is calculated by multiplying by 1,000,000 and then using
modulo 1,000,000, val2 holds micro units.

However, IIO_VAL_INT_PLUS_NANO tells the IIO core to format val2 as nano
units (billionths).

Does this mismatch cause the sysfs scale attribute to expose a value that is
exactly 1000 times smaller than intended for these channels?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260522115337.18188-1-antoniu.miclaus@analog.com?part=2