Re: [PATCH v8 v8 06/11] iio: adc: hx711: introduce hx711_chip_info structure

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] The probe function unconditionally fails if `device_get_match_data()` returns NULL, breaking legacy non-DT platform instantiations.
- [Low] The `__counted_by_ptr` macro is undefined in the Linux kernel headers, causing a build failure.
--

commit c0e541cc80817071cecb490c0510348025ebed7a
Author: Piyush Patle <piyushpatle228@gmail.com>

iio: adc: hx711: introduce hx711_chip_info structure

This commit adds a per-variant static configuration structure and populates
the IIO device fields from it at probe time to prepare for more hardware
variants.

> diff --git a/drivers/iio/adc/hx711.c b/drivers/iio/adc/hx711.c
> index 823fd74f9cf95..062d6860da7d6 100644
> --- a/drivers/iio/adc/hx711.c
> +++ b/drivers/iio/adc/hx711.c

[ ... ]

> +/**
> + * struct hx711_chip_info - per-variant static configuration
> + * @name: IIO device name
> + * @channels: channel specification array
> + * @num_channels: number of entries in @channels
> + * @iio_info: IIO info ops for this variant
> + */
> +struct hx711_chip_info {
> +	const char			*name;
> +	const struct iio_chan_spec	*channels __counted_by_ptr(num_channels);

Will this cause a build failure?

It looks like __counted_by_ptr is undefined in the kernel headers. The
compiler might treat this as an undeclared identifier.

[ ... ]

> @@ -473,6 +498,12 @@ static int hx711_probe(struct platform_device *pdev)
>  
>  	mutex_init(&hx711_data->lock);
>  
> +	chip_info = device_get_match_data(dev);
> +	if (!chip_info)
> +		return dev_err_probe(dev, -ENODEV, "missing driver data\n");

Does this introduce a regression for legacy platform instantiations?

For platform devices instantiated by name without a firmware node,
device_get_match_data() returns NULL. Prior to this patch, these devices
probed successfully using hardcoded defaults.

Since this driver does not provide a platform_device_id table, is it
possible we need a fallback to a default configuration, such as &hx711_chip,
to avoid breaking backwards compatibility for non-DT users?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260511174342.123820-1-piyushpatle228@gmail.com?part=6