Re: [PATCH v4 2/4] iio: adc: qcom-pm8xxx-xoadc: remove redundant error logs when reading values

Re: [PATCH v4 2/4] iio: adc: qcom-pm8xxx-xoadc: remove redundant error logs when reading values

[Andy Shevchenko]

On Sat, Apr 25, 2026 at 08:39:16PM +0000, Antony Kurniawan Soemardi wrote:
> Drop dev_err() logging for -EINVAL and -ETIMEDOUT cases and rely on
> return values to report errors, reducing unnecessary log noise.

Assuming that this error codes go to the user space without changes,
this LGTM, FWIW,
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>

The EINVAL --> ENOENT can be discussed / changed separately later on.

...

> +	if (!wait_for_completion_timeout(&adc->complete,
> +					 VADC_CONV_TIME_MAX_US)) {

I would even dare to make this a single line (82 characters).

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v4 3/4] iio: adc: qcom-pm8xxx-xoadc: constify pointers in read operations

[Andy Shevchenko]

On Sat, Apr 25, 2026 at 08:39:20PM +0000, Antony Kurniawan Soemardi wrote:
> These functions do not modify the indio_dev & adc structure, so they
> should be marked const to allow the compiler to catch any accidental
> modifications and to document the read only nature of these functions.

...

> -	struct pm8xxx_xoadc *adc = iio_priv(indio_dev);
> +	const struct iio_dev *indio_dev = d;
> +	const struct pm8xxx_xoadc *adc = iio_priv(indio_dev);
>  
>  	complete(&adc->complete);

Pointer to a constant object assumes no modification to that, isn't it? Please,
compile your code first.


-- 
With Best Regards,
Andy Shevchenko