Re: [PATCH] hwmon: (ntc_thermistor): try reading processed

[Guenter Roeck <linux@roeck-us.net>]

On 12/19/20 2:41 PM, Linus Walleij wrote:
> Before trying the custom method of reading the sensor
> as raw and then converting assuming 1000 scaling, just
> use iio_read_channel_processed() which first tries to
> see if the ADC can provide a processed value directly,
> else reads raw and applies scaling inside of IIO
> using the scale attributes of the ADC.
> 
> The code that hardcodes scaling to 1000 and assumes
> a 12bit ADC is very dubious. I keep it around here
> but I have a strong urge to just delete it.
> 
> This gives correct readings on the AB8500 thermistor
> inputs used in the Ux500 HREFP520 platform for reading
> battery and board temperature.
> 
> Cc: Peter Rosin <peda@axentia.se>
> Cc: Chris Lesiak <chris.lesiak@licor.com>
> Cc: Jonathan Cameron <jic23@cam.ac.uk>
> Cc: linux-iio@vger.kernel.org
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> ---
>  drivers/hwmon/ntc_thermistor.c | 29 ++++++++++++++++++++---------
>  1 file changed, 20 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/hwmon/ntc_thermistor.c b/drivers/hwmon/ntc_thermistor.c
> index 3aad62a0e661..ac0d80faddf6 100644
> --- a/drivers/hwmon/ntc_thermistor.c
> +++ b/drivers/hwmon/ntc_thermistor.c
> @@ -326,18 +326,29 @@ struct ntc_data {
>  static int ntc_adc_iio_read(struct ntc_thermistor_platform_data *pdata)
>  {
>  	struct iio_channel *channel = pdata->chan;
> -	int raw, uv, ret;
> +	int uv, ret;
>  
> -	ret = iio_read_channel_raw(channel, &raw);
> +	/* A processed voltage channel will return microvolts */
> +	ret = iio_read_channel_processed(channel, &uv);
>  	if (ret < 0) {
> -		pr_err("read channel() error: %d\n", ret);
> -		return ret;
> -	}
> +		int raw;
>  
> -	ret = iio_convert_raw_to_processed(channel, raw, &uv, 1000);
> -	if (ret < 0) {
> -		/* Assume 12 bit ADC with vref at pullup_uv */
> -		uv = (pdata->pullup_uv * (s64)raw) >> 12;
> +		/*
> +		 * FIXME: This fallback to using a raw read and then right
> +		 * out assume the ADC is 12 bits and hard-coding scale
> +		 * to 1000 seems a bit dangerous. Should it simply be
> +		 * deleted?
> +		 */

The hwmon ABI specifically supports unscaled values, which can then be
scaled in userspace using the sensors configuration file.
Given that we return the pseudo-scaled value to userspace today,
it seems to me that it would do more harm to change that instead of just
leaving it in place.

Either case, calling iio_convert_raw_to_processed() does not seem to add
value here. This is already done by iio_read_channel_processed().
The best we can do is to use the original fallback.

Guenter

> +		ret = iio_read_channel_raw(channel, &raw);
> +		if (ret < 0) {
> +			pr_err("read channel() error: %d\n", ret);
> +			return ret;
> +		}
> +		ret = iio_convert_raw_to_processed(channel, raw, &uv, 1000);
> +		if (ret < 0) {
> +			/* Assume 12 bit ADC with vref at pullup_uv */
> +			uv = (pdata->pullup_uv * (s64)raw) >> 12;
> +		}
>  	}
>  
>  	return uv;
>