On Tue, 07 Jan 2025 21:50:21 +0100
Javier Carrasco <javier.carrasco.cruz@xxxxxxxxx> wrote:
> The configuration registers are not volatile and are not affected
> by read operations (i.e. not precious), making them suitable to be
> cached in order to reduce the number of accesses to the device.
>
> Add support for regfields as well to simplify register operations,
> taking into account the different fields for the veml6030/veml7700 and
> veml6035.
>
> Signed-off-by: Javier Carrasco <javier.carrasco.cruz@xxxxxxxxx>
> ---
> drivers/iio/light/veml6030.c | 141 +++++++++++++++++++++++++++++++++++--------
> 1 file changed, 116 insertions(+), 25 deletions(-)
>
> diff --git a/drivers/iio/light/veml6030.c b/drivers/iio/light/veml6030.c
> index 9b71825eea9bee2146be17ed2f30f5a8f7ad37e3..a6385c6d3fba59a6b22845a3c5e252b619faed65 100644
> --- a/drivers/iio/light/veml6030.c
> +++ b/drivers/iio/light/veml6030.c
> @@ -65,6 +65,11 @@ enum veml6030_scan {
> VEML6030_SCAN_TIMESTAMP,
> };
>
> +struct veml6030_rf {
> + struct regmap_field *it;
> + struct regmap_field *gain;
> +};
> +
> struct veml603x_chip {
> const char *name;
> const int(*scale_vals)[][2];
> @@ -75,6 +80,7 @@ struct veml603x_chip {
> int (*set_info)(struct iio_dev *indio_dev);
> int (*set_als_gain)(struct iio_dev *indio_dev, int val, int val2);
> int (*get_als_gain)(struct iio_dev *indio_dev, int *val, int *val2);
> + int (*regfield_init)(struct iio_dev *indio_dev);
With only two fields, why use a callback rather than just adding the two
const struct reg_field into this structure directly?
I'd also be tempted to do the caching and regfield changes as separate patches.
Jonathan
