Re: [PATCH v2] iio: light sensor: Improve granularity of tsl2583lux values.

[Jonathan Cameron]

On 12/02/2011 10:19 PM, Bryan Freed wrote:
> When illuminance0_calibbias gets 4000 (for a 4x multiplier), we see lux
> granularity of 4.  Reversing the order of the right shift and multiplication
> retains the precision of the unadjusted lux value.
Looks sensible to me, but having said that, Greg, please leave a little
while for Jon Brenner to get to it as he clearly knows this driver
a lot better than me!
> 
> Signed-off-by: Bryan Freed <bfreed@xxxxxxxxxxxx>
Acked-by: Jonathan Cameron <jic23@xxxxxxxxxx>
> ---
>  drivers/staging/iio/light/tsl2583.c |   17 ++++++++++++++---
>  1 files changed, 14 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/staging/iio/light/tsl2583.c b/drivers/staging/iio/light/tsl2583.c
> index 80f77cf..25f4c1d 100644
> --- a/drivers/staging/iio/light/tsl2583.c
> +++ b/drivers/staging/iio/light/tsl2583.c
> @@ -194,6 +194,7 @@ static int taos_get_lux(struct iio_dev *indio_dev)
>  {
>  	u16 ch0, ch1; /* separated ch0/ch1 data from device */
>  	u32 lux; /* raw lux calculated from device data */
> +	u64 lux64;
>  	u32 ratio;
>  	u8 buf[5];
>  	struct taos_lux *p;
> @@ -297,9 +298,19 @@ static int taos_get_lux(struct iio_dev *indio_dev)
>  		lux = (lux + (chip->als_time_scale >> 1)) /
>  			chip->als_time_scale;
>  
> -	/* adjust for active gain scale */
> -	lux >>= 13; /* tables have factor of 8192 builtin for accuracy */
> -	lux = (lux * chip->taos_settings.als_gain_trim + 500) / 1000;
> +	/* Adjust for active gain scale.
> +	 * The taos_device_lux tables above have a factor of 8192 built in,
> +	 * so we need to shift right.
> +	 * User-specified gain provides a multiplier.
> +	 * Apply user-specified gain before shifting right to retain precision.
> +	 * Use 64 bits to avoid overflow on multiplication.
> +	 * Then go back to 32 bits before division to avoid using div_u64().
> +	 */
> +	lux64 = lux;
> +	lux64 = lux64 * chip->taos_settings.als_gain_trim;
> +	lux64 >>= 13;
> +	lux = lux64;
> +	lux = (lux + 500) / 1000;
>  	if (lux > TSL258X_LUX_CALC_OVER_FLOW) { /* check for overflow */
>  return_max:
>  		lux = TSL258X_LUX_CALC_OVER_FLOW;

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/