On Mon, Apr 28, 2014 at 10:58:26AM +0530, Keerthy wrote:
> index 6a79328..5ddb220 100644
> --- a/drivers/regulator/Kconfig
> +++ b/drivers/regulator/Kconfig
> @@ -384,6 +384,18 @@ config REGULATOR_PALMAS
> on the muxing. This is handled automatically in the driver by
> reading the mux info from OTP.
>
> +config REGULATOR_TPS65917
> + tristate "TI TPS65917 PMIC Regulators"
> + depends on MFD_TPS65917
Please keep the Kconfig and Makefile sorted.
> +static const struct regs_info tps65917_regs_info[] = {
> + {
> + .name = "SMPS1",
> + .sname = "smps1-in",
> + .vsel_addr = TPS65917_SMPS1_VOLTAGE,
> + .ctrl_addr = TPS65917_SMPS1_CTRL,
> + .sleep_id = TPS65917_EXTERNAL_REQSTR_ID_SMPS1,
> + },
> +static int tps65917_is_enabled_smps(struct regulator_dev *dev)
> +{
> + struct tps65917_pmic *pmic = rdev_get_drvdata(dev);
> + int id = rdev_get_id(dev);
> + unsigned int reg;
> +
> + tps65917_smps_read(pmic->tps65917,
> + tps65917_regs_info[id].ctrl_addr, ®);
> +
> + reg &= TPS65917_SMPS1_CTRL_STATUS_MASK;
> + reg >>= TPS65917_SMPS1_CTRL_STATUS_SHIFT;
> +
> + return !!(reg);
> +}
A lot of these functions look like they could be implemented using the
standard helpers rather than open coded - the helpers now support
multi-bit enable fields.
> + if (selector == 0)
> + return 0;
> + else if (selector < 6)
> + return 500000 * mult;
> + else
> + /* Voltage is linear mapping starting from selector 6,
> + * volt = (0.49V + ((selector - 5) * 0.01V)) * RANGE
> + * RANGE is either x1 or x2
> + */
> + return (490000 + ((selector - 5) * 10000)) * mult;
> +}
This looks like the driver could just use linear ranges.
> +static struct of_device_id of_tps65917_match_tbl[] = {
> + { .compatible = "ti,tps65917-pmic", },
> +};
Missing null entry at the end of the table here.
Attachment:
signature.asc
Description: Digital signature
