On Tue, Sep 28, 2021 at 02:22:41AM -0700, Zev Weiss wrote:
> The appropriate mantissa values for the lm25066 family's direct-format
> current and power readings are a function of the sense resistor
> employed between the SENSE and VIN pins of the chip. Instead of
> assuming that resistance is always the same 1mOhm as used in the
> datasheet, allow it to be configured via a device-tree property
> ("shunt-resistor-micro-ohms").
>
> Signed-off-by: Zev Weiss <zev@xxxxxxxxxxxxxxxxx>
> ---
> Documentation/hwmon/lm25066.rst | 2 ++
> drivers/hwmon/pmbus/lm25066.c | 12 ++++++++++++
> 2 files changed, 14 insertions(+)
>
> diff --git a/Documentation/hwmon/lm25066.rst b/Documentation/hwmon/lm25066.rst
> index 9f1d7e4d3ca1..a2098eb24090 100644
> --- a/Documentation/hwmon/lm25066.rst
> +++ b/Documentation/hwmon/lm25066.rst
> @@ -79,6 +79,8 @@ This driver does not auto-detect devices. You will have to instantiate the
> devices explicitly. Please see Documentation/i2c/instantiating-devices.rst for
> details.
>
> +The shunt (sense) resistor value can be configured by a device tree property;
> +see Documentation/devicetree/bindings/hwmon/pmbus/ti,lm25066.yaml for details.
>
> Platform data support
> ---------------------
> diff --git a/drivers/hwmon/pmbus/lm25066.c b/drivers/hwmon/pmbus/lm25066.c
> index 18d5a76f346d..29e848bcd436 100644
> --- a/drivers/hwmon/pmbus/lm25066.c
> +++ b/drivers/hwmon/pmbus/lm25066.c
> @@ -458,6 +458,7 @@ MODULE_DEVICE_TABLE(of, lm25066_of_match);
> static int lm25066_probe(struct i2c_client *client)
> {
> int config;
> + u32 shunt;
> struct lm25066_data *data;
> struct pmbus_driver_info *info;
> const struct __coeff *coeff;
> @@ -535,6 +536,17 @@ static int lm25066_probe(struct i2c_client *client)
> info->b[PSC_POWER] = coeff[PSC_POWER].b;
> }
>
> + /*
> + * Values in the TI datasheets are normalized for a 1mOhm sense
> + * resistor; assume that unless DT specifies a value explicitly.
> + */
> + if (of_property_read_u32(client->dev.of_node,
> + "shunt-resistor-micro-ohms", &shunt))
Space vs. tabs again. Never mind, fixed up and applied.
Thanks,
Guenter
> + shunt = 1000;
> +
> + info->m[PSC_CURRENT_IN] = info->m[PSC_CURRENT_IN] * shunt / 1000;
> + info->m[PSC_POWER] = info->m[PSC_POWER] * shunt / 1000;
> +
> return pmbus_do_probe(client, info);
> }
>
