Re: [PATCH v2 1/1] hwmon: (isl28022) Fix current reading calculation

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

On 6/10/25 05:27, Geert Uytterhoeven wrote:
> Hi Yikai,
> 
> On Mon, 19 May 2025 at 10:48, Yikai Tsai <yikai.tsai.wiwynn@gmail.com> wrote:
>> According to the ISL28022 datasheet, bit15 of the current register is
>> representing -32768. Fix the calculation to properly handle this bit,
>> ensuring correct measurements for negative values.
>>
>> Signed-off-by: Yikai Tsai <yikai.tsai.wiwynn@gmail.com>
> 
> 
>> --- a/drivers/hwmon/isl28022.c
>> +++ b/drivers/hwmon/isl28022.c
>> @@ -154,6 +154,7 @@ static int isl28022_read_current(struct device *dev, u32 attr, long *val)
>>          struct isl28022_data *data = dev_get_drvdata(dev);
>>          unsigned int regval;
>>          int err;
>> +       u16 sign_bit;
>>
>>          switch (attr) {
>>          case hwmon_curr_input:
>> @@ -161,8 +162,9 @@ static int isl28022_read_current(struct device *dev, u32 attr, long *val)
>>                                    ISL28022_REG_CURRENT, &regval);
>>                  if (err < 0)
>>                          return err;
>> -               *val = ((long)regval * 1250L * (long)data->gain) /
>> -                       (long)data->shunt;
>> +               sign_bit = (regval >> 15) & 0x01;
>> +               *val = (((long)(((u16)regval) & 0x7FFF) - (sign_bit * 32768)) *
>> +                       1250L * (long)data->gain) / (long)data->shunt;
> 
> Isn't this complex operation to convert the 16-bit register value to
> a two-complement signed number equivalent to a simple cast?
> 
>      (s16)regval
> 
> isl28022_read_in() has similar code, but as the sign bit is not always
> the MSB, it needs two additional shifts:
> 
>      ((s16)(regval << shift)) >> shift
> 

Yes, I know. This could all be simplified using sign_extend32(), but I really
wanted to be able to apply it. If someone sends me a register map, maybe I can
write unit test code and use it to simplify the driver.

Guenter