Re: [PATCH] (ina2xx) Drop bus_voltage_shift configuration

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

On 3/2/26 08:02, Jonas Rebmann wrote:
> Hi Guenter,
> 
> On 2026-03-02 16:50, Guenter Roeck wrote:
>> Hi,
>>
>> On 3/2/26 07:26, Jonas Rebmann wrote:
>>> The INA219 has the lowest three bits of the bus voltage register
>>> zero-reserved and the bus_voltage_shift ina2xx_config field was
>>> introduced to accommodate for that.
>>>
>>> The INA234 has four bits of the bus voltage, of the shunt voltage, and
>>> of the current registers zero-reserved but the latter two were
>>> implemented by choosing a 16x higher conversion constant instead of a
>>> separate field specifying a bit shift.
>>>
>>> For consistency and simplicity, drop bus_voltage_shift and adapt the
>>> conversion constants for INA219 and INA234 accordingly, yielding the
>>> same measurement values.
>>>
>>
>> This isn't about simplicity, it is about correctness.
>>
>> The datasheet for INA234 says for the lower 4 bits:
>>
>>       Always returns 0. Remove these bits from the full result by doing a
>>       right arithmetic shift
>>
>> which is what we should do for all chips with reserved bits instead of
>> assuming that the reserved bits always return 0.
> 
> It says that for the reserved-zero bits at the lower end of the shunt
> voltage and current registers of INA234 as well as for the bus voltage
> register.
> 
> If bus_voltage_shift is kept, wouldn't a shunt_voltage_shift and
> current_shift need to be introduced to support those registers of the
> INA234 with the same notion of correctness?
> 

shunt_voltage_shift, yes, but the current register is not used,
so a current_shift would be unnecessary.

Thanks,
Guenter