adc interface design

See ADC interface API for details on the interface and how to use it.

Discussing read_t

read_t only has a singular field and that is percentage. The reason a floating point number between -1.0 and 1.0 is to erase information like voltage and bit-resolution.

Erasing Voltage

The actual voltage on an ADC pin is not usually determinable by a ADC driver. The only thing an ADC peripheral can determine is the proportion of the read voltage relative to a voltage reference. That voltage reference could be anything, and therefore, there's no reason to try to embed that information into the ADC interface.

Erasing Bit Resolution

In general, analog to digital converters have a fixed bit resolution. In what cases would knowing the ADC resolution at runtime be useful? It wouldn't be useful in terms of calculating the voltage because you can use the floating point value for that.

No, the only reason why a developer or driver would need the bit resolution to discriminate against an ADC, emitting an error that indicates that the resolution of an ADC is not high enough for the driver or application to work. But this would violate the D.4 Safe design philosophy that drivers should NOT hold code that attempts to check for problems of architecture.