On Tue, Apr 30, 2024 at 9:59 AM Jonathan Cameron <Jonathan.Cameron@xxxxxxxxxx> wrote: > > On Tue, 30 Apr 2024 09:13:30 -0400 > marc.ferland@xxxxxxxxx wrote: > > > From: Marc Ferland <marc.ferland@xxxxxxxxxxxx> > > > > For temperature readings, the remainder is returned as nano Celsius > > _but_ we mark it as IIO_VAL_INT_PLUS_MICRO. This results in incorrect > > temperature reporting through hwmon for example. I have a board here > > which reports the following when running 'sensors': > > > > iio_hwmon-isa-0000 > > Adapter: ISA adapter > > temp1: +93.3°C > > > > With the patch applied, it returns the correct temperature: > > > > iio_hwmon-isa-0000 > > Adapter: ISA adapter > > temp1: +30.5°C > > > > Signed-off-by: Marc Ferland <marc.ferland@xxxxxxxxxxxx> > > IIO temperature units are milli celcius, so I'm not following > the argument here. The driver might be reporting in pico celcius > I suppose? Call out that this is the scale factor though, so > it corresponds to 1LSB hence a small number is certainly plausible.. > > Reasonable to argue it's taking the integer and dividing by 10^9 hence > INT_PLUS_NANO is the right answer, but it isn't nano celsius. > Oups! you're obviously right. I'll update my commit message and resend. Marc
