On Wed, Apr 7, 2021 at 10:26 AM Lars-Peter Clausen <lars@xxxxxxxxxx> wrote:
>
> On 4/6/21 8:28 PM, Puranjay Mohan wrote:
> > +
> > +static int tmp117_write_raw(struct iio_dev *indio_dev,
> > + struct iio_chan_spec const *channel, int val,
> > + int val2, long mask)
> > +{
> > + struct tmp117_data *data = iio_priv(indio_dev);
> > + s16 off;
> > +
> > + switch (mask) {
> > + case IIO_CHAN_INFO_CALIBBIAS:
> > + off = clamp(val, -32768, 32767);
> > + if (off == data->calibbias)
>
> data->calibbias is only set in probe() and always 0. I'm not sure we
> need to cache the value. Reading it back from the device seems fine.
I forgot to update it, thanks for finding the bug.
Actually, I wanted to update the calibbias register only if a
different value is being written.
If the same value is written to the device repeatedly then we can save
some I2C writes using this.
and while reading the calibbias, it is read from the device only.
I will fix the bug in the next revision.
>
> > + return 0;
> > + return i2c_smbus_write_word_swapped(data->client,
> > + TMP117_REG_TEMP_OFFSET, off);
> > +
> > + default:
> > + return -EINVAL;
> > + }
> > +}
> > +
>
--
Thanks and Regards
Yours Truly,
Puranjay Mohan
