On Mon, 23 Mar 2020 14:39:34 +0000
Eugene Zalkonnikov <ez@xxxxxxxxxxxxx> wrote:
> Hi Jonathan,
>
> > On 22 Mar 2020, at 17:17, Jonathan Cameron <jic23@xxxxxxxxxx> wrote:
> >> +static int hdc2010_write_raw(struct iio_dev *indio_dev,
> >> + struct iio_chan_spec const *chan,
> >> + int val, int val2, long mask)
> >> +{
> >> + struct hdc2010_data *data = iio_priv(indio_dev);
> >> + int new, ret = -EINVAL;
> >
> > Unless I'm half asleep (which is possible :) ret is always
> > written below before being used. Hence this init isn't needed.
> >
> I suspect it’s a vestige from before all stylistic rewrites in recent patches.
> Same for the shared returns. Will change that.
>
> >> +
> >> + indio_dev->dev.parent = &client->dev;
> >> + indio_dev->name = dev_name(&client->dev);
> >
> > That should be the part number. IIRC that's not what you'll get
> > from dev_name called on the i2c device. Note we let a few of
> > these in in the past and are now stuck with them...
> >
> My access to actual hw is limited now, so to reduce the iterations here: what
> is going to be a suitable device name? Would “hdc20x0” do? “hdc2010”?
Should match the part so hdc2010 or hdc2080 as appropriate. No wild cards.
>
>
> Regards,
> Eugene.
