On Mon, Aug 16, 2021 at 4:07 PM Liam Beguin <liambeguin@xxxxxxxxx> wrote:
> On Mon Aug 16, 2021 at 8:48 AM EDT, Andy Shevchenko wrote:
> > On Mon, Aug 16, 2021 at 3:39 PM Liam Beguin <liambeguin@xxxxxxxxx>
> > wrote:
> > > On Mon Aug 16, 2021 at 4:04 AM EDT, Andy Shevchenko wrote:
> > > > On Mon, Aug 16, 2021 at 12:35 AM Liam Beguin <liambeguin@xxxxxxxxx>
> > > > wrote:
...
> > > > > + tmp = 4096000;
> > > > > + ret = device_property_read_u32(dev, "adi,internal-ref-microvolt", &tmp);
> > > >
> > > > > + if (ret < 0 && ret != -EINVAL) {
> > >
> > > Hi Andy,
> > >
> > > >
> > > > What does this check (second part) is supposed to mean?
> > > > The first part will make it mandatory, is it the goal?
> > > >
> > >
> > > device_property_read_u32() will return -EINVAL if the property isn't
> > > found in the devicetree.
> > >
> > > This checks for errors when the property is defined while keeping it
> > > optional.
> >
> > Don't assign and don't check the error code of the API. As simply as
> > that.
>
> I'm not against getting rid of it, but I was asked to check for these
> errors in earlier revisions of the patch.
Okay, I leave it to you, guys, to decide, just note that the usual
pattern for optional stuff
a) either check for (!ret);
b) or ignore the returned value completely.
> > > > > + dev_err(dev, "invalid value for adi,internal-ref-microvolt\n");
> > > > > + return ret;
> > > > > + }
--
With Best Regards,
Andy Shevchenko
