RE: [PATCH v5 3/3] iio: dac: ad3530r: Add driver for AD3530R and AD3531R

[Paller, Kim Seer]

> -----Original Message-----
> From: Andy Shevchenko <andy@xxxxxxxxxx>
> Sent: Thursday, April 24, 2025 12:50 AM
> To: Paller, Kim Seer <KimSeer.Paller@xxxxxxxxxx>
> Cc: Jonathan Cameron <jic23@xxxxxxxxxx>; Lars-Peter Clausen
> <lars@xxxxxxxxxx>; Hennerich, Michael <Michael.Hennerich@xxxxxxxxxx>;
> Rob Herring <robh@xxxxxxxxxx>; Krzysztof Kozlowski <krzk+dt@xxxxxxxxxx>;
> Conor Dooley <conor+dt@xxxxxxxxxx>; David Lechner
> <dlechner@xxxxxxxxxxxx>; Nuno Sá <noname.nuno@xxxxxxxxx>; Sa, Nuno
> <Nuno.Sa@xxxxxxxxxx>; linux-iio@xxxxxxxxxxxxxxx; linux-
> kernel@xxxxxxxxxxxxxxx; devicetree@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v5 3/3] iio: dac: ad3530r: Add driver for AD3530R and
> AD3531R
> 
> [External]
> 
> On Wed, Apr 23, 2025 at 07:53:37AM +0000, Paller, Kim Seer wrote:
> > > From: Andy Shevchenko <andy@xxxxxxxxxx>
> > > Sent: Tuesday, April 22, 2025 11:11 PM
> > > On Mon, Apr 21, 2025 at 12:24:54PM +0800, Kim Seer Paller wrote:
> 
> First of all, there are a lot of comments left without replies while some of
> them commented as "agree I will follow your advice". This is confusing. The
> rule of thumb is to not reply with positive at all, just only for the things
> that you want to clarify. And with that remove a lot of unneeded (you agree
> with) context!
> 
> ...
> 
> > > > +		st->vref_mv = range_multiplier * vref / 1000;
> > >
> > > MILLI?
> >
> > Yes this is milli, will change this also to vref_mV
> >
> > 	st->vref_mV = range_multiplier * vref_mV / 1000;
> 
> Ah, I was not clear enough, MILLI in capital letters is defined constant which
> you may use instead of 1000.
> 
> ...
> 
> > > > +	vref = devm_regulator_get_enable_read_voltage(dev, "ref");
> > > > +	if (vref < 0 && vref != -ENODEV)
> > > > +		return vref;
> > > > +
> > > > +	has_external_vref = vref != -ENODEV;
> > >
> > > Wouldn't be better just make this 0 when it's == -ENODEV and check just
> the
> > > value without having this additional boolean variable (note, I haven't
> checked
> > > the meaning of Vref == 0 in case it's possible in real life and hardware
> behaves
> > > adequately)?
> >
> > I think it could be simpler to set vref to 0 when it's -ENODEV and check its
> value directly
> > without having additional boolean variable. I'll try this approach.
> 
> But double check that hardware doesn't support Vref == 0 in real life.

I confirm that the hardware doesn't support Vref == 0, as also stated in the datasheet.
 
> 
> --
> With Best Regards,
> Andy Shevchenko
>