Received-SPF: pass (google.com: domain of linux-kernel+bounces-178004-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240510141836.1624009-1-adureghello@baylibre.org>
 <20240510141836.1624009-3-adureghello@baylibre.org> <CAMknhBGU8bXg7obzyjzb7a4AUbjnw_0b+mqEAYJJekAK2CB-CQ@mail.gmail.com>
 <20240513185231.GA2920495-robh@kernel.org>
In-Reply-To: <20240513185231.GA2920495-robh@kernel.org>
From: David Lechner <dlechner@baylibre.com>
Date: Mon, 13 May 2024 14:03:35 -0500
Message-ID: <CAMknhBH6CrUmoUiSmU-EGc0eLw-HbaO30gAcsLBQppe3uYHpWw@mail.gmail.com>
Subject: Re: [PATCH 3/3] dt-bindings: iio: dac: fix ad3552r gain parameter names
To: Rob Herring <robh@kernel.org>
Cc: Angelo Dureghello <adureghello@baylibre.com>, jic23@kernel.org, krzk+dt@kernel.org, 
	conor+dt@kernel.org, nuno.sa@analog.com, lars@metafoo.de, 
	Michael.Hennerich@analog.com, linux-iio@vger.kernel.org, 
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Mon, May 13, 2024 at 1:52=E2=80=AFPM Rob Herring <robh@kernel.org> wrote=
:
>
> On Fri, May 10, 2024 at 10:43:18AM -0500, David Lechner wrote:
> > On Fri, May 10, 2024 at 9:19=E2=80=AFAM Angelo Dureghello
> > <adureghello@baylibre.com> wrote:
> > >
> > > From: Angelo Dureghello <adureghello@baylibre.com>
> > >
> > > The adi,gain-scaling-p/n values are an inverted log2,
> > > so initial naiming was set correct, but the driver uses just
> > > adi,gain-scaling-p/n, so uniforming documentation, that seems
> > > a less-risk fix for future rebases, and still conformant to datasheet=
.
> > >
> > > Signed-off-by: Angelo Dureghello <adureghello@baylibre.com>
> > > ---
> > >  .../devicetree/bindings/iio/dac/adi,ad3552r.yaml | 16 ++++++++------=
--
> > >  1 file changed, 8 insertions(+), 8 deletions(-)
> > >
> > > diff --git a/Documentation/devicetree/bindings/iio/dac/adi,ad3552r.ya=
ml b/Documentation/devicetree/bindings/iio/dac/adi,ad3552r.yaml
> > > index 17442cdfbe27..9e3dbf890bfa 100644
> > > --- a/Documentation/devicetree/bindings/iio/dac/adi,ad3552r.yaml
> > > +++ b/Documentation/devicetree/bindings/iio/dac/adi,ad3552r.yaml
> > > @@ -94,13 +94,13 @@ patternProperties:
> > >              maximum: 511
> > >              minimum: -511
> > >
> > > -          adi,gain-scaling-p-inv-log2:
> > > -            description: GainP =3D 1 / ( 2 ^ adi,gain-scaling-p-inv-=
log2)
> > > +          adi,gain-scaling-p:
> > > +            description: GainP =3D 1 / ( 2 ^ adi,gain-scaling-p)
> > >              $ref: /schemas/types.yaml#/definitions/uint32
> > >              enum: [0, 1, 2, 3]
> > >
> > > -          adi,gain-scaling-n-inv-log2:
> > > -            description: GainN =3D 1 / ( 2 ^ adi,gain-scaling-n-inv-=
log2)
> > > +          adi,gain-scaling-n:
> > > +            description: GainN =3D 1 / ( 2 ^ adi,gain-scaling-n)
> > >              $ref: /schemas/types.yaml#/definitions/uint32
> > >              enum: [0, 1, 2, 3]
> > >
> > > @@ -109,8 +109,8 @@ patternProperties:
> > >
> > >          required:
> > >            - adi,gain-offset
> > > -          - adi,gain-scaling-p-inv-log2
> > > -          - adi,gain-scaling-n-inv-log2
> > > +          - adi,gain-scaling-p
> > > +          - adi,gain-scaling-n
> > >            - adi,rfb-ohms
> > >
> > >      required:
> > > @@ -214,8 +214,8 @@ examples:
> > >                  reg =3D <1>;
> > >                  custom-output-range-config {
> > >                      adi,gain-offset =3D <5>;
> > > -                    adi,gain-scaling-p-inv-log2 =3D <1>;
> > > -                    adi,gain-scaling-n-inv-log2 =3D <2>;
> > > +                    adi,gain-scaling-p =3D <1>;
> > > +                    adi,gain-scaling-n =3D <2>;
> > >                      adi,rfb-ohms =3D <1>;
> > >                  };
> > >              };
> > > --
> > > 2.45.0.rc1
> > >
> > >
> >
> > The DT bindings are generally considered immutable. So unless we can
> > prove that no one has ever put adi,gain-scaling-n-inv-log2 in a .dtb
> > file,
>
> You can't ever prove that.
>
> > we probably need to fix this in the driver rather than in the
> > bindings. (The driver can still handle adi,gain-scaling-p in the
> > driver for backwards compatibility but the official binding should be
> > what was already accepted in the .yaml file)
>
> If we can reasonable assume that the Linux driver is the only consumer,
> there are no upstream dts users (in kernel or other opensource
> projects), and/or the property is somewhat recent, then that's good
> enough IMO.
>
> Rob

Ack. I stand corrected then.