[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220518161348.GC3302100-robh@kernel.org>
Date: Wed, 18 May 2022 11:13:48 -0500
From: Rob Herring <robh@...nel.org>
To: Stephen Boyd <swboyd@...omium.org>
Cc: linux-kernel@...r.kernel.org, patches@...ts.linux.dev,
devicetree@...r.kernel.org,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
chrome-platform@...ts.linux.dev,
Guenter Roeck <groeck@...omium.org>,
Douglas Anderson <dianders@...omium.org>,
Craig Hesling <hesling@...omium.org>,
Tom Hughes <tomhughes@...omium.org>,
Alexandru M Stan <amstan@...omium.org>,
Tzung-Bi Shih <tzungbi@...nel.org>,
Matthias Kaehlcke <mka@...omium.org>,
Benson Leung <bleung@...omium.org>,
Lee Jones <lee.jones@...aro.org>
Subject: Re: [PATCH v5 3/3] dt-bindings: cros-ec: Add ChromeOS fingerprint
binding
On Mon, May 16, 2022 at 12:23:03PM -0700, Stephen Boyd wrote:
> Quoting Rob Herring (2022-05-16 08:20:03)
> > On Wed, May 11, 2022 at 06:39:21PM -0700, Stephen Boyd wrote:
> > > diff --git a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml
> > > index 409ecef967ce..e5fe60beb9fe 100644
> > > --- a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml
> > > +++ b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml
> > > @@ -19,15 +19,19 @@ description:
> > > properties:
> > > compatible:
> > > oneOf:
> > > - - description:
> > > - For implementations of the EC is connected through I2C.
> > > - const: google,cros-ec-i2c
> > > - - description:
> > > - For implementations of the EC is connected through SPI.
> > > - const: google,cros-ec-spi
> > > - - description:
> > > - For implementations of the EC is connected through RPMSG.
> > > - const: google,cros-ec-rpmsg
> > > + # For implementations of the EC is connected through I2C.
> >
> > Using 'description' was preferred.
>
> I just learned that I should use anyOf instead of oneOf. Confusing!
Uhh, no. It can only be one of these entries at a time.
> >
> > > + - items:
> > > + - const: google,cros-ec-i2c
> > > + # For implementations of the FPMCU connected through SPI.
> > > + - items:
> > > + - const: google,cros-ec-fp
> > > + - const: google,cros-ec-spi
> > > + # For implementations of the EC is connected through SPI.
> > > + - items:
> > > + - const: google,cros-ec-spi
> > > + # For implementations of the EC is connected through RPMSG.
> > > + - items:
> > > + - const: google,cros-ec-rpmsg
> > >
> > > controller-data:
> > > description:
> > > @@ -176,6 +189,37 @@ allOf:
> > > - reg
> > > - interrupts
> > >
> > > + - if:
> > > + properties:
> > > + compatible:
> > > + contains:
> > > + const: google,cros-ec-fp
> > > + then:
> > > + properties:
> > > + '#address-cells': false
> > > + '#size-cells': false
> > > + typec: false
> > > + ec-pwm: false
> > > + keyboard-controller: false
> > > + proximity: false
> > > + codecs: false
> > > + cbas: false
> > > +
> > > + patternProperties:
> > > + "^i2c-tunnel[0-9]*$": false
> > > + "^regulator@[0-9]+$": false
> > > + "^extcon[0-9]*$": false
> >
> > Is the list of what's allowed shorter? If so, you could list those
> > properties and use 'additionalProperties: false'.
>
> Yes.
>
> >
> > Or maybe this is a sign that this should be a separate schema document.
>
> I couldn't figure that out. I tried to add new properties here but it
> didn't work.
Like this?:
then:
properties:
allowed-prop-1: true
allowed-prop-2: true
allowed-prop-3: true
additionalProperties: false
> And then when I tried to make a different schema document
> it complained that the example for google,cros-ec-spi in here had a
> problem.
You probably need a custom 'select' so that both schemas aren't
matching. Otherwise a schema is applied if any of the compatible strings
match.
> Can properties be defined in this section?
Yes, but discouraged for vendor specific properties. The if/then should
really be just additional constraints.
Rob
Powered by blists - more mailing lists