Re: [PATCH 3/3] dt-bindings: gpio: pcf857x: Convert to json-schema

[Rob Herring <robh@kernel.org>]

On Fri, May 21, 2021 at 12:23:47PM +0200, Geert Uytterhoeven wrote:
> Hi Linus,
> 
> On Fri, May 21, 2021 at 12:04 PM Linus Walleij <linus.walleij@linaro.org> wrote:
> > On Fri, May 21, 2021 at 9:54 AM Geert Uytterhoeven
> > <geert+renesas@glider.be> wrote:
> > > Convert the PCF857x-compatible I/O expanders Device Tree binding
> > > documentation to json-schema.
> > >
> > > Document missing compatible values, properties, and gpio hogs.
> > >
> > > Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> >
> > (...)
> > > Perhaps the "ti,pcf8575" construct should be removed, and the few users
> > > fixed instead?
> >
> > You would rather list it as deprecated I think?
> > It is ABI...
> 
> All DTS files use the "nxp,pcf8575" fallback, except for
> arch/x86/platform/ce4100/falconfalls.dts.
> The latter ain't working with Linux, as the Linux driver doesn't
> match against "ti,pcf8575"...

Perhaps can it just be removed?

> 
> > > +  gpio-controller: true
> >
> > So this is implicitly using the generic schema in
> > /dtschema/schemas/gpio/gpio.yaml
> 
> if you leave it out:
> 
>     Documentation/devicetree/bindings/gpio/nxp,pcf8575.yaml: ignoring,
> error in schema: properties
>     warning: no schema found in file:
> Documentation/devicetree/bindings/gpio/nxp,pcf8575.yaml
>     Documentation/devicetree/bindings/gpio/nxp,pcf8575.yaml:
> properties: 'gpio-controller' is a dependency of '#gpio-cells'
>     from schema $id: http://devicetree.org/meta-schemas/gpios.yaml#
> 
> > > +  lines-initial-states:
> > > +    $ref: /schemas/types.yaml#/definitions/uint32
> > > +    description:
> > > +      Bitmask that specifies the initial state of each line.
> > > +      When a bit is set to zero, the corresponding line will be initialized to
> > > +      the input (pulled-up) state.
> > > +      When the  bit is set to one, the line will be initialized to the
> > > +      low-level output state.
> > > +      If the property is not specified all lines will be initialized to the
> > > +      input state.
> >
> > Is this something we standardized or something that should
> > actually be a custom "nxp," property we just missed it?
> > (Looks like the latter... oh well, now it is there.)
> 
> Too late for an "nxp," prefix.
> See the NOTE in drivers/gpio/gpio-pcf857x.c:
> 
>         /* NOTE:  these chips have strange "quasi-bidirectional" I/O pins.
>          * We can't actually know whether a pin is configured (a) as output
>          * and driving the signal low, or (b) as input and reporting a low
>          * value ... without knowing the last value written since the chip
>          * came out of reset (if any).  We can't read the latched output.
>          *
>          * In short, the only reliable solution for setting up pin direction
>          * is to do it explicitly.  The setup() method can do that, but it
>          * may cause transient glitching since it can't know the last value
>          * written (some pins may need to be driven low).
>          *
>          * Using n_latch avoids that trouble.  When left initialized to zero,
>          * our software copy of the "latch" then matches the chip's all-ones
>          * reset state.  Otherwise it flags pins to be driven low.
>          */
> 
> > > +patternProperties:
> > > +  "^(hog-[0-9]+|.+-hog(-[0-9]+)?)$":
> > > +    type: object
> >
> > But this is already in
> > /dtschema/schemas/gpio/gpio-hog.yaml
> > for nodename, isn't that where it properly belongs?
> >
> > I'm however confused here Rob will know what to do.

This one is a bit odd.

> If we leave this out, something still has to refer to it?
> I see no other binding doing that...

It's selected by 'gpio-hog' being present, but here you need to make 
sure that's the case.

And I would hope you could define the node name to be just 1 of the 2 
cases.

Rob