[PATCH v4 1/3] i2c: cadence: Document device tree bindings
Gerhard Sittig
gsi at denx.de
Fri Apr 4 12:17:57 PDT 2014
On Thu, 2014-04-03 at 10:59 -0700, Soren Brinkmann wrote:
>
> Add device tree binding documentation for the Cadence I2C controller.
>
> [ ... ]
>
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/i2c/i2c-cadence.txt
> @@ -0,0 +1,21 @@
> +Binding for the Cadence I2C controller
> +
> +Required properties:
> + compatible: Compatibility string. Must be 'cdns,i2c-r1p10'.
> + clocks: From common clock bindings. Phandle to input clock.
the usual complaint: 'clocks' items are not just phandles (your
example even suggests this); either adjust the description to
something correct, or just refer to the common clock bindings to
not duplicate their description
but I guess the I2C controller's binding should explicitly state
which clocks are required, and you might want to consider
'clock-names', too
> +
> +Optional properties:
> + clock-frequency: Desired operating frequency, in Hz, of the bus (actual may
> + be lower). Defaults to 400000 if not specified.
is the value default a feature of the Linux implementation, or
consciously designed to be in the binding spec? and I agree that
the default should be the standard I2C speed instead of fast mode
> +
> +Example:
> +
> + i2c at e0004000 {
> + compatible = "cdns,i2c-r1p10";
> + clocks = <&clkc 38>;
> + interrupts = <0 25 4>;
> + reg = <0xE0004000 0x1000>;
> + clock-frequency = <400000>;
> + #address-cells = <1>;
> + #size-cells = <0>;
> + };
use lower case hex digits, consider symbolic identifiers for
clocks and interrupt flags
the example has many more properties than the binding describes,
the additional items aren't even optional -- you might want to
refer to a few more common or general bindings
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: office at denx.de
More information about the linux-arm-kernel
mailing list