On Thu, Mar 17, 2016 at 06:23:40PM +0100, Arnd Bergmann wrote:
> The regmap binding talks about one thing, which is register
> endianess, and it gets almost every aspect of it wrong.
>
> This replaces the current text of the file with a version
> that makes more sense and that matches what we implement
> now.
>
> Signed-off-by: Arnd Bergmann <a...@arndb.de>
> Fixes: a06c488da0b0 ("regmap: Add explict native endian flag to DT bindings")
> Fixes: 275876e208e2 ("regmap: Add the DT binding documentation for
> endianness")
[...]
> +Note:
> +Regmap defaults to little-endian register access on MMIO based
> +devices, this is by far the most common setting. On CPU
This seems a bit Linux specific to me, but I guess if we are defining
the binding to be this way it is fine.
Acked-by: Rob Herring <r...@kernel.org>
> +architectures that typically run big-endian operating systems
> +(e.g. PowerPC), registers can be defined as big-endian and must
> +be marked that way in the devicetree.
