On Thu, Feb 14, 2013 at 6:48 PM, Stephen Warren <[email protected]> wrote:

> This is actually why I rather dislike the idea that device tree has an
> immutable bindings. To some exten, I imagine that much of
> Documentation/stable_api_nonsense.txt could apply against device tree!
> We can't fix problems like this by changing the definition of that
> binding to remove the cd-inverted property and rely solely on the GPIO
> specifier flags.

Atleast it could use an "unstable, testing" phase I think.
But there are others who think breaking support for an old
device trees is like breaking userspace.

If we shall set DT bindings in stone they need to go through as
much scrutiny as these IEEE committee products that are
over at openfirmware.org, so we can at least mitigate the risk
of horrid mistakes.

I think it was Wolfram who mentioned some DT binding work
being rushed by engineers constantly keeping a finger
on the fast-forward button and I agree with that observation. :-(

Yours,
Linus Walleij
_______________________________________________
devicetree-discuss mailing list
[email protected]
https://lists.ozlabs.org/listinfo/devicetree-discuss

Reply via email to