Hi Russell,

On Thursday 20 March 2014 18:18:20 Russell King - ARM Linux wrote:

On Thu, Mar 20, 2014 at 07:16:29PM +0100, Laurent Pinchart wrote:

quoted

On Thursday 20 March 2014 17:54:31 Grant Likely wrote:

quoted

On Wed, 12 Mar 2014 10:25:56 +0000, Russell King - ARM Linux wrote:

[snip]

quoted

quoted

quoted

I believe trying to do this according to the flow of data is just
wrong. You should always describe things from the primary device for
the CPU towards the peripheral devices and never the opposite
direction.

Agreed.

Absolutely not agreed. The whole concept of CPU towards peripherals only
makes sense for very simple devices and breaks as soon as the hardware
gets more complex. There's no such thing as CPU towards peripherals when
peripherals communicate directly.

Please consider use cases more complex than just a display controller and
an encoder, and you'll realize how messy not being able to parse the
whole graph at once will become. Let's try to improve things, not to make
sure to prevent support for future devices.

That's odd, I did.

Please draw some (ascii) diagrams of the situations you're saying this
won't work for, because at the moment all I'm seeing is some vague
hand-waving rather than anything factual that I can relate to.  Help
us to actually _see_ the problem you have with this approach so we can
understand it.

Working on it. Given my drawing skills this will take a bit of time.

-- 
Regards,

Laurent Pinchart

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html