On 07/03/14 19:06, Grant Likely wrote:

On Thu, 27 Feb 2014 10:36:36 +0200, Tomi Valkeinen [off-list ref] wrote:

quoted

On 26/02/14 16:48, Philipp Zabel wrote:

quoted

quoted

I would like the document to acknowledge the difference from the
phandle+args pattern used elsewhere and a description of when it would
be appropriate to use this instead of a simpler binding.

Alright. The main point of this binding is that the devices may have
multiple distinct ports that each can be connected to other devices.

The other main point with this binding are multiple endpoints per port.
So you can have, say, a display controller, with single port, which has
two endpoints going to two separate LCD panels.

In physical level that would usually mean that the same pins from the
display controller are connected to two panels. Most likely this would
mean that only one panel can be used at a time, possibly with different
settings (say, 16 RGB pins for one panel, 24 RGB pins for the other).

What device is in control in that scenario?

The endpoints in a single port are exclusive, only one can be active at
a time. So the control for the active path would be no different than in
single panel case (for which people have different opinions).

 Tomi