Re: [PATCH 06/17] Document the linux,network-index property.

[Segher Boessenkool <segher@kernel.crashing.org>]

>>> Segher, what is the the 'other alias' mechanism you are referring
>>> to that should be dropped?  Is it this proposed linux,network-index
>>> property?  or something else?
>>
>> Just the
>>
>> 	pic0: pic@700 {
>> 		...
>> 	}
>>
>> labeling thing -- it becomes redundant when the flat tree
>> stuff would support OF-style aliases, so it can be phased
>> out then.
>
> How?  You still need integer-encoded phandles, as that's how various
> properties are defined.

Of course.  But DTC could derive them automatically.

> Why should every node that needs a phandle
> reference have to go in /aliases?

They don't.  They don't need a label now, either.  It's just
handy for many/most things to go in /aliases.

> They seem to be solving two different
> problems to me.

Yeah, one a superset of the other.  Except for labels
on / into properties as David mentioned, which is a
flat-tree specific solution for a flat-tree specific
problem.

> As for using aliases for network indices, it could be done, but it'd be
> more complicated to implement due to string manipulation.

It's more complex because it's more flexible, yes.  I think
it'll pay off -- and how complex is using a sprintf() really.


Segher