=20

-----Original Message-----
From: Wood Scott-B07421=20
Sent: Monday, October 29, 2007 2:44 PM
To: Yoder Stuart-B08248
Cc: Matt Sealey; Dale Farnsworth; Linuxppc-dev@ozlabs.org
Subject: Re: RFC: replace device_type with new "class" property?
=20
Yoder Stuart-B08248 wrote:

quoted

Here's an example of what I'm trying to get at-- take=20
a node from a FSL device tree.  The ideas I've heard
for expressing the class are like this--
=20
#1  don't express any class at all:
=20
  ucc@2200 {
        compatible =3D "fsl,ucc_geth";
        model =3D "UCC";
        device-id =3D <3>;
        reg =3D <2200 200>;
        interrupts =3D <22>;
        interrupt-parent =3D < &qeic >;
        mac-address =3D [ 00 00 00 00 00 00 ];
        local-mac-address =3D [ 00 00 00 00 00 00 ];
        rx-clock =3D <19>;
        tx-clock =3D <1a>;
        phy-handle =3D < &phy3 >;
        pio-handle =3D < &pio3 >;

=20
Of course, this should properly be something like
=20
ethernet@2200 {
	compatible =3D "fsl,mpc8360-qe-enet", "fsl,qe-enet";
	local-mac-address =3D [ 00 00 00 00 00 00 ];
	...
};
=20
With no "mac-address", and a more useful "model" if any at all.

That's fine, but as a human reader there is nothing
that would tell me this node implemented a defined
binding that specified required properties.

Stuart