On Wed, Oct 24, 2007 at 11:19:33AM -0400, Jon Smirl wrote:
> On 10/24/07, Grant Likely <[EMAIL PROTECTED]> wrote:
> > On 10/24/07, Timur Tabi <[EMAIL PROTECTED]> wrote:
> > > > codec0: [EMAIL PROTECTED] {
> > > > compatible = "ti,tas5508";
> > > > reg = <0>;
> > > > i2s-handle = <&[EMAIL PROTECTED]>;
> > > > };
> > >
> > > I'd do this the other way around -- that is:
> > >
> > > [EMAIL PROTECTED] { // PSC2
> > > compatible = "fsl,mpc5200b-psc-i2s","fsl,mpc5200-psc-i2s";
> > > ...
> > > i2c-handle = <&codec0>; /* Or something like that */
> >
> > i2c-handle is a poor property name here. It should be 'codec-handle'.
> > The codec could theoretically live on just about *any* control bus;
> > not just i2c.
>
> That's one of the reasons I put the second option in the post.
>
> In the second option the i2s driver would instantiate first. Next the
> generic code would get loaded. The generic codec will know the control
> but for the device and it can go look in the i2c node for the address.
> i2c node still lists all of the devices on the i2c bus. But the codecs
> are in the i2c-handle property so they don't trigger a second loading
> of the codec.
>
> A fundamental question is, which bus should trigger the loading of the
> generic codec driver. The answer to this determines how the device
> tree should look.
No! Device tree layout should not be determined by the instantiation
model used by Linux drivers right now.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-dev
