David Gibson wrote:
> On Tue, Nov 06, 2007 at 09:06:34PM +0100, Marian Balakowicz wrote:
>> Add device tree source file for Motion-PRO board.
> [snip]
>> + [EMAIL PROTECTED] { // Motion-PRO status LED
>> + compatible = "promess,motionpro-statusled";
>> + reg = <660 10>;
>> + interrupts = <1 f 0>;
>> + interrupt-parent = <&mpc5200_pic>;
>> + blink-delay = <64>; // 100 msec
>> + };
>> +
>> + [EMAIL PROTECTED] { // Motion-PRO ready LED
>> + compatible = "promess,motionpro-readyled";
>> + reg = <670 10>;
>> + interrupts = <1 10 0>;
>> + interrupt-parent = <&mpc5200_pic>;
>> + };
>
> Is there actually any difference in behaviour betweeen the two LEDs?
> If not, they should probably have the same compatible value, and
> perhaps a "label" property or some such to associate them with a
> particular LED.
In fact they are identical, so I have used the same compatible
property and added label property do distinguish them.
> [snip]
>> + [EMAIL PROTECTED] {
>> + device_type = "spi";
>
> No device_type!
Removed.
>> + compatible = "mpc5200b-spi","mpc5200-spi";
>> + reg = <f00 20>;
>> + interrupts = <2 d 0 2 e 0>;
>> + interrupt-parent = <&mpc5200_pic>;
>> + };
>> +
>> + [EMAIL PROTECTED] {
>> + device_type = "usb-ohci-be";
>
> Nor here.
Removed.
> [snip]
>> + // PSC2 in spi master mode
>> + [EMAIL PROTECTED] { // PSC2
>> + compatible = "mpc5200b-psc-spi","mpc5200-psc-spi";
>> + cell-index = <1>;
>
>>From your description, this is an incorrect usage of cell-index - it
> should *only* be used to index into SoC shared registers; never for
> logical numbering.
Again true, but the situation is similar to the one with the "serial"
node.
Cheers,
m.
_______________________________________________
Linuxppc-dev mailing list
[email protected]
https://ozlabs.org/mailman/listinfo/linuxppc-dev
