Re: [PATCH v2 3/4] arm: dts: Add USB phy nodes for AM33XX

[George Cherian <george.cherian@ti.com>]

On 7/20/2013 12:03 AM, Sergei Shtylyov wrote:
> Hello.
>
> On 07/19/2013 06:20 PM, Sebastian Andrzej Siewior wrote:
>
>>>> diff --git a/arch/arm/boot/dts/am33xx.dtsi
>>>> b/arch/arm/boot/dts/am33xx.dtsi
>>>> index 8e1248f..e3890c4 100644
>>>> --- a/arch/arm/boot/dts/am33xx.dtsi
>>>> +++ b/arch/arm/boot/dts/am33xx.dtsi
>>>> @@ -326,21 +326,59 @@
>>>>                status = "disabled";
>>>>            };
>>>>
>>>> -        usb@47400000 {
>>>> -            compatible = "ti,musb-am33xx";
>>>> -            reg = <0x47400000 0x1000    /* usbss */
>>>> -                   0x47401000 0x800        /* musb instance 0 */
>>>> -                   0x47401800 0x800>;    /* musb instance 1 */
>>>> -            interrupts = <17        /* usbss */
>>>> -                      18        /* musb instance 0 */
>>>> -                      19>;        /* musb instance 1 */
>>>> -            multipoint = <1>;
>>>> -            num-eps = <16>;
>>>> -            ram-bits = <12>;
>>>> -            port0-mode = <3>;
>>>> -            port1-mode = <3>;
>>>> -            power = <250>;
>>>> -            ti,hwmods = "usb_otg_hs";
>>>> +        phy1: am335x-usb0@44e10620 {
>
>>>     Shouldn't the PHYs be *under* the usb0/1 device nodes in the 
>>> hierarchy?
>>> They're not on the same bus as the MUSB controllers for sure.
>
>> I redo the complete thing. I have now:
>
>> usb: usb@47400000 {
>>     compatible = "ti,am33xx-usb";
>>
>>     usb0_phy: phy@47401300 {
>>         compatible = "ti,am335x-usb-phy";
>>     }
>>     usb0: usb@47401000 {
>>         musb0: usb@47401400 {
>>             compatible = "mg,musbmhdrc";
>>         }
>>     }
>>     usb1_phy: phy@47402300 {
>>         compatible = "ti,am335x-usb-phy";
>>     }
>>     usb1: usb@47402000 {
>>         musb1: usb@47402400 {
>>             compatible = "mg,musbmhdrc";
>>         }
>>     }
>> }
>
>> And you want usb0_phy to be child of usb0? In the TRM they are all in
>> the same block.
>
>    Ah, the fact that PHYs didn't have the "reg" property got me 
> muddled, I didn't pay attention to the address part of the node 
> names... BTW, where is the "reg" prop? I see PHYs share the address 
> space with "omap-control-usb@44e10620" device -- what's the point with 
> this?
In control module(CM) each USB has got 2 registers  in which one is a 
shared register( wakeup register)
So all the CM access are done using the control-usb driver 
(phy-omap-control-usb.c). Same reason why phy's dont have a reg property.
>
>> Sebastian
>
> WBR, Sergei
>


-- 
-George