Re: [PATCH 3/9] phy-rcar-usb-gen2: add device tree support

[Ben Dooks <ben.dooks@codethink.co.uk>]

On 06/03/14 19:16, Sergei Shtylyov wrote:
> Hello.
>
> On 03/06/2014 09:01 PM, Ben Dooks wrote:
>
>> Add support for the phy-rcar-gen2-usb driver to be probed from device
>> tree.
>
>> Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
>> Reviewed-by: Ian Molton <ian.molton@codethink.co.uk>
>> ---
>> Fixes from v2:
>>     - fix missed of_match_ptr()
>>     - fix names of channel selection booleans
>>     - updated and merged documentation for dt entries
>>
>> Fixes from v2:
>>     - fix missing of_if patch
>>
>> Fixes from v1:
>>     - use of_property_reasd-bool()
>>     - remove unused of_id variable
>>
>> Cc: Felipe Balbi <balbi@ti.com>
>> Cc: linux-usb@vger.kernel.org
>> Cc: linux-sh@vger.kernel.org
>> Cc: Magnus Damm <magnus.damm@gmail.com>
>> Cc: Simon Horman <horms@verge.net.au>
>> Cc: devicetree@vger.kernel.org
>
>> Conflicts:
>>     drivers/usb/phy/phy-rcar-gen2-usb.c
>> ---
>>   .../bindings/usb/renesas,rcar-gen2-usb-phy.txt     | 36
>> ++++++++++++++++++++++
>>   drivers/pci/host/pci-rcar-gen2.c                   |  1 +
>
>     Eh? What does this file have to do with USB PHY?

Ah, it was a fixup for a missing header that got merged
into the wrong file. Will fix that.

>>   drivers/usb/phy/phy-rcar-gen2-usb.c                | 34
>> +++++++++++++++++---
>>   3 files changed, 66 insertions(+), 5 deletions(-)
>>   create mode 100644
>> Documentation/devicetree/bindings/usb/renesas,rcar-gen2-usb-phy.txt
>
>> diff --git
>> a/Documentation/devicetree/bindings/usb/renesas,rcar-gen2-usb-phy.txt
>> b/Documentation/devicetree/bindings/usb/renesas,rcar-gen2-usb-phy.txt
>> new file mode 100644
>> index 0000000..5351a30
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/usb/renesas,rcar-gen2-usb-phy.txt
>> @@ -0,0 +1,36 @@
>> +Renesas RCar gen2 USB PHY bindings
>> +----------------------------------
>> +
>> +Bindings for the USB PHY block used in some Renesas SoCs.
>> +
>> +Required properties:
>> + - compatible:  "renesas,usb-phy-r8a7790" for the R8A7790 SoC
>> +        "renesas,usb-phy-r8a7791" for the R8A7791 SoC
>> + - reg : A single region to access device registers
>> + - clocks : The reference to the clock to use for this block
>> + - clock-names : The name for the clock at index 0 (must be "usbhs")
>> +
>> +Optional properties:
>> +
>> + - renesas,usb0-device: boolean, if present USB0 is connected to HS
>> device
>> +            otherwise the USB0 is connected to OHCI/EHCI host.
>
>     IIUC, the testing has shown that USBHS is dual-role controller in
> that case, i.e. supports both host and device roles (the manual has the
> host controller details too). Vladimir, is it so?

Currently there is no auto-detection for this, so it gets set at
start time.

>> + - renesas,usb2-xhci: boolean, if present USB2 is connected to XHCI
>> controller
>> +              otherwise the USB2 is connected to OHCI/EHCI host.
>> +
>> +
>> +Example device node for SoC dtsi file:
>> +
>> +    usbphy: usbphy@e6590100 {
>> +        compatible = "renesas,usb-phy-r8a7790";
>> +        clocks = <&mstp7_clks R8A7790_CLK_HSUSB>;
>> +        clock-names = "usbhs";
>> +        reg = < 0x0 0xe6590100 0x0 0x100>;
>> +        status = "disabled";
>> +    };
>> +
>> +Example board file:
>> +
>> +&usbphy {
>> +    status = "okay";
>> +};
>
>     These are usually merged into one node for the example.

Much nicer if separate.

-- 
Ben Dooks				http://www.codethink.co.uk/
Senior Engineer				Codethink - Providing Genius