Re: [RFC PATCH v1 1/2] dt-bindings: usb: samsung,exynos-xhci: support Samsung Exynos xHCI Controller

[Krzysztof Kozlowski]

On 05/12/2022 08:48, Jung Daehwan wrote:
> On Mon, Dec 05, 2022 at 08:31:46AM +0100, Krzysztof Kozlowski wrote:
>> On 05/12/2022 03:06, Jung Daehwan wrote:
>>> On Thu, Dec 01, 2022 at 09:59:06AM +0100, Krzysztof Kozlowski wrote:
>>>> On 01/12/2022 03:13, Daehwan Jung wrote:
>>>>> Add the Samsung Exynos xHCI Controller bindings with DT schema format.
>>>>>
>>>>> Signed-off-by: Daehwan Jung <dh10.jung@xxxxxxxxxxx>
>>>>> ---
>>>>>  .../bindings/usb/samsung,exynos-xhci.yaml     | 25 +++++++++++++++++++
>>>>>  1 file changed, 25 insertions(+)
>>>>>  create mode 100644 Documentation/devicetree/bindings/usb/samsung,exynos-xhci.yaml
>>>>>
>>>>> diff --git a/Documentation/devicetree/bindings/usb/samsung,exynos-xhci.yaml b/Documentation/devicetree/bindings/usb/samsung,exynos-xhci.yaml
>>>>> new file mode 100644
>>>>> index 000000000000..c5dde53b6491
>>>>> --- /dev/null
>>>>> +++ b/Documentation/devicetree/bindings/usb/samsung,exynos-xhci.yaml
>>>>> @@ -0,0 +1,25 @@
>>>>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
>>>>> +%YAML 1.2
>>>>> +---
>>>>> +$id: https://protect2.fireeye.com/v1/url?k=7899b46f-19e45c17-78983f20-74fe485fffb1-728a1b33a5d009dd&q=1&e=bdc50247-e986-43da-a15e-03ac6c3a25e8&u=http%3A%2F%2Fdevicetree.org%2Fschemas%2Fusb%2Fsamsung%2Cexynos-xhci.yaml%23
>>>>> +$schema: https://protect2.fireeye.com/v1/url?k=ea1282f0-8b6f6a88-ea1309bf-74fe485fffb1-536f21757c62f28b&q=1&e=bdc50247-e986-43da-a15e-03ac6c3a25e8&u=http%3A%2F%2Fdevicetree.org%2Fmeta-schemas%2Fcore.yaml%23
>>>>> +
>>>>> +title: Samsung Exynos xHCI
>>>>> +
>>>>> +maintainers:
>>>>> +  - Daehwan Jung <dh10.jung@xxxxxxxxxxx>
>>>>> +
>>>>> +properties:
>>>>> +  compatible:
>>>>> +    const: samsung,exynos-xhci
>>>>> +
>>>>> +required:
>>>>> +  - compatible
>>>>> +
>>>>> +additionalProperties: false
>>>>> +
>>>>
>>>> These do not look like complete bindings... What type of device has no
>>>> resources at all, just compatible?
>>>>
>>>> Best regards,
>>>> Krzysztof
>>>>
>>>>
>>>
>>> It gets resources from dwc->xhci_resources as you can see
>>> dwc3_host_init(usb/dwc3/host.c). I think it doesn't need to get another resource.
>>
>> You refer to driver, but we talk about hardware. Not driver. Your
>> hardware has no resources, so this does not look like complete binding.
>>
>> Best regards,
>> Krzysztof
>>
>>
> 
> It actually doesn't get new resources but shares resources of dwc3 driver.
> You mean it's not complete binding without resources? Is it okay if I
> add description about it? If not, could you suggest a good way?

No, description is not okay, because I doubt that your device does not
have IO space, clocks, interrupts and phys.

Therefore description changes nothing - binding still does not describe
the hardware. How to proceed? Write binding for a real hardware, not
wakelock-Android-driver-override.

Best regards,
Krzysztof