[EXT] Re: [PATCH v2 1/3] dt-bindings: usb: cdns-imx8qm: add imx8qm cdns3 glue bindings
Krzysztof Kozlowski
krzysztof.kozlowski at linaro.org
Mon Mar 20 09:27:22 PDT 2023
On 20/03/2023 17:22, Frank Li wrote:
>>>>>>> + assigned-clocks:
>>>>>>> + items:
>>>>>>> + - description: Phandle and clock specifier of
>> IMX_SC_PM_CLK_PER.
>>>>>>> + - description: Phandle and clock specifoer of
>>>> IMX_SC_PM_CLK_MISC.
>>>>>>> + - description: Phandle and clock specifoer of
>>>>>> IMX_SC_PM_CLK_MST_BUS.
>>>>>>> +
>>>>>>> + assigned-clock-rates:
>>>>>>> + items:
>>>>>>> + - description: Must be 125 Mhz.
>>>>>>> + - description: Must be 12 Mhz.
>>>>>>> + - description: Must be 250 Mhz.
>>>>>>
>>>>>> I would argue that both properties above are not needed. If your
>>>>>> hardware requires fixed frequencies, clock provider can fix them, can't
>> it?
>>>>>
>>>>> Clock provider don't know fixed value and turn on only used by client.
>>>>
>>>> So maybe fix the clock provider? Or this device driver? Requiring by
>>>> binding specific frequencies for every board is a bit redundant.
>>>
>>> It is not for every boards, it is common for a chip family. Only a place to set
>> for
>>> QM and QXP.
>>>
>>> The similar case is network driver, which require a specific frequency at
>> clock assign.
>>> Generally frequency is fixed, clock source name may change at difference
>> chips.
>>
>> If frequency is always fixed, I don't understand why this is in DT
>> bindings. I would even say it should not be in DTS. We don't put into
>> DTS properties which are always the same, because otherwise they would
>> grow crazy big.
>
> Although frequency is fixed, clock name may change for difference platform.
>
> assigned-clocks = <&clk IMX_SC_R_USB_2 IMX_SC_PM_CLK_PER>,
> <&clk IMX_SC_R_USB_2 IMX_SC_PM_CLK_MISC>,
> <&clk IMX_SC_R_USB_2 IMX_SC_PM_CLK_MST_BUS>;
> assigned-clock-rates = <125000000>, <12000000>, <250000000>;
>
> some platform use IMX_SC_R_USB_2, other platform may use IMX_SC_R_USB_3.
This I understand, you wrote it above, so nothing new and my concerns
are still there.
Best regards,
Krzysztof
More information about the linux-arm-kernel
mailing list