Re: [PATCH v4 10/16] dt-bindings: phy: tegra-xusb: Add nvidia,pmc prop

[JC Kuo]

On 11/14/20 12:20 AM, Thierry Reding wrote:
> On Mon, Oct 19, 2020 at 04:40:46PM -0500, Rob Herring wrote:
>> On Fri, Oct 16, 2020 at 09:07:20PM +0800, JC Kuo wrote:
>>> This commit describes the "nvidia,pmc" property for Tegra210 tegra-xusb
>>> PHY driver. It is a phandle and specifier referring to the Tegra210
>>> pmc@7000e400 node.
>>>
>>> Signed-off-by: JC Kuo <jckuo@xxxxxxxxxx>
>>> ---
>>> v4:
>>>    new change to document "nvidia,pmc" prop
>>>
>>>  .../devicetree/bindings/phy/nvidia,tegra124-xusb-padctl.txt      | 1 +
>>>  1 file changed, 1 insertion(+)
>>>
>>> diff --git a/Documentation/devicetree/bindings/phy/nvidia,tegra124-xusb-padctl.txt b/Documentation/devicetree/bindings/phy/nvidia,tegra124-xusb-padctl.txt
>>> index 38c5fa21f435..ea559baeb546 100644
>>> --- a/Documentation/devicetree/bindings/phy/nvidia,tegra124-xusb-padctl.txt
>>> +++ b/Documentation/devicetree/bindings/phy/nvidia,tegra124-xusb-padctl.txt
>>> @@ -54,6 +54,7 @@ For Tegra210:
>>>  - avdd-pll-uerefe-supply: PLLE reference PLL power supply. Must supply 1.05 V.
>>>  - dvdd-pex-pll-supply: PCIe/USB3 PLL power supply. Must supply 1.05 V.
>>>  - hvdd-pex-pll-e-supply: High-voltage PLLE power supply. Must supply 1.8 V.
>>> +- nvidia,pmc: phandle and specifier referring to the Tegra210 pmc@7000e400 node.
>>
>> 'Tegra210 pmc@7000e400' is kind of specific. Going to update this for 
>> every address and chip?
>>
>> If there's only one PMC, you can just find the compatible PMC node. Then 
>> you don't need a DT update.
> 
> I did advise against doing the lookup by compatible string because I
> think it has a couple of downsides. On one hand it's going to create an
> additional maintenance burden on the XUSB pad controller driver because
> the PMC is usually not compatible between SoC versions, so for every new
> SoC generation we'll have to add the PMC compatible string to the XUSB
> pad controller driver.
> 
> On the other hand, this is new functionality and we can easily gate that
> on the existence of this phandle, so it doesn't impact DT backwards
> compatibility. That also has the advantage of explicitly listing the
> dependency between the PMC and the XUSB pad controller in DT, so if we
> ever do get around to come up with an even better mechanism to resolve
> these dependencies than deferred probe, we do have the data already
> available.
> 
> Thierry
> 
Hi Thierry,
Thanks for the ideas. I will stay with the adding "nvidia,pmc" phandle property
implementation. As for the DT document change, I will modify it with below.

nvidia,pmc: phandle and specifier referring to the Tegra210 PMC node.

Thanks,
JC