Re: [PATCH 1/2] dt-bindings: reset: Add document for Broadcom STB reset controller

[Florian Fainelli]

On 1/3/19 11:19 AM, Rob Herring wrote:
> On Thu, Jan 03, 2019 at 10:53:25AM -0800, Florian Fainelli wrote:
>> On 1/3/19 9:41 AM, Rob Herring wrote:
>>> On Thu, Dec 20, 2018 at 05:34:08PM -0800, Florian Fainelli wrote:
>>>> Add a binding document for the Broadcom STB reset controller, also known
>>>> as SW_INIT-style reset controller.
>>>>
>>>> Signed-off-by: Florian Fainelli <f.fainelli@xxxxxxxxx>
>>>> ---
>>>>  .../devicetree/bindings/reset/brcm,reset.txt  | 27 +++++++++++++++++++
>>>>  1 file changed, 27 insertions(+)
>>>>  create mode 100644 Documentation/devicetree/bindings/reset/brcm,reset.txt
>>>>
>>>> diff --git a/Documentation/devicetree/bindings/reset/brcm,reset.txt b/Documentation/devicetree/bindings/reset/brcm,reset.txt
>>>> new file mode 100644
>>>> index 000000000000..6e5341b4f891
>>>> --- /dev/null
>>>> +++ b/Documentation/devicetree/bindings/reset/brcm,reset.txt
>>>> @@ -0,0 +1,27 @@
>>>> +Broadcom STB SW_INIT-style reset controller
>>>> +===========================================
>>>> +
>>>> +Broadcom STB SoCs have a SW_INIT-style reset controller with separate
>>>> +SET/CLEAR/STATUS registers and possibly multiple banks, each of 32 bit
>>>> +reset lines.
>>>> +
>>>> +Please also refer to reset.txt in this directory for common reset
>>>> +controller binding usage.
>>>> +
>>>> +Required properties:
>>>> +- compatible: should be brcm,brcmstb-reset
>>>> +- reg: register base and length
>>>> +- #reset-cells: must be set to 1
>>>> +
>>>> +Example:
>>>> +
>>>> +	reset: reset-controller@8404318 {
>>>> +		compatible = "brcm,brcmstb-reset";
>>>> +		reg = <0x8404318 0x30>;
>>>
>>> Based on this address, should this be a sub-node of something else? Or 
>>> not even a sub-node and just make the parent be a reset provider?
>>
>> The reset controller is part of a larger "sundry" node which has a
>> collection of functionality, from pinmux/pinctrl, reset controller,
>> spare bits, chicken bits, anything the designers forgot to put somewhere
>> else and decided to put there.
>>
>> If there is one thing consistent though is that given a set of 32-bit
>> register groups, they have a self contained functionality such that you
>> can break up the larger "sundry" space into smaller sub-blocks which
>> have one an only one functionality. Do you think this warrants a
>> different representation in Device Tree?
> 
> With pinctrl in the mix, you're going to need sub-nodes anyways. So just 
> define what this is a sub-node of.

pinctrl is not necessarily something I want the kernel to control, since
we have a high level scripting language without our bootloader that
makes sure pinctrl is properly configured from early boot on all the way
to the kernel, and preserved across suspend/resume states.
pinctrl-single does work, and was occasionally used. Everything else is
typically muxes that the kernel does not need to touch/see/be aware of.

> 
> Also, I'd prefer to have complete example for the "sundry" node and 
> child nodes than partial examples spread across the tree.

I am afraid I can't provide that example because the sundry node is
really changing from chip to chip, and there is a gazillion of things in
there that the kernel typically does not even touch, like
pinmuxing/pinctrl, various mux selections etc. I could provide the
following example if that is what you are requesting?:


sun-top-ctrl: simple-bus@8404000 {
	compatible = "brcm,brcmstb-sun-top-ctrl", "simple-bus";
	reg = <0x8404000 0x708>;

	reset: reset-controller@318 {
		compatible = "brcm,brcmstb-reset";
		reg = <0x318 0x30>;
		#reset-cells = <1>;
	};
};

Would that be what you expect to see?
-- 
Florian