Re: [RFC PATCH 6/7] ARM: dts: stm32: add ETZPC as a system bus for STM32MP15x boards

[Gatien CHEVALLIER]

On 12/22/22 16:21, Krzysztof Kozlowski wrote:
> On 22/12/2022 15:42, Gatien CHEVALLIER wrote:
> > Hello,
> >
> >
> > On 12/22/22 11:30, Krzysztof Kozlowski wrote:
> > > On 21/12/2022 18:30, Gatien Chevallier wrote:
> > > > The STM32 System Bus is an internal bus on which devices are connected.
> > > > ETZPC is a peripheral overseeing the firewall bus that configures
> > > > and control access to the peripherals connected on it.
> > > >
> > > > For more information on which peripheral is securable, please read
> > > > the STM32MP15 reference manual.
> > > >
> > > > Signed-off-by: Gatien Chevallier <gatien.chevallier@xxxxxxxxxxx>
> > > > ---
> > > >    arch/arm/boot/dts/stm32mp151.dtsi | 2737 +++++++++++++++--------------
> > > >    1 file changed, 1406 insertions(+), 1331 deletions(-)
> > > >    
> > > > -		lptimer1: timer@40009000 {
> > > > +		etzpc: etzpc@5c007000 {
> > >
> > > Node names should be generic.
> > > https://devicetree-specification.readthedocs.io/en/latest/chapter2-devicetree-basics.html#generic-names-recommendation
> >
> > Will change to etzpc: bus@5c007000 in V3
> >
> > > > +			compatible = "st,stm32mp15-sys-bus";
> > > > +			reg = <0x5c007000 0x400>;
> > > >    			#address-cells = <1>;
> > > > -			#size-cells = <0>;
> > > > -			compatible = "st,stm32-lptimer";
> > > > -			reg = <0x40009000 0x400>;
> > > > -			interrupts-extended = <&exti 47 IRQ_TYPE_LEVEL_HIGH>;
> > > > -			clocks = <&rcc LPTIM1_K>;
> > > > -			clock-names = "mux";
> > > > -			wakeup-source;
> > > > -			status = "disabled";
> > >
> > > Why entire bus is disabled? What resources do you miss?
> >
> > Here, I did not want to modify the status of the nodes in the device
>
> By making it disabled you actually modify the status of any enabled
> node. By making it enabled you do not cause any change.
>
> > tree. Nodes that are default enabled in the SoC device tree stay enabled
> > and vice-versa.
>
> No, because parent is disabled.
>
> > IMO think this patch should not include these kind of
> > changes in its scope. I describe the bus and the peripherals connected
> > to it, nothing more :)
>
> I am not proposing to change existing nodes. I am asking why new node is
> being disabled (because it is a new node, isn't it?).

I'm sorry but the ETZPC node is not disabled. The patch is quite hard to 
read as I'm moving all securable/MCU isolation capable peripheral nodes 
under ETZPC.

>>>> -			#size-cells = <0>;
>>>> -			compatible = "st,stm32-lptimer";
>>>> -			reg = <0x40009000 0x400>;
>>>> -			interrupts-extended = <&exti 47 IRQ_TYPE_LEVEL_HIGH>;
>>>> -			clocks = <&rcc LPTIM1_K>;
>>>> -			clock-names = "mux";
>>>> -			wakeup-source;
>>>> -			status = "disabled";

are part of lptimer1 node, which is moved under ETZPC. ETZPC node is 
default enabled.

Result after patch is:

etzpc: etzpc@5c007000 {
	compatible = "st,stm32mp15-sys-bus";
	reg = <0x5c007000 0x400>;
	#address-cells = <1>;
	#size-cells = <1>;
	feature-domain-controller;
	#feature-domain-cells = <2>;
	ranges;

	timers2: timer@40000000 {
(...)

Or do I misunderstand you comment?

> Best regards,
> Krzysztof