Re: [PATCH v3 5/9] dt-bindings: timer: Add schema for realtek,otto-timer

From: Chris Packham
Date: Sun Jun 30 2024 - 22:15:19 EST



On 30/06/24 08:40, Sander Vanheule wrote:
> Hi Chris,
>
> Thanks for submitting these patches!
>
> On Thu, 2024-06-27 at 16:33 +1200, Chris Packham wrote:
>> Add the devicetree schema for the realtek,otto-timer present on a number
>> of Realtek SoCs.
>>
>> Signed-off-by: Chris Packham <chris.packham@xxxxxxxxxxxxxxxxxxx>
>> ---
> [...]
>
>> +
>> +  reg:
>> +    items:
>> +      - description: timer0 registers
>> +      - description: timer1 registers
>> +      - description: timer2 registers
>> +      - description: timer3 registers
>> +      - description: timer4 registers
>> +
>> +  clocks:
>> +    maxItems: 1
>> +
>> +  interrupts:
>> +    items:
>> +      - description: timer0 interrupt
>> +      - description: timer1 interrupt
>> +      - description: timer2 interrupt
>> +      - description: timer3 interrupt
>> +      - description: timer4 interrupt
> Instead of providing a (SoC dependent) number of reg and interrupt items, can't we just
> provide one reg+interrupt per timer and instantiate one block for however many timers the
> SoC has?
>
>> +
>> +additionalProperties: false
>> +
>> +examples:
>> +  - |
>> +    timer@3200 {
>> +      compatible = "realtek,rtl9302-timer", "realtek,otto-timer";
>> +      reg = <0x3200 0x10>, <0x3210 0x10>, <0x3220 0x10>,
>> +            <0x3230 0x10>, <0x3240 0x10>;
>> +
>> +      interrupt-parent = <&intc>;
>> +      interrupts = <7>, <8>, <9>, <10>, <11>;
>> +      clocks = <&lx_clk>;
>> +    };
> So this would become:
> timer@3200 {
> compatible = ...
> reg = <0x3200 0x10>;
> interrupt-parent = <&intc>;
> interrupts = <7>;
> ...
> };
> timer@3210 {
> compatible = ...
> reg = <0x3210 0x10>;
> interrupt-parent = <&intc>;
> interrupts = <8>;
> ...
> };
> ...
>
> More verbose, but it also makes the binding a bit simpler. The driver can then still do
> whatever it wants with all the timers that are registered, although some more resource
> locking might be required.

I kind of prefer the single entry for the whole TCU. If we were to fold
the watchdog into this then we could have a single larger range that
covered all the timers similar to the ingenic,tcu. But that would
technically be a breaking change at this point.