Re: [PATCH v1 2/2] riscv: dts: starfive: add the missing monitor core

[<Conor.Dooley@microchip.com>]

On 13/07/2022 16:12, Icenowy Zheng wrote:
> 在 2022-07-13星期三的 15:09 +0000，Conor.Dooley@microchip.com写道：
>>
>>
>> On 13/07/2022 16:02, Icenowy Zheng wrote:
>>> 在 2022-07-13星期三的 14:55 +0000，Conor.Dooley@microchip.com写道：
>>>> On 13/07/2022 15:26, Icenowy Zheng wrote:
>>>>>
>>>>> 在 2022-07-11星期一的 19:43 +0100，Conor Dooley写道：
>>>>>> From: Conor Dooley <conor.dooley@microchip.com>
>>>>>>
>>>>>> The JH7100 has a 32 bit monitor core that is missing from the
>>>>>> device
>>>>>> tree. Add it (and its cpu-map entry) to more accurately
>>>>>> reflect
>>>>>> the
>>>>>> actual topology of the SoC.
>>>>>>
>>>>>> Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
>>>>>> ---
>>>>>>  arch/riscv/boot/dts/starfive/jh7100.dtsi | 21
>>>>>> +++++++++++++++++++++
>>>>>>  1 file changed, 21 insertions(+)
>>>>>>
>>>>>> diff --git a/arch/riscv/boot/dts/starfive/jh7100.dtsi
>>>>>> b/arch/riscv/boot/dts/starfive/jh7100.dtsi
>>>>>> index c617a61e26e2..92fce5b66d3d 100644
>>>>>> --- a/arch/riscv/boot/dts/starfive/jh7100.dtsi
>>>>>> +++ b/arch/riscv/boot/dts/starfive/jh7100.dtsi
>>>>>> @@ -67,6 +67,23 @@ cpu1_intc: interrupt-controller {
>>>>>>                         };
>>>>>>                 };
>>>>>>
>>>>>> +               E24: cpu@2 {
>>>>>> +                       compatible = "sifive,e24", "riscv";
>>>>>> +                       reg = <2>;
>>>>>> +                       device_type = "cpu";
>>>>>> +                       i-cache-block-size = <32>;
>>>>>> +                       i-cache-sets = <256>;
>>>>>> +                       i-cache-size = <16384>;
>>>>>> +                       riscv,isa = "rv32imafc";
>>>>>> +                       status = "disabled";
>>>>>> +
>>>>>> +                       cpu2_intc: interrupt-controller {
>>>>>> +                               compatible = "riscv,cpu-
>>>>>> intc";
>>>>>> +                               interrupt-controller;
>>>>>> +                               #interrupt-cells = <1>;
>>>>>> +                       };
>>>>>> +               };
>>>>>> +
>>>>>>                 cpu-map {
>>>>>>                         cluster0 {
>>>>>>                                 core0 {
>>>>>> @@ -76,6 +93,10 @@ core0 {
>>>>>>                                 core1 {
>>>>>>                                         cpu = <&U74_1>;
>>>>>>                                 };
>>>>>> +
>>>>>> +                               core2 {
>>>>>> +                                       cpu = <&E24>;
>>>>>> +                               };
>>>>>
>>>>> Sorry but I think this change makes the topology more
>>>>> inaccurate.
>>>>>
>>>>> The E24 core is very independent, just another CPU core
>>>>> connected
>>>>> the
>>>>> same bus -- even no coherency (E24 takes AHB, which is not
>>>>> coherency-
>>>>> sensible). Even the TAP of it is independent with the U74 TAP.
>>>>>
>>>>> And by default it does not boot any proper code (if a debugger
>>>>> is
>>>>> attached, it will discover that the E24 is in consistently
>>>>> fault at
>>>>> 0x0
>>>>> (mtvec is 0x0 and when fault it jumps to 0x0 and fault again),
>>>>> until
>>>>> its clock is just shutdown by Linux cleaning up unused clocks.)
>>>>>
>>>>> Personally I think it should be implemented as a remoteproc
>>>>> instead.
>>>>
>>>> Maybe I am missing something, but I don't quite get what the
>>>> detail
>>>> of how we access this in code has to do with the devicetree?
>>>> It is added here in a disabled state, and will not be used by
>>>> Linux.
>>>> The various SiFive SoCs & SiFive corecomplex users that have a
>>>> hart
>>>> not capable of running Linux also have that hart documented in
>>>> the
>>>> devicetree.
>>>> To me, what we are choosing to do with this hart does not really
>>>> matter very much, since this is a description of what the
>>>> hardware
>>>> actually looks like.
>>>
>>> The E24 is not in the core complex at all. It's just a dedicate CPU
>>> connected to another bus (well as I saw the document says the E24
>>> bus
>>> is maximum 2G, I doubt whether it's the same bus with the U74 one).
>>>
>>> The U74 MC only allows S5 management cores to be part of it, not
>>> E24.
>>
>> So is the correct topology more like:
>> cpu-map {
>>         cluster0 {
>>                 core0 {
>>                         cpu = <&U74_0>;
>>                 };
>>                 core1 {
>>                         cpu = <&U74_1>;
>>                 };
>>         };
>>         cluster1 {
>>                 core0 {
>>                         cpu = <&E24>;
>>                 };
>>         };
>> };
> 
> Considering E24 seems to see a total different bus connected to it, I
> don't think it even proper to add it to cpus node.

Well, it is a CPU is it not? How one is supposed to document that a
CPU is not attached to the same buses I do not know however.

> 
> And I don't think it has a hart id of 2, as your node describes.

Do you have any idea what it would be then?