[PATCH 2/2] arm64: dts: hi6220: add acpu_sctrl

[zhangfei]

On 2017年08月15日 23:01, Clément Bœsch wrote:
> On Tue, Aug 15, 2017 at 02:32:43PM +0100, Wei Xu wrote:
>> Hi Zhangfei,
>>
>> On 2017/4/18 2:17, Zhangfei Gao wrote:
>>> Add acpu_sctrl clock node
>>>
>>> Signed-off-by: Zhangfei Gao <zhangfei.gao at linaro.org>
>>> Signed-off-by: Li Pengcheng <lipengcheng8 at huawei.com>
>>> ---
>>>   arch/arm64/boot/dts/hisilicon/hi6220.dtsi | 6 ++++++
>>>   1 file changed, 6 insertions(+)
>>>
>>> diff --git a/arch/arm64/boot/dts/hisilicon/hi6220.dtsi b/arch/arm64/boot/dts/hisilicon/hi6220.dtsi
>>> index 470461d..710cc34 100644
>>> --- a/arch/arm64/boot/dts/hisilicon/hi6220.dtsi
>>> +++ b/arch/arm64/boot/dts/hisilicon/hi6220.dtsi
>>> @@ -262,6 +262,12 @@
>>>   			#clock-cells = <1>;
>>>   		};
>>>   
>>> +		acpu_sctrl: acpu_sctrl at f7032000 {
> I think the unit address is incorrect here (this value has been copied
> from the pm controller while it should match the 1st one the reg).
oops,
Yes, you are right, will update.
>>> +			compatible = "hisilicon,hi6220-acpu-sctrl", "syscon";
>>> +			reg = <0x0 0xf6504000 0x0 0x1000>;
>>> +			#clock-cells = <1>;
>>> +		};
>>> +
>>>   		medianoc_ade: medianoc_ade at f4520000 {
>>>   			compatible = "syscon";
>>>   			reg = <0x0 0xf4520000 0x0 0x4000>;
>>>
> Also I'm sorry I'm going to ask stupid questions:
>
> - why 's' in "sctrl" while all the other node controllers declared around
>    seems to use "ctrl"?
Just from hisilicon convention, sctrl means system control.
acpu_ctrl seems better, more match spec.
However clk part using this name already been merged, so will keep use this.
>
> - is it possible to have access to the specs where those addresses can be
>    found? The only public doc I could find is the "Hi6220V100 Multi-Mode
>    Application Processor" one where this information doesn't seem present.
https://github.com/96boards/documentation/tree/master/ConsumerEdition/HiKey/HardwareDocs
Looks only this part.
>
> - is there any work pending adding the pmu counters node (interrupt
>    GCI_SPI 99) ? AFAIK the ACPU controller added here can be used to enable
>    the counters.
This patch is part of enabling core-sight, not touching pmu.
>> Thanks!
>> Applied to the hisilicon dt tree since the driver part is in the mainline now.
>>
> Is this refering to https://github.com/hisilicon/linux-hisi/commits/next/dt64 ?
Yes.
>
> [...]
>