Re: [PATCH] ARM: tegra: Fix restoration of PLLM when exiting suspend

[Dmitry Osipenko <digetx@gmail.com>]

11.12.2019 11:50, Peter De Schrijver пишет:
> On Tue, Dec 10, 2019 at 11:29:42PM +0300, Dmitry Osipenko wrote:
>> External email: Use caution opening links or attachments
>>
>>
>> 10.12.2019 22:28, Dmitry Osipenko пишет:
>>> Hello Jon,
>>>
>>> PLLM's enable-status could be defined either by PMC or CaR. Thus at
>>> first you need to check whether PMC overrides CaR's enable and then
>>> judge the enable state based on PMC or CaR state respectively.
>>>
>>
>> Actually, now I think that it doesn't make sense to check PMC WB0 state
>> at all. IIUC, PLLM's state of the WB0 register defines whether Boot ROM
>> should enable PLLM on resume from suspend. Thus it will be correct to
>> check only the CaR's enable-state of PLLM.
>>
>> I'm not sure what's the idea of WB0 overriding, maybe to resume faster.
>> Peter, could you please clarify that?
> 
> I don't know why these overriding bits exist. The code for them was in
> the downstream driver so I implemented the same in the upstream driver
> :)

Okay, I'll try to figure out how to clean up it properly.