Re: [PATCH v6 00/12] Unregister critical branch clocks + some RPM

[Konrad Dybcio]

On 14.01.2024 05:53, Dmitry Baryshkov wrote:
> On Sat, 13 Jan 2024 at 16:51, Konrad Dybcio <konrad.dybcio@xxxxxxxxxx> wrote:
>>
>> On Qualcomm SoCs, certain branch clocks either need to be always-on, or
>> should be if you're interested in touching some part of the hardware.
>>
>> Using CLK_IS_CRITICAL for this purpose sounds like a genius idea,
>> however that messes with the runtime pm handling - if a clock is
>> marked as such, the clock controller device will never enter the
>> "suspended" state, leaving the associated resources online, which in
>> turn breaks SoC-wide suspend.
>>
>> This series aims to solve that on a couple SoCs that I could test the
>> changes on and it sprinkles some runtime pm enablement atop these drivers.
> 
> Probably it is out of scope for this
> I wonder if it makes sense to route (some) of the clocks properly.
> Should we use GCC_foo_SLEEEP_CLK as a sleep clock for the
> corresponding device?
> I'm not sure about the AHB and XO clocks.
> 
> Another question is regarding the suspended state. Wouldn't leaving
> GCC_foo_XO clocks enabled keep the XO enabled as well?

Doesn't seem to be the case

Konrad