Re: [PATCH V2 6/7] thermal/drivers/cpu_cooling: Introduce the cpu idle cooling driver

[Daniel Lezcano]

On 27/03/2018 14:28, Juri Lelli wrote:
> Hi Daniel,
> 
> On 27/03/18 12:26, Daniel Lezcano wrote:
>> On 27/03/2018 04:03, Leo Yan wrote:
>>> Hi Daniel,
>>>
>>> On Wed, Feb 21, 2018 at 04:29:27PM +0100, Daniel Lezcano wrote:
>>>> The cpu idle cooling driver performs synchronized idle injection across all
>>>> cpus belonging to the same cluster and offers a new method to cool down a SoC.
>>>>
>>>> Each cluster has its own idle cooling device, each core has its own idle
>>>> injection thread, each idle injection thread uses play_idle to enter idle.  In
>>>> order to reach the deepest idle state, each cooling device has the idle
>>>> injection threads synchronized together.
>>>>
>>>> It has some similarity with the intel power clamp driver but it is actually
>>>> designed to work on the ARM architecture via the DT with a mathematical proof
>>>> with the power model which comes with the Documentation.
>>>>
>>>> The idle injection cycle is fixed while the running cycle is variable. That
>>>> allows to have control on the device reactivity for the user experience. At
>>>> the mitigation point the idle threads are unparked, they play idle the
>>>> specified amount of time and they schedule themselves. The last thread sets
>>>> the next idle injection deadline and when the timer expires it wakes up all
>>>> the threads which in turn play idle again. Meanwhile the running cycle is
>>>> changed by set_cur_state.  When the mitigation ends, the threads are parked.
>>>> The algorithm is self adaptive, so there is no need to handle hotplugging.
>>>
>>> The idle injection threads are RT threads (FIFO) and I saw in
>>> play_idle() set/clear flag PF_IDLE for it.  Will these idle injection
>>> threads utilization be accounted into RT utilization?
>>>
>>> If idle injection threads utilization is accounted as RT tasks
>>> utilization, will this impact CPUFreq governor 'schedutil' for OPP
>>> selection?
>>
>> Hi Leo,
>>
>> The idle injection task has a very low utilization when it is not in the
>> play_idle function, basically it wakes up, sets a timer and play_idle().
>>
>> Regarding the use case, the idle injection is the base brick for an
>> combo cooling device with cpufreq + cpuidle. When the idle injection is
>> used alone, it is because there is no cpufreq driver for the platform.
>> If there is a cpufreq driver, then we should endup with the cpu cooling
>> device where we have control of the OPP (and there is no idle injection
>> threads) or the combo cooling device.
>>
>> Except I'm missing something, the idle injection threads won't impact
>> the OPP selection.
> 
> Mmm, they actually might. schedutil selects max OPP as soon as it sees
> an RT thread. I fear these guys might generate unwanted spikes. Maybe
> you can filter them out?

Yes, absolutely. Leo pointed it also.

We might want to change the check at:

https://elixir.bootlin.com/linux/v4.16-rc7/source/kernel/sched/cpufreq_schedutil.c#L364

in order to ignore PF_IDLE tagged tasks.



-- 
 <http://www.linaro.org/> Linaro.org â Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog