Re: [RFC/RFT][PATCH] cpufreq: schedutil: Reduce frequencies slower

From: Rafael J. Wysocki
Date: Sat Apr 01 2017 - 22:04:59 EST


On Sun, Apr 2, 2017 at 1:29 AM, Andres Oportus
<andresoportus@xxxxxxxxxxx> wrote:
> On Sat, Apr 1, 2017 at 1:39 PM, Andres Oportus
> <andresoportus@xxxxxxxxxxx> wrote:
>> Hi Rafael, Juri,

[cut]

>>>
>>> > As we discussed at the last LPC, having an energy model handy and use
>>> > that to decide how quickly to ramp up or slow down seems the desirable
>>> > long term solution, but we probably need something (as you are
>>> > proposing) until we get there.
>>>
>>> Well, we definitely need something to address real use cases, like the one
>>> that
>>> I responded to with this patch. :-)
>>
>> I don't know the history/intent behind schedutil rate limiting, but if we

Basically, the hardware cannot be requested to change the frequency
too often as that would be too expensive in general.

>> make it to be only "down" as Juri mentioned we would not be adding a new
>> tunable but rather changing the current one to be more restricted (maybe
>> some renaming would be in order if this is done), this would provide
>> hysteresis to reduce this problem without locking the amount of the
>> hysteresis which may not work for all platforms. I also agree that "it is
>> difficult to imagine that the same values will always be suitable for every
>> workload", but without any value to control the whole system, we get nothing
>> in between. Ultimately I also think we should solve the hysteresis problem
>> at the root, i.e. the input to the governor in the form of util/load that
>> has not only hysteresis and energy model, but also any other behavioral
>> inputs built-in.

That's long-term, though, and besides knobs only help if users change
the defaults. From my experience they usually don't do that.

Thanks,
Rafael