Intel P-State Gets More Cleanups & Optimizations

[geearf]

I'm on PulseAudio too, but with default settings.

Could you please elaborate what "pstate passive" means?

Edit: What does i7z tell you? Is there a noteable difference in clock/vcore/states with the patch compared to without?
i7z terminal seems to be the only tool that reads out correct values for me when the pstate module isn't loaded.

If you don't mind, please re-try your SCII tests with the same resampler

I'm not exactly sure what pstate passive is, on the linux boot line instead of intel_pstate=enable or disable you put passive, this will allow you to use the acpi-cpufreq governors, but with some stuff still from pstate. From the dev: 'With acpi-cpufreq "schedutil" may under-provision things due to the way in which turbo P-states are represented in the ACPI tables.' (*maybe less with intel_pstate=passive*)

As for i7z, I'll get that information for you.

[aufkrawall]

I would have to install SC2 again. Is there really a notable quality improvement with speex-float-10 over the default?

[aufkrawall]

Some update: I noticed worse performance with schedutil in Unigine Heaven compared to pstate performance.

[geearf]

I would have to install SC2 again. Is there really a notable quality improvement with speex-float-10 over the default?

So the clock is different, which explains why it works better. What the patch does is slow the throttling of the CPU is I understand it correctly, so it makes sense to have a higher clock on average.

As for audio quality, frankly that's a great question, and I have no idea Since pulseaudio usually uses around 3% of my cpu on max I never really looked into it

[aufkrawall]

Oh latest comment I got from the dev, if the patch that fixes my issue with schedutil is accepted upstream (it may not, as it may increase power consumption...),

Hi again,
there were some changes for 4.12. Do you think the more aggressive behavior made it?

[geearf]

I do not know, there was a discussion about it on LKML, but it never concluded.
Let's wait for the merge window to close and we can check the source code to see if the patch is there.

[geearf]

Hi again,
there were some changes for 4.12. Do you think the more aggressive behavior made it?

Well I just checked and it was checked in in April, so we should be good.