Benchmarks - Is PowerTOP Tuning Worthwhile For Modern AMD Linux Laptops?

[phoronix]

Phoronix: Benchmarks - Is PowerTOP Tuning Worthwhile For Modern AMD Linux Laptops?

While PowerTOP was immensely helpful when the Intel open-source project started out in 2007 for reporting untuned kernel parameters and noting what's keeping the CPU from reaching its deeper sleep states, over the past decade Linux has greatly improved when it comes to power management and better behavior out-of-the-box. PowerTOP continues to see occasional commits and new releases, but there's less talk about it these days than going back a number of years when it was a must-have for x86_64 laptops. In any case I was curious to see if following its tips still provided any meaningful difference on a modern AMD Ryzen powered laptop...

https://www.phoronix.com/scan.php?page=news_item&px=PowerTOP-ThinkPad-Ryzen-5000

[brent]

This is my experience as well. All the tunings that actually make a difference are either default or part of power profiles in firmware and kernel nowadays. powertop tunables have very little, if any, effect. The same is true for specialized tuning software like TLP to a lesser degree. TLP offers some low power "tunings" that are often harmful for battery life, even.

[sireangelus]

that's not my experience at all. i have a zephytus g14 and powertop saves me from 4 to 6w of power. You should avoid taking such conclusions on such a small sample.

[liskin]

Tweaking PowerTOP tunables primarily changes idle power, but the article only mentions benchmark results. No wonder these weren't affected positively. If anything, tuning the laptop for power saving would result in worse benchmark results, not better.

[brent]

that's not my experience at all. i have a zephytus g14 and powertop saves me from 4 to 6w of power. You should avoid taking such conclusions on such a small sample.

Then you either have buggy firmware, buggy hardware or it is a kernel bug. The kernel nowadays enables all the important things like PCIe ASPM, USB autosuspend, NVME APST and so on by default, if it deems it safe to do so. Cooperation by the firmware (ACPI) is also required to some degree. Sure you can force on some specific power saving setting, but usually there is a good reason if the kernel doesn't enable it by default.

A typical example would be PCIe ASPM: on some buggy ACPI implementations, ASPM is configured to an "always on" configuration. ASPM works fine, but the firmware simply doesn't configure a suitable policy. If you have such a probkem, file an issue with the vendor, or just make sure to buy Linux compatible hardware beforehand.

Personally I don't have much tolerance for these kinds of problems anymore. If hardware has obvious firmware bugs or Linux compatibility issues, I send it back.

[Vorpal]

Tlp is still useful in that it allows to control charge limits on the two batteries in my ThinkPad T480 though. But I should test and see if the other features are still useful or not.

[phoronix_anon]

Charging thresholds were exposed by the kernel a couple of years ago so even that TLP feature is kind of pointless now.

[Michael]

Tweaking PowerTOP tunables primarily changes idle power, but the article only mentions benchmark results. No wonder these weren't affected positively. If anything, tuning the laptop for power saving would result in worse benchmark results, not better.

There are idle times as part of the power/thermal data.

[bug77]

I think powertop's main use case is checking out you don't peripherals that won't powersave properly. Especially on laptops, where manufacturers tend to include all sorts of "custom" hardware. Lenovo is one of the manufacturers that are actually Linux-friendly, no wonder you didn't have any need for powertop there.

As a side note, toggling everything you see as bad isn't how you use powertop. I've made my mouse basically unusable, enabling one too many things on that list.

[royce]

Tweaking PowerTOP tunables primarily changes idle power, but the article only mentions benchmark results. No wonder these weren't affected positively. If anything, tuning the laptop for power saving would result in worse benchmark results, not better.

Idle power is important - unless you're background-folding proteins or mining coin, a typical workstation is going to be idle most of the time.