I am running Fedora 27 on a 2014 11 inch MacBook Air, using GNOME 3.26. I only average about 5 hours on battery instead of the 9 hours this model would last running Mac OS. I know getting the 9 hours of battery life on Linux is not feasible, but I was hoping for more than I have now. Surprisingly, according to PowerTOP, the most significant drain on the battery is the keyboard and trackpad, which are apparently a single unit.

The first lines of output from PowerTOP:

The battery reports a discharge rate of 4.77 W
The power consumed was 253 J
The estimated remaining time is 8 hours, 22 minutes

Summary: 176.4 wakeups/second,  0.0 GPU ops/seconds, 0.0 VFS ops/sec and 6.3% CPU use

Power est.              Usage       Events/s    Category       Description
  2.41 W    100.0%                      Device         USB device: Apple Internal Keyboard / Trackpad (Apple Inc.)
  1.77 W      5.0%                      Device         Display backlight
  211 mW      0.9 ms/s      41.5        Timer          tick_sched_timer
  128 mW     16.3 ms/s      15.5        Process        [PID 1854] /usr/bin/gnome-shell
 90.9 mW    213.3 µs/s      18.0        Process        [PID 8] [rcu_sched]

I have checked PowerTOP on a regular basis for several weeks, and power usage for the keyboard and trackpad are typically about 2.4 Watts as shown above, although I have seen it as low as about 250 mWatts and as high as 4 Watts. Running LXQt instead of GNOME did not make a difference, which causes me to suspect this is a driver issue. I tried Ubuntu 16.04 from a live USB for comparison, but Ubuntu's version of PowerTOP did not provide a wattage breakdown. In another Linux on Mac-hardware power consumption question, power consumption by the keyboard and trackpad was only 188 mW.

How can I reduce the keyboard and trackpads' power consumption to a reasonable level? Or is PowerTOP being inaccurate?

  • @dsstorefile Yes, there is an "autosuspend" tunable for the device. Unfortunately, enabling it makes no difference in power consumption for it. – Daniel O Feb 28 '18 at 3:46
  • @dsstorefile Yes PowerTOP consistently estimates too high for the power-remaining statistic. The sample was also taken with nothing significant running. The best I’ve actually achieved was about 6 hours. – Daniel O Feb 28 '18 at 4:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.