You are not logged in.
Scaling driver: intel_pstate
Scaling available governors: performance powersave
Energy performance available preferences: \
default
performance
balance_performance
balance_power
power
When i use performance scaling governer
/sys/devices/system/cpu/cpu0/cpufreq λ cat energy_performance_preference
performanceWhich is guess is normal
But, when i use powersave scaling governer
/sys/devices/system/cpu/cpu0/cpufreq λ cat energy_performance_preference
balance_performancewhich I expect it to be power instead of balance_performance
is There a way to change that? XD Thanks alot.
Last edited by longshot (2024-04-30 12:16:39)
~ A penguin geek who wants to know every file in sysfs
Offline
I don't know is there a way to configure this properly, but udev rule should work:
# /etc/udev/rules.d/80-cpu-performance.rules
KERNEL=="cpu[0-9]*", SUBSYSTEM=="cpu", ATTR{cpufreq/scaling_governor}=="powersave", ATTR{cpufreq/energy_performance_preference}="power"You may want to trigger udev to apply this right away.
Offline
I don't know is there a way to configure this properly, but udev rule should work:
# /etc/udev/rules.d/80-cpu-performance.rules KERNEL=="cpu[0-9]*", SUBSYSTEM=="cpu", ATTR{cpufreq/scaling_governor}=="powersave", ATTR{cpufreq/energy_performance_preference}="power"You may want to trigger udev to apply this right away.
Thanks alot, that acually worked when i trigger udev.
But now I can only switch Performance and Power, and I can't do balance_performance or balance_power options.
I have read some docs on intel_pstate, and I have come up with this script.
#!/bin/sh
cpupower frequency-info -p | grep -q powersave && state=performance || state=powersave
preference=$(echo -e "performance\npower\nbalance_performance\nbalance_power" | dmenu -i -l 5 -p "Using $state governer. Choose an energy preference >> " )
if [ "$preference" = performance ]
then
cpupower frequency-set -g performance
elif [ -n "$preference" ]
then
cpupower frequency-set -g powersave
echo setting preference to $preference
echo -n "$preference" | tee /sys/devices/system/cpu/cpufreq/policy*/energy_performance_preference
fi
# docs.kernel.org:
# The powersave P-state selection algorithm provided by intel_pstate is not a
# counterpart of the generic powersave governor (roughly, it corresponds to the
# schedutil and ondemand governors).
# https://docs.kernel.org/admin-guide/pm/intel_pstate.html?highlight=intel_pstate#active-mode-with-hwpand that solved the Problem.
Thanks for the influence bro @yataro
Last edited by longshot (2024-04-30 16:42:34)
~ A penguin geek who wants to know every file in sysfs
Offline