[Testing - Trixie] [Solved] Cpu governors absent in 6.6.13

[TrixieTest]

Hello. I updated from kernel 6.5.13-1 to 6.6.13-1 and noticed my cpu to not drop frequency like it usually did before (i have ondemand governor by default), so i started investigating a problem and realized i have less available governors in newer kernel. Performance became new default one.
6.5.13-1

Code: Select all

user1@debian-pc:~$ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors 
powersave conservative userspace ondemand performance schedutil 

6.6.13-1

Code: Select all

user1@debian-pc:~$ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors 
performance schedutil 

I tried installing linux-cpupower package like it said in this topic, but it didn't help
I'm on Debian Testing with Ryzen 5 5600, latest not Beta bios installed.

[Aki]

Hello,

I'm on Debian Testing with Ryzen 5 5600, latest not Beta bios installed.

moved from "System and Network Configuration" to "Testing and Unstable" sub-forum.

[fabien]

6.6.13-1

Code: Select all

user1@debian-pc:~$ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors 
performance schedutil 

Check if the governors are not available anyway. On my system (stable kernel and different CPU):

Code: Select all

#> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors 
performance schedutil
#> cpupower frequency-info -p
analyzing CPU 0:
  current policy: frequency should be within 1.60 GHz and 3.90 GHz.
                  The governor "schedutil" may decide which speed to use
                  within this range.

Code: Select all

#> cpupower frequency-set -g powersave
Setting cpu: 0
[...]
#> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors 
powersave performance schedutil
#> cpupower frequency-info -p
analyzing CPU 0:
  current policy: frequency should be within 1.60 GHz and 3.90 GHz.
                  The governor "powersave" may decide which speed to use
                  within this range.

Code: Select all

#> cpupower frequency-set -g conservative
Setting cpu: 0
[...]
#> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors 
conservative powersave performance schedutil
#> cpupower frequency-info -p
analyzing CPU 0:
  current policy: frequency should be within 1.60 GHz and 3.90 GHz.
                  The governor "conservative" may decide which speed to use
                  within this range.

Code: Select all

#> cpupower frequency-set -g ondemand
Setting cpu: 0
[...]
#> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors
ondemand conservative powersave performance schedutil
#> cpupower frequency-info -p
analyzing CPU 0:
  current policy: frequency should be within 1.60 GHz and 3.90 GHz.
                  The governor "ondemand" may decide which speed to use
                  within this range.

schedutil is a very efficient governor anyway, probably better than ondemand.

[TrixieTest]

Check if the governors are not available anyway. On my system (stable kernel and different CPU):

After trying to manually change governor to any on the absent ones, they appeared and are available for usage.

Code: Select all

#> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors 
userspace ondemand conservative powersave performance schedutil 

Code: Select all

#> cpupower frequency-info -p
analyzing CPU 5:
  current policy: frequency should be within 2.20 GHz and 3.50 GHz.
                  The governor "ondemand" may decide which speed to use
                  within this range.

schedutil is a very efficient governor anyway, probably better than ondemand.

Thanks, will consider switching to it.

Also found, that cpupower tries to work with more cores than CPU actually has, need to check this.

Code: Select all

:~# cpupower frequency-set -g ondemand
Setting cpu: 0
[...]
Setting cpu: 11
Following CPUs are offline:
12-31
cpupower set operation was not performed on them