On 8/23/21 5:05 PM, Emanuel Berg wrote:
David Christensen wrote:
But changing the profile (governor) doesn't produce any
(noticable?) sound level change and also the temperature of
the CPU and the GPU seem unaffected.
You will not notice a change in CPU fan temperature or speed
profiles as a function of the Linux governor setting until
the motherboard CPU fan speed control is working and you
change the CPU load.
Here is a Perl one-liner that should peg one core:
$ perl -e "1 while 1"
OK, I wrote a script [last] and here is the result
$ temp-gov
CPU C (33.50 45.75) (100 iterations) conservative
CPU C (45.88 46.38) (100 iterations) userspace
CPU C (46.38 46.75) (100 iterations) powersave
CPU C (46.75 47.00) (100 iterations) ondemand
CPU C (47.00 47.00) (100 iterations) performance
CPU C (47.00 47.25) (100 iterations) schedutil
Hm ... the result seems to be pretty much identical for the
governors anyway?
#! /bin/zsh
#
# this file:
# https://dataswamp.org/~incal/conf/.zsh/temp
temp-gov () {
local t=100
local cpu
local cpu_min
local cpu_max
perl -e '1 while 1' &
local pid=$!
sleep 10
local i
local g
for g in $(cpufreq-info -g); do
sudo cpufreq-set -g $g
cpu_min=999
cpu_max=0
for i in {0..$t}; do
cpu=$(sensors -j | jq -a '.["k10temp-pci-00c3"].Tdie.temp1_input')
(( $cpu < $cpu_min )) && cpu_min=$cpu
(( $cpu > $cpu_max )) && cpu_max=$cpu
done
printf "CPU C (%.2f %.2f) (%d iterations) %s\n" $cpu_min $cpu_max $t $g
done
kill $pid
}
For a CPU with N cores (N=4 for an AMD Ryzen 3 3200G?) and an otherwise
unloaded system, your test procedure should be something like:
loop over governor choices
set governor
loop 3 times
sleep 60 seconds
print statistics
endloop
loop from 1 to N
start background process
loop 10 times
sleep 6 seconds
print statistics
endloop
endloop
kill all background processes
endloop
"print statistics" should include time, governor setting, number of
background processes running, and CPU temperature. If would be nice to
also include system loading percent, CPU frequency, and CPU fan speed.
David
