On 2022-01-28 08:40, hdv@gmail wrote:
I am reasonably sure the problem lies in some form or combination of
software. Sadly, my expertise in that area is insufficient to find out
what it is exactly.
What kernel/OS/driver are you using if it is software I can try to
reproduce since I have a pretty similar card.
On the other hand: what goes in my case is not necessarily valid in
yours.
In case it is is due to overheating this script will log temperature,
power management and fan settings.
#!/bin/bash
cd /sys/class/drm/card0/device/hwmon/hwmon0
while `true`;
do date;
for x in fan1_enable fan1_input fan1_target pwm1 pwm1_enable
temp1_input;
do echo -n "$x: "; cat $x;
done;
echo; sleep 1;
done
Output will be the following about once a second:
Fri 28 Jan 2022 09:28:12 AM EST
fan1_enable: 0
fan1_input: 1714
fan1_target: 1714
pwm1: 0
pwm1_enable: 2
temp1_input: 36000
Also you can max out the fan to see if that helps:
echo 0 > /sys/class/drm/card0/device/hwmon/hwmon0/pwm1_enable
(If you can pwm1_enable after it will show 1 and not 0, but if you look
at pwm1 it will be at max of 255).
Bijan