You are not logged in.
I only just diagnosed the supposed "why" behind all these odd failures to wake from suspend and its showing as
kernel: thermal thermal_zone0: acpitz: critical temperature reached
kernel: reboot: HARDWARE PROTECTION shutdown (Temperature too high)
kernel: reboot: Hardware protection timed-out. Trying forced poweroff
despite the fact that temperatures are 100% under control while on and it did NOT do this with to 6.8.1-arch1-1 (which I wasnt paying enough attention and forgot to save a separate copy so now I dont have it to downgrade... GDI. If someone knows a way to download a copy of the older 6.8.1 package, I would appreciate.)
This has basically made suspending majority unreliable and I honestly have no idea what change could have caused this. At the moment, the best idea I have is to try the LTS but since its lasted for 3 minor patches... its possible that eventually this would get back to me.
Hardware: Ryzen 5950x, Gigabyte x570 Aorus Master(rev 1.0), Sapphire pulse 7900xt.
Last edited by Versa (2024-04-09 15:40:13)
Offline
You can obtains older packages from the ALA You can also find bisection kernels between 6.8.1 and 6.8.2 to locate the cause in https://bbs.archlinux.org/viewtopic.php?id=294310 Follow along with the bisection results until yours differ.
Last edited by loqs (2024-04-09 15:55:16)
Online
You can obtains older packages from the ALA You can also find bisection kernels between 6.8.1 and 6.8.2 to locate the cause in https://bbs.archlinux.org/viewtopic.php?id=294310 Follow along with the bisection results until yours differ.
Thank you for pointing me at the ALA, somehow, despite using arch for 5 years already ive never come across that.
At some point when I have a bit more time than today, ill check about those bisection kernels and seeing about where it came in. So far, its been VERY easy to replicate. As long as I wake it up within a few minutes of putting it to sleep, nearly 100% replication.
Offline
ok, after downgrading again, I think I am coming to the conclusion that more than just 6.8.2, I just got lucky with 6.8.1 before this.
When deliberately trying to trigger it, it happens even on 6.8.1 AND LTS so im going to have to check more stuff it seems.
Its almost like when waking up, the crit temp it checks is lower than intended and oddly, even setting thermal.nocrt=1 in grub command line doesnt prevent it. It wont tell me the actual crit temp it reached either.
Last edited by Versa (2024-04-09 20:11:46)
Offline
Please let me know if you have any new findings, I'm having the same issue.
Happened again today, Linux 6.12.0-rc2-1-mainline.
Almost same hardware, gpu is 7900xtx.
Hardware: Ryzen 5950x, Gigabyte x570 Aorus Master(rev 1.0), Sapphire pulse 7900xtx.
Offline
Looks like a problem with the motherboard sensor reporting.
thermal thermal_zone0: acpitz: critical temperature reached
acpitz-acpi-0
Adapter: ACPI interface
temp1: +16.8°C
temp2: +16.8°C
It always shows 16.8°C. And triggers thermal shutdown on wake from suspend.
https://forum.manjaro.org/t/acpi-therma … /154502/16
In this forum post, they just disabled the option by,
echo disabled > /sys/class/thermal/thermal_zone1/mode
I had to change thermal_zone1 to thermal_zone0.
And add to my startup scripts.
Offline