You are not logged in.
Pages: 1
Hi,
Hope someone can help me out with some overheating issues (95c+) I am encountering in arch, but I am not seeing in windows.
In windows I used prime95 to stress test and didnt get temps above 70, meanwhile using stress --cpu 4 in arch reaches very high temperatures quickly
sensors output
ERROR: Can't get value of subfeature temp1_max_alarm: Can't read
ERROR: Can't get value of subfeature temp1_max_alarm: Can't read
coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +101.0°C (high = +110.0°C, crit = +110.0°C)
Core 0: +85.0°C (high = +110.0°C, crit = +110.0°C)
Core 1: +86.0°C (high = +110.0°C, crit = +110.0°C)
Core 2: +85.0°C (high = +110.0°C, crit = +110.0°C)
Core 3: +85.0°C (high = +110.0°C, crit = +110.0°C)
Core 4: +85.0°C (high = +110.0°C, crit = +110.0°C)
Core 5: +85.0°C (high = +110.0°C, crit = +110.0°C)
Core 6: +85.0°C (high = +110.0°C, crit = +110.0°C)
Core 7: +85.0°C (high = +110.0°C, crit = +110.0°C)
Core 8: +97.0°C (high = +110.0°C, crit = +110.0°C)
Core 12: +101.0°C (high = +110.0°C, crit = +110.0°C)
Core 16: +92.0°C (high = +110.0°C, crit = +110.0°C)
Core 20: +90.0°C (high = +110.0°C, crit = +110.0°C)
Core 24: +100.0°C (high = +110.0°C, crit = +110.0°C)
Core 28: +100.0°C (high = +110.0°C, crit = +110.0°C)
Core 32: +82.0°C (high = +110.0°C, crit = +110.0°C)
Core 33: +82.0°C (high = +110.0°C, crit = +110.0°C)
ucsi_source_psy_USBC000:001-isa-0000
Adapter: ISA adapter
in0: 20.00 V (min = +5.00 V, max = +0.00 V)
curr1: 2.25 A (max = +0.00 A)
spd5118-i2c-3-50
Adapter: SMBus I801 adapter at efa0
temp1: +48.2°C (low = +0.0°C, high = +55.0°C)
(crit low = +0.0°C, crit = +85.0°C)
BAT0-acpi-0
Adapter: ACPI interface
in0: 15.96 V
power1: 0.00 W
ucsi_source_psy_USBC000:002-isa-0000
Adapter: ISA adapter
in0: 0.00 V (min = +0.00 V, max = +0.00 V)
curr1: 0.00 A (max = +0.00 A)
iwlwifi_1-virtual-0Then while being this hot, the fans are not actually going at full speed, they seem very quiet.
The laptop is less than 2 years old and I have been running arch on it since I got it. I don't recall having overheating issues until in maybe 2 months ago it started happening.
Thanks for helping out.
Offline
Don't compare apples and oranges.
https://wiki.archlinux.org/title/Stress_testing#MPrime
Generally, if you're throwing infinite load at the CPU it will heat up until it runs into TDP constraints and then recruit the fans to keep it within limits.
If the CPU doesn't heat up on windows you're not equally stressing it.
Next to the running task the scheduler (performance ./. powersaving), turbo-boost and security mitigations may impose performance (thus energy demand, thus heat) limits.
https://wiki.archlinux.org/title/CPU_frequency_scaling
https://wiki.archlinux.org/title/Improv … rmance#CPU
Do you have (actual) problems outside the synthetic tests?
Offline
I have the overheating issue when I do pacman --Syu and am compiling rust code. I have turbo turned off in the bios and have tried the different modes (performance, balanced and powersaving), but doesnt seem to make much of a difference. I am using Coretemp in windows to check the temperatures and it does say load 100%. Could it be that windows limits the tdp and arch doesnt?
Offline
Does the CPU heat up and clock down rather than the fans engaging?
Did you configure the BIOS/UEFI to let the OS control the fans?
https://wiki.archlinux.org/title/Fan_speed_control
Offline
The cpu does not clock down and the fans only starts slowly spinning when reaching above 95c and above 100 I think its max speed for the fans.
I have no fan options in the bios. I have tried to configure nbfc, but I found no matching config, not even in the revive project.
Offline
fans only starts slowly spinning when reaching above 95c and above 100 I think its max speed
So they start really late and ramp up fast where on windows they'd start earlier and increment slower (or never)?
You could try to lie to the ACPI and see whether that impacts the behavior,
acpi_osi=! acpi_osi="Windows 2015"https://learn.microsoft.com/en-us/windo … inacpi-osi
I don't recall having overheating issues until in maybe 2 months ago it started happening.
Any kernel, firmware or microcode updates during that time? Or did you just not stress the system?
nb. the system isn't technically overheating if the CPU doesn't emergency-shutdown and also the sensor read-out might be wrong (maybe it's 100°F
)
Does the system get notably hotter to the touch?
Ceterum censeo: 3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.
Offline
Pages: 1