You are not logged in.

#1 2014-12-21 23:52:28

mrgreen7
Member
Registered: 2014-12-21
Posts: 13

[SOLVED] What to do about overheating CPU?

I kept getting

Package temperature above threshold, cpu clock throttled.

so I thought I'd set the maximum cpu frequency from 3700000 to 2700000 which should basically disable the turbo boost by intel. I tried

cpupower frequency-set -u 2700000

as it is suggested in https://wiki.archlinux.org/index.php/CP … requencies.

However doing

sudo cat /sys/devices/system/cpu/*/cpufreq/scaling_max_freq

after a reboot still gives back 3700000. Did my change via cpupower not take effect? Is the command I used for checking the current maximum frequency wrong?

Last edited by mrgreen7 (2015-11-30 22:14:06)

Offline

#2 2014-12-22 09:52:20

Raynman
Member
Registered: 2011-10-22
Posts: 1,539

Re: [SOLVED] What to do about overheating CPU?

It probably did take effect, but then you rebooted.

Read the second item of the note in the wiki section. And then you should enable cpupower.service, I think.

Last edited by Raynman (2014-12-22 09:53:02)

Offline

#3 2014-12-22 13:45:29

ANOKNUSA
Member
Registered: 2010-10-22
Posts: 2,141

Re: [SOLVED] What to do about overheating CPU?

If you've got an Intel CPUl, trying to manually lock down the processor speed is the last thing you should do. The first thing is clean out the machine case/chassis; the second is replace the thermal paste on the CPU and heat sink; and the third is figure out what the average running temperature of the CPU is after you've done the first two. Only after that should you consider manually throttling anything. Turbo boost shouldn't be in effect long enough to cause any dangerous heat spikes, and the in-kernel P-state driver intelligently handles throttling on its own. What you're trying to do is akin to breaking a marathon runner's legs so she won't get heat stroke while running.

Offline

#4 2014-12-22 17:04:47

mrgreen7
Member
Registered: 2014-12-21
Posts: 13

Re: [SOLVED] What to do about overheating CPU?

Thank you for the hints. By the way the machine is a T440p with a Intel(R) Core(TM) i7-4800MQ. The machine is about 4-5 months old.

I cleaned the fan but the overheating is still happening. I've got not thermal paste at hand at the moment. Frankly, I expect this not to be necessary for a machine of this age.

In case it helps somehow: Here are the relevant parts of dmesg:

[  940.111653] CPU4: Core temperature above threshold, cpu clock throttled (total events = 39584)
[  940.111654] CPU5: Core temperature above threshold, cpu clock throttled (total events = 39584)
[  940.111667] CPU5: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.111670] CPU4: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.111672] CPU1: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.111673] CPU0: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.111674] CPU3: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.111676] CPU6: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.111677] CPU7: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.111679] CPU2: Package temperature above threshold, cpu clock throttled (total events = 46523)
[  940.112635] CPU4: Core temperature/speed normal
[  940.112636] CPU7: Package temperature/speed normal
[  940.112637] CPU5: Core temperature/speed normal
[  940.112638] CPU6: Package temperature/speed normal
[  940.112649] CPU5: Package temperature/speed normal
[  940.112651] CPU4: Package temperature/speed normal
[  940.112654] CPU0: Package temperature/speed normal
[  940.112655] CPU2: Package temperature/speed normal
[  940.112656] CPU1: Package temperature/speed normal
[  940.112656] CPU3: Package temperature/speed normal
[ 1049.924457] mce: [Hardware Error]: Machine check events logged

mce log:

Hardware event. This is not a software error.
MCE 0
CPU 2 THERMAL EVENT TSC 29bd9a79e6 
TIME 1419179950 Sun Dec 21 17:39:10 2014
Processor 2 heated above trip temperature. Throttling enabled.
Please check your system cooling. Performance will be impacted
STATUS 88080c03 MCGSTATUS 0
MCGCAP c09 APICID 2 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60
Hardware event. This is not a software error.
MCE 1
CPU 3 THERMAL EVENT TSC 29bd9ac0f3 
TIME 1419179950 Sun Dec 21 17:39:10 2014
Processor 3 heated above trip temperature. Throttling enabled.
Please check your system cooling. Performance will be impacted
STATUS 88080c03 MCGSTATUS 0
MCGCAP c09 APICID 3 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60
Hardware event. This is not a software error.
MCE 2
CPU 2 THERMAL EVENT TSC 29bdc2f9e9 
TIME 1419179950 Sun Dec 21 17:39:10 2014
Processor 2 below trip temperature. Throttling disabled
STATUS 880a0802 MCGSTATUS 0
MCGCAP c09 APICID 2 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60
Hardware event. This is not a software error.
MCE 3
CPU 3 THERMAL EVENT TSC 29bdc326f2 
TIME 1419179950 Sun Dec 21 17:39:10 2014
Processor 3 below trip temperature. Throttling disabled
STATUS 880a0802 MCGSTATUS 0
MCGCAP c09 APICID 3 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60
Hardware event. This is not a software error.
MCE 4
CPU 5 THERMAL EVENT TSC 1415d339d71 
TIME 1419180396 Sun Dec 21 17:46:36 2014
Processor 5 heated above trip temperature. Throttling enabled.
Please check your system cooling. Performance will be impacted
STATUS 88080803 MCGSTATUS 0
MCGCAP c09 APICID 5 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60
Hardware event. This is not a software error.
MCE 5
CPU 4 THERMAL EVENT TSC 1415d33c726 
TIME 1419180396 Sun Dec 21 17:46:36 2014
Processor 4 heated above trip temperature. Throttling enabled.
Please check your system cooling. Performance will be impacted
STATUS 88080803 MCGSTATUS 0
MCGCAP c09 APICID 4 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60
Hardware event. This is not a software error.
MCE 6
CPU 4 THERMAL EVENT TSC 1415d5ca4ba 
TIME 1419180396 Sun Dec 21 17:46:36 2014
Processor 4 below trip temperature. Throttling disabled
STATUS 88090802 MCGSTATUS 0
MCGCAP c09 APICID 4 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60
Hardware event. This is not a software error.
MCE 7
CPU 5 THERMAL EVENT TSC 1415d5cb200 
TIME 1419180396 Sun Dec 21 17:46:36 2014
Processor 5 below trip temperature. Throttling disabled
STATUS 88090802 MCGSTATUS 0
MCGCAP c09 APICID 5 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 60

For testing I started burnP6 from cpuburn 4 times (quad core) and so far it seems as if the overheating occurs very quickly after I started the burnP6 programs i.e. when the fan is still accelerating and did not reach the rpm level on which it stays roughly constant. Once it reaches that level no overheating messages occur. Could it thus be that something is wrong with the sensors/the fan in the sense that the fan reacts too late? If so, what can I do about it?/ Is this controlled by the OS or is that completely hardware bound?

Offline

#5 2014-12-24 01:48:22

Buddlespit
Member
From: Chesapeake, Va.
Registered: 2014-02-07
Posts: 501

Re: [SOLVED] What to do about overheating CPU?

I think you're trying to reach around your butt to scratch your ear. 1 month or 10 years old, the first thing you do with an overheating issue is clean the fan/heat sink. The second thing is re-paste the cpu. After that, you need to contact the manufacturer and notify them that your 5 month old computer is overheating.
In all honesty, it really sounds like the thermal paste is working poorly (bad paste, bad application).

Offline

Board footer

Powered by FluxBB