Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Intel core i7 7th Gen Overheating
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
GenHeDi
Tux's lil' helper
Tux's lil' helper


Joined: 02 Jan 2016
Posts: 122

PostPosted: Tue Jun 19, 2018 5:39 pm    Post subject: Intel core i7 7th Gen Overheating Reply with quote

Hello,

Recently I bought a TUXEDO Book BX1506 which is equipped with an Intel Kaby Lake CPU: Intel Core i7-7700T (4x 2.90-3.80 GHz Quad-Core, 8-Threads, 8 MB Cache, 35W TDP) processor.

After installing Gentoo and KUbuntu 18.04 LTS, I notice that I get a lot of overheating notifications in my Gentoo dmesg log. This happens mainly during the compilation of large programs (MAKEOPTS="-j5").

When I'm surfing under KDE with Firefox and looking at youtube films, then the fan always turns on (but do not get any dmesg notifications yet).

Dmesg snippet:
Code:
[  901.311962] CPU6: Core temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311963] CPU2: Core temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311964] CPU3: Package temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311965] CPU1: Package temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311966] CPU7: Package temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311967] CPU0: Package temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311967] CPU5: Package temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311968] CPU4: Package temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311969] CPU2: Package temperature above threshold, cpu clock throttled
(total events = 1)
[  901.311971] mce: [Hardware Error]: Machine check events logged
[  901.315933] CPU2: Core temperature/speed normal
[  901.315934] CPU1: Package temperature/speed normal
[  901.315935] CPU3: Package temperature/speed normal
[  901.315936] CPU0: Package temperature/speed normal
[  901.315936] CPU5: Package temperature/speed normal
[  901.315937] CPU7: Package temperature/speed normal
[  901.315938] CPU4: Package temperature/speed normal
[  901.315938] CPU2: Package temperature/speed normal
[  901.315940] mce: [Hardware Error]: Machine check events logged
[  901.343044] CPU6: Core temperature/speed normal
[ 1196.527385] kworker/dying (5165) used greatest stack depth: 12440 bytes
left
[ 1201.316957] CPU3: Package temperature above threshold, cpu clock throttled
(total events = 35574)
[ 1201.316959] CPU0: Package temperature above threshold, cpu clock throttled
(total events = 35574)
[ 1201.316960] CPU5: Package temperature above threshold, cpu clock throttled
(total events = 35574)
[ 1201.316962] CPU2: Core temperature above threshold, cpu clock throttled
(total events = 27443)
[ 1201.316963] CPU4: Package temperature above threshold, cpu clock throttled
(total events = 35574)
[ 1201.316964] CPU1: Package temperature above threshold, cpu clock throttled
(total events = 35574)
[ 1201.316965] CPU7: Package temperature above threshold, cpu clock throttled
(total events = 35574)
[ 1201.316966] CPU2: Package temperature above threshold, cpu clock throttled
(total events = 35574)
[ 1201.316968] mce_notify_irq: 1 callbacks suppressed
[ 1201.316969] mce: [Hardware Error]: Machine check events logged
[ 1201.347961] CPU6: Core temperature/speed normal
[ 1201.347964] CPU6: Package temperature/speed normal
[ 1201.347966] mce: [Hardware Error]: Machine check events logged
[ 1501.322945] CPU2: Core temperature/speed normal
[ 1501.322946] CPU3: Package temperature/speed normal
[ 1501.322947] CPU1: Package temperature/speed normal
[ 1501.322948] CPU4: Package temperature/speed normal
[ 1501.322948] CPU5: Package temperature/speed normal
[ 1501.322949] CPU7: Package temperature/speed normal
[ 1501.322950] CPU0: Package temperature/speed normal
[ 1501.322953] CPU2: Package temperature/speed normal
[ 1501.322955] mce: [Hardware Error]: Machine check events logged
[ 1501.347933] CPU6: Core temperature above threshold, cpu clock throttled
(total events = 56135)
[ 1501.347979] CPU6: Package temperature above threshold, cpu clock throttled
(total events = 69968)
[ 1501.351817] mce: [Hardware Error]: Machine check events logged
[ 1501.373973] CPU6: Core temperature/speed normal
[ 1501.373977] CPU6: Package temperature/speed normal
[ 1801.324929] CPU0: Package temperature/speed normal
[ 1801.324930] CPU4: Package temperature/speed normal
[ 1801.324931] CPU7: Package temperature/speed normal
[ 1801.324932] CPU3: Package temperature/speed normal
[ 1801.324933] CPU2: Core temperature/speed normal
[ 1801.324936] CPU2: Package temperature/speed normal
[ 1801.324940] mce_notify_irq: 1 callbacks suppressed
[ 1801.324940] mce: [Hardware Error]: Machine check events logged
[ 1801.324958] CPU1: Package temperature/speed normal
[ 1801.324959] CPU5: Package temperature/speed normal
[ 1801.390935] CPU6: Core temperature/speed normal
[ 1801.390938] CPU6: Package temperature/speed normal
[ 1801.390941] mce: [Hardware Error]: Machine check events logged
[ 1808.075933] CPU7: Core temperature above threshold, cpu clock throttled
(total events = 25567)
[ 1808.075934] CPU3: Core temperature above threshold, cpu clock throttled
(total events = 25567)
[ 1808.076951] CPU3: Core temperature/speed normal
[ 1808.076952] CPU7: Core temperature/speed normal


I have installed the Intel Microcode package, but this solves nothing yet.

When I'm using KUbuntu, then the fan is switched on much less.

Is there someone who can help me out?

Regards,

GenHeDi
_________________
Nulla tenaci invia est via
Back to top
View user's profile Send private message
bunder
Bodhisattva
Bodhisattva


Joined: 10 Apr 2004
Posts: 5843

PostPosted: Tue Jun 19, 2018 6:06 pm    Post subject: Reply with quote

Try turning off turbo boost.

Code:
echo 1 > /sys/devices/system/cpu/intel_pstate/no_turbo

_________________
Neddyseagoon wrote:
The problem with leaving is that you can only do it once and it reduces your influence.

banned from #gentoo since sept 2017
Back to top
View user's profile Send private message
GenHeDi
Tux's lil' helper
Tux's lil' helper


Joined: 02 Jan 2016
Posts: 122

PostPosted: Wed Jun 20, 2018 4:27 pm    Post subject: Reply with quote

Hello bunder,

I had a look in the directory butthis file does not exist in my installation

Code:
# ls -l /sys/devices/system/cpu/
 totaal 0
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu0
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu1
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu2
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu3
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu4
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu5
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu6
 drwxr-xr-x  9 root root    0 20 jun  2018 cpu7
 drwxr-xr-x 10 root root    0 20 jun 18:17 cpufreq
 drwxr-xr-x  2 root root    0 20 jun 18:17 cpuidle
 drwxr-xr-x  2 root root    0 20 jun 18:17 hotplug
 -r--r--r--  1 root root 4096 20 jun 18:17 isolated
 -r--r--r--  1 root root 4096 20 jun 18:17 kernel_max
 drwxr-xr-x  2 root root    0 20 jun 18:17 microcode
 -r--r--r--  1 root root 4096 20 jun 18:17 modalias
 -r--r--r--  1 root root 4096 20 jun 18:17 offline
 -r--r--r--  1 root root 4096 20 jun 18:17 online
 -r--r--r--  1 root root 4096 20 jun 18:17 possible
 drwxr-xr-x  2 root root    0 20 jun 18:17 power
 -r--r--r--  1 root root 4096 20 jun 18:17 present
 -rw-r--r--  1 root root 4096 20 jun  2018 uevent
 drwxr-xr-x  2 root root    0 20 jun 18:17 vulnerabilities


In my KUbuntu installation it does exist and it has the value of 0

How do I get my Gentoo directory structure up to spec's?

Regards
_________________
Nulla tenaci invia est via
Back to top
View user's profile Send private message
bunder
Bodhisattva
Bodhisattva


Joined: 10 Apr 2004
Posts: 5843

PostPosted: Wed Jun 20, 2018 4:46 pm    Post subject: Reply with quote

I think you might be missing CONFIG_X86_INTEL_PSTATE=y

Cheers
_________________
Neddyseagoon wrote:
The problem with leaving is that you can only do it once and it reduces your influence.

banned from #gentoo since sept 2017
Back to top
View user's profile Send private message
GenHeDi
Tux's lil' helper
Tux's lil' helper


Joined: 02 Jan 2016
Posts: 122

PostPosted: Sat Jun 23, 2018 8:07 am    Post subject: Reply with quote

You were right, it was not set. I turned it into y and now I can turn off turbo boost.
Although after a reboot it is turned on again, how to deal with that?

Yesterday I forced a huge compiling job on this laptop but unfortunately I have still overheating messages in my dmeg.
Although way less than before.

I think that still other kernel parameters are not properly set yet, like "Processor Clocking Controle interface driver" is not set yet.
Is there a proper way to double check the necessary settings / value's? I followed the instructions of the Gentoo AMD64 Handbook.

Regards
_________________
Nulla tenaci invia est via
Back to top
View user's profile Send private message
bunder
Bodhisattva
Bodhisattva


Joined: 10 Apr 2004
Posts: 5843

PostPosted: Sat Jun 23, 2018 9:28 am    Post subject: Reply with quote

I wrote a script for this... https://github.com/bunder2015/scripts/blob/master/linux/no-turbo.sh

if you put it into /etc/local.d then it should run on startup, but you can also add it to cron so it runs periodically (as sometimes it can turn itself back on, my laptop does when I unplug/replug).
_________________
Neddyseagoon wrote:
The problem with leaving is that you can only do it once and it reduces your influence.

banned from #gentoo since sept 2017
Back to top
View user's profile Send private message
GenHeDi
Tux's lil' helper
Tux's lil' helper


Joined: 02 Jan 2016
Posts: 122

PostPosted: Sun Jun 24, 2018 11:10 am    Post subject: Reply with quote

Thanks a lot,

I did put your script in my /etc/local.d and it works like a charm (after I put a .start suffix in the filename :oops: )

But after this modification I see still overheating messages in my dmesg.

Any idea how to proceed?
_________________
Nulla tenaci invia est via
Back to top
View user's profile Send private message
Jaglover
Watchman
Watchman


Joined: 29 May 2005
Posts: 7090
Location: Saint Amant, Acadiana

PostPosted: Sun Jun 24, 2018 11:16 am    Post subject: Reply with quote

Use distcc, reduce make instances, force powersave governor. Laptops are not made for compiling.
_________________
Please learn how to denote units correctly!
Back to top
View user's profile Send private message
GenHeDi
Tux's lil' helper
Tux's lil' helper


Joined: 02 Jan 2016
Posts: 122

PostPosted: Sun Jun 24, 2018 11:26 am    Post subject: Reply with quote

What is your advice regarding the "force powersave governor" part?

Regards,
_________________
Nulla tenaci invia est via
Back to top
View user's profile Send private message
bunder
Bodhisattva
Bodhisattva


Joined: 10 Apr 2004
Posts: 5843

PostPosted: Sun Jun 24, 2018 12:22 pm    Post subject: Reply with quote

I haven't had to do anything other than turn turbo off. Are you sure your cooling is working okay?

This thing looks like it was made by Clevo. If you press Fn + 1 does the fan spin up to full speed?
_________________
Neddyseagoon wrote:
The problem with leaving is that you can only do it once and it reduces your influence.

banned from #gentoo since sept 2017
Back to top
View user's profile Send private message
GenHeDi
Tux's lil' helper
Tux's lil' helper


Joined: 02 Jan 2016
Posts: 122

PostPosted: Sun Jun 24, 2018 1:06 pm    Post subject: Reply with quote

Fn + 1 does the fan spin up to full speed > confirm

I have also KUbuntu 18.04 LTS running on same system. When I'm running KUbuntu there are no logs in dmesg of thermal heating at all.

Regards,
_________________
Nulla tenaci invia est via
Back to top
View user's profile Send private message
Hu
Moderator
Moderator


Joined: 06 Mar 2007
Posts: 13836

PostPosted: Sun Jun 24, 2018 3:04 pm    Post subject: Reply with quote

When you run Kubuntu 18.04, are you using the system in a way that keeps the CPUs running at maximum load for minutes or hours at a time? The binary distributions generally don't put nearly as much load on your system as a big Gentoo emerge job.
Back to top
View user's profile Send private message
GenHeDi
Tux's lil' helper
Tux's lil' helper


Joined: 02 Jan 2016
Posts: 122

PostPosted: Sun Jun 24, 2018 4:43 pm    Post subject: Reply with quote

Your right, I don't do heavy stuff on KUbuntu like the emerge jobs on Gentoo.
_________________
Nulla tenaci invia est via
Back to top
View user's profile Send private message
fxkrait
n00b
n00b


Joined: 30 Jul 2018
Posts: 15
Location: DRPK

PostPosted: Thu Aug 02, 2018 4:38 am    Post subject: Reply with quote

I have a Clevo laptop as well, and have no issues with my CPU overheating.

Laptops are just fine for compiling all day.
I can turbo (not at max) and stay under 90 C.

Look into re-applying your thermal paste with "liquid metal", it will drastically improve your cooling capabilities.
_________________
A great painter does not content himself by affecting us with his masterpieces; ultimately, he succeeds in changing the landscape of our minds. Once a miniaturist's artistry enters our souls this way, it becomes the criterion for the beauty of our world.
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2868
Location: Bay Area, CA

PostPosted: Fri Aug 03, 2018 6:03 am    Post subject: Reply with quote

Try undervolting your CPU. See: https://github.com/georgewhewell/undervolt

I am doing a -140mV undervolt on my Dell XPS and its plenty stable. It may sound counter-intuitive but undervolting the CPU actually gave me 15% more performance from the CPU because temperature thresholds where throttling happened, were achieved with a higher CPU frequency.

BTW, throttling is a protection mechanism. CPU microcode just degrades the performance by few hundred Mhz's to let the CPU cool down. So, as such its not a longevity hit on the CPU as much as its a peak performance hit.
Back to top
View user's profile Send private message
n05ph3r42
Tux's lil' helper
Tux's lil' helper


Joined: 11 Jul 2016
Posts: 103

PostPosted: Mon Aug 06, 2018 9:22 pm    Post subject: Reply with quote

I can also recommend to force powersave manager( disable all the rest in your kernel) + install thermald. Recently i added new info about it to power management r page: Also, for Intel processors used in laptops since 2011 you may want to enable
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum