Laptop CPU+GPU overheating after update to 18.04 LTS

Question

I had this same issue in my last attempt to upgrade (with a clean install) from 16.04 LTS to 18.04 LTS. When I am browsing internet and specially when I am playing videos my CPU+GPU raises to fast to high levels causing throttling to fire and producing intolerable lags, leaving the system unusable for about 60 seconds or more.

With Ubuntu 16.04 LTS I never experienced this problem. Of course temperatures raised and throttling fired frequently (after all this is a laptop), but system was never unusable and video playback was smooth.

My previous solution was to downgrade to 16.04 LTS again, but now I need the newer version so I want to find a solution.

System specs:

• CPU: Intel Core i7-6500U
• GPU: Intel HD Graphics 520
• Ubuntu: 18.04 LTS
• Linux: 4.15.0-30
• Mesa: 18.0.5
• Hardware acceleration: on

UPDATE 1

Definitively it is a problem with frequency throttling. Instead of throttling CPU frequency to avoid overheating the procesor, the system is waiting until temperature reaches a limit (default of 85ºC) and then intel_powerclamp module fires, turning the system unusable for 60 seconds or more.

I would expect a progressive performance degradation from intel_powerclamp, but no, it is all or nothing. ¿Can this be configured?

My current solution is a modification of the script of this answer. It seems to work, but it is a big hack. There must be a conventional way to tweak system configuration to accomplish the same result.

UPDATE 2

If I turn off frequency turbo boost (echo "1" | sudo tee /sys/devices/system/cpu/intel_pstate/no_turbo) the problem completely disappears. Temperature never reaches the limit and the script introduced on update 1 nor intel_powerclamp fires and the laptop is smooth and usable.

Problem now is that turbo boost is disabled. With Ubuntu 16.04 LTS I never had this issue.

This is clearly a bad configuration of the thermal management subsystem of Ubuntu 18.04 LTS. Is there a way to fix it?

UPDATE 3

System is still unusable. Installing Ubuntu 18.04 LTS server and anything else than Gnome3 works. Well, it's not a novelty that Gnome sucks, but it was usable with 16.04...

Solution 1

Disable intel_pstate at boot by editing your grub configuration.

Open /etc/default/grub with your preferred text editor and locate the line that begins with:

GRUB_CMDLINE_LINUX_DEFAULT

And add intel_pstate=disable at the end of that string. Here's how mine looks after that:

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash intel_pstate=disable"

Now, execute:

sudo update-grub

And reboot your laptop. That's it. It should work as smooth and cool as with 16.04.

Some additional information: intel_pstate governor seems to be buggy, or just plain stupid and unable to detect quick temperature raises and reduce the processor speed in consequence. It seems it'd just pray to Saint Cooling Device for it to keep the CPU below the shutdown threshold. In a desktop computer with a DC9 turbine like fan such leap of faith maybe justified, but in laptops with one small fan like my XPS13 it's better to be skeptic and leave those things to the good old ACPI (Atheist Configuration and Power Interface). At least until the intel_pstate governor grow some brains.

And the turbo boost is mostly useless. Chances are that you won't lose anything for having it disabled. That feature was useful in the days of monocore 32 bit processor, but not in the 64 bit multicore world we're living on right now.

Solution 2

I had issues with the i7-6500u as well.

Changes in the Linux kernel improved performance of Intel chips, with the side effect of causing thermal clamping on devices with a low thermal limit.

I found TLP helped a lot with toning down the behavior. Turning off turbo boost is another way to tame it. You can also set the fans to constantly run or come on at a lower temperature, which raises the thermal limit of the device.

Solution 3

tlp

As mentioned on this thread tlp works wonders for keeping system under control. It works with thermald, Intel Powerclamp, Battery vs AC for USB power, etc. Although highly configurable I've never had to change the configuration settings for a pleasant Out-Of-The-Box experience. Prior to using it I had all kinds of problems with an IvyBridge laptop overheating all the time. I have it on my new Skylake laptop and the fans NEVER run except when doing Ubuntu 16.04 LTS to 18.04 upgrade.

You can get a very detailed write-up with installation instructions here: Stop cpu from overheating

---

As an aside I've read a few questions the last two weeks from people wanting to override the minimum frequency, maximum frequency or to fix the current frequency at a specific speed. As such I've decided to create a new bash+yad GUI project. A tab to display temperatures for all thermal zones and issuing notify-send (pop-up bubble messages) will be part of it. As well some sort of session log file for when you weren't watching the screen might be kept.

I just had the idea 2 hours ago whilst rereading this thread. No promises on doing the project but I think a few people would like it and I would have a lot of fun :)

Comments

[Dave A]

I experience the same issue with a i7-6500U when running CPU intensive tasks such as compressing/decompressing large files and especially compiling software with make. I tried limiting make to only use one process but that results in a longer runtime, more heat. In my case it does not throttle, but shuts the entire laptop down harshly.

[Fran Marzoa]

Turbo boost is actually pretty useless in a 64 bit system. Unless you run some 32 bit applications often, there's no real disadvantage on keeping it disabled.

[Admin]

Is there anything new to this?? I have similar problems. Ubuntu 18.04 or 20.04 overheat whereas 16.04 worked very fine, all this with default configuration. TLP doesn't solve this (even with configuring it powersave), nor the intel_pstate disabling. Lubuntu also overheats. Not a problem with the kernel version. No process taking all the power. So where is the problem???

[user3368561]

Problem is still here; only less frequent.

[WinEunuuchs2Unix]

Everyone should install TLP even before heat problems start up. Keeps system cooler, keeps fans off or low speed more often and generally everything purring along nicely. Plus it does extra stuff for when on battery power. Besides deactivating turbo boost, consider lowering the maximum frequency from say 3,500 MHz to 3,000 Mhz or 2,500 MHz. It makes things slower of course but has less heat.

[user3368561]

@WinEunuuchs2Unix The thing is that I don't need any workaround on Ubuntu 16.04. Why Ubuntu 18.04 needs them? Who knows...

[WinEunuuchs2Unix]

@user3368561 You can try the same kernel version you were using under 16.04. Something like 4.4.0-131 or 4.13.0-36. Ubuntu 16.04.5 LTS uses the 18.04 kernel now 4.15.0-29 / -30 and I had to move off of it because it prevented Laptop from suspending. I went back to older 4.13.0-36 which works fine.

[user3368561]

@WinEunuuchs2Unix I'm using 4.15.0-30 without problems. They changed something else on 18.04 that screwed up a bunch of laptops (just look at bug reports, there are hundreds of users with similar issues).

[WinEunuuchs2Unix]

Well you can't say you don't have a problem under 4.15.0-30 because your laptop is overheating and it didn't do that before...

[user3368561]

This solution didn't work on my laptop.

[user3368561]

@WinEunuuchs2Unix I was using Ubuntu 16.04 with latest kernel, so I was already using 4.15.0 without issues. Now I'm using Ubuntu 18.04 server with a light window manager and all issues are gone. Definitively this is related with Gnome 3.

[WinEunuuchs2Unix]

@user3368561 I use tlp with 16.04 and my fans never come on with my new Skylake laptop except during 16.04 to 18.04 upgrade. It runs for an hour and all 8 CPUs hit 100% load frequently. On my old Ivybridge laptop with 16.04 before tlp fan was constantly running and machine was constantly hot. Powerclamp would kick in to throttle back the CPU's with idle states. After putting in tlp temps dropped extremely and Powerlcamp hardly ever if never ran again. Powerclamp has never engaged on the new laptop. 4.15.0-29 & -30 have suspend/resume issues for me so I'm running on 4.13-0-36.

[Fran Marzoa]

It works pretty well in my XPS13, nonetheless it still enables turbo mode from time to time. You may try leaving intel_pstate enabled and disabling turbo mode altogether, since turbo mode doesn't actually improve real application working speed and it's the main cause of overheating. Just google "Ubuntu disable intel turbo boost", there are many answers on how to do so and add it to boot scripts. Good luck!

[Fran Marzoa]

BTW, there's also a Gnome Shell extension that may help you deal manually with the issue, or at least doing some testing to get a better understanding on where your problem lies on. It's called "CPU Power Manager". For both, disabling the turbo boost and using this extension, you need to have intel_pstate enabled, so if you changed your grub as suggested in my answer you'll have to reenable again by removing that intel_pstate=disable string from your GRUB_CMDLINE_LINUX_DEFAULT and executing sudo update-grub. Don't forget to reboot after that.

[user3368561]

@WinEunuuchs2Unix I tested all the fixes proposed here, tlp included. None of them worked with Ubuntu 18.04 + Gnome3.

[mohamed khalifa]

Any source on that "turbo frequency is useless" claim? I find 37% raw frequency (3.7 GHz vs. 2.7 GHz) quite significant!

[Fran Marzoa]

You missed a relevant "mostly" there. I think it depends on what you do. I did some benchmarking with multi-threaded apps with and without turbo in my i7 laptop and didn't notice any difference at all. Unfortunately I haven't found any comprehensive benchmarking comparing different intel processors with turbo enabled vs disabled, so I cannot say it's completely useless.

[Mahdi mehrabi]

thank you, but tlp just solved my CPU overheating for GPU installing the proper driver solved my problem

[Lexible]

TLP does little to nothing for overheating on my newish high end system.

[WinEunuuchs2Unix]

@Lexible My skylake system never overheats. Very new systems sometimes have to wait awhile for tlp updates. Some say they are better without tlp and built in manufacturing cooling works fine as is. Just because it doesn't work for you isn't a reason to downvote this answer. You should complain to tlp developers instead.

[Lexible]

Been waiting over a year. Is that long enough?