Processor stuck at min frequency...

Hi all,

Since last Friday, I’ve been having a very weird issue with my computer. I woke it up from sleep and it began to behave completely erratically: the LED for “battery charging” was always on and it was extremely slow. I of course tried to reboot it, but the issues remained: slowness and (possibly unrelated?) it somehow thinks to be always charging, whether AC is plugged or not (funny thing: the LED correctly changes its colour when charged, from orange to green and then back to orange when it starts discharging, all still thinking AC is always on!!).

I tried everything over the weekend to find the reason for these weird issues. I tested on a USB drive system, it was still a bit slow, but hard to tell whether it was because of being on USB or not, I reinstalled the system, disconnected/reconnected the battery, changed the SSD for another (and reinstalled again!), changed the RAM, flashed the BIOS, all of it to no avail. The laptop is still slow.

But now, I have found, at least, the reason for it being slow: the CPU frequency is always at its “resting” minimal frequency (400Mhz). I tried to force it through software by changing the governor or directly the frequency (using cpupower frequency-set), but of course, it didn’t do anything. I looked at the temperature, it’s not because the temperature sensor is broken (temperature is normal using “sensors”).

Any advice at this point? I believe it all points to this being purely a hardware problem (and thus that I’ll have to buy another laptop…), but I you guys have an idea of something I could do to fix this mess, I’d very gladly read it!

Hi
Checked the BIOS settings? What model laptop and cpu?

Yes, I checked the BIOS settings, nothing there to set the CPU frequency, a very basic BIOS. Sorry, I should have added the laptop spec: it’s an Asus Transformer Flip TP301UA. Below the output of KInfoCenter:

Operating System: openSUSE Tumbleweed 20201014
KDE Plasma Version: 5.20.0
KDE Frameworks Version: 5.75.0
Qt Version: 5.15.1
Kernel Version: 5.8.14-1-default
OS Type: 64-bit
Processors: 4 × Intel® Core™ i5-6200U CPU @ 2.30GHz
Memory: 7.7 Gio of RAM
Graphics Processor: Mesa DRI Intel® HD Graphics 520

(The bug arose before the update to Plasma 5.20 by the way… I merely updated the system, in the off-chance it would automagically fix the system, you never know)

Hi
You don’t have any additional boot line commands like acpi ones?


cat /etc/default/grub | grep  GRUB_CMDLINE_LINUX_DEFAULT
lsmod | grep intel

If nothing acpi in the grub output, could try disabling intel_pstate to see if that changes it, append the following to the kernel optional parameters via YaST booloader;


intel_pstate=disable

Thanks for your advice! Here are the information you requested (nothing out of the ordinary, since I reinstalled the system):


cat /etc/default/grub | grep  GRUB_CMDLINE_LINUX_DEFAULT

**GRUB_CMDLINE_LINUX_DEFAULT**="splash=silent quiet mitigations=auto"


lsmod | grep intel

snd_soc_acpi_**intel**_match    45056  1 snd_soc_skl 
snd_soc_acpi           16384  2 snd_soc_acpi_**intel**_match,snd_soc_skl 
snd_hda_**intel**          57344  3 
snd_**intel**_dspcfg       24576  2 snd_hda_**intel**,snd_soc_skl 
snd_hda_codec         167936  5 snd_hda_codec_generic,snd_hda_codec_hdmi,snd_hda_**intel**,snd_hda_codec_realtek,snd_soc_hdac_hda 
bt**intel**                32768  1 btusb 
**intel**_powerclamp       20480  0 
snd_hda_core          110592  8 snd_hda_codec_generic,snd_hda_codec_hdmi,snd_hda_**intel**,snd_hda_ext_core,snd_hda_codec,snd_hda_codec_realtek,snd_soc_hdac_h
da,snd_soc_skl 
bluetooth             708608  43 btrtl,bt**intel**,btbcm,bnep,btusb,rfcomm 
**intel**_pmc_bxt          16384  1 iTCO_wdt 
kvm_**intel**             331776  0 
**intel**_rapl_msr         20480  0 
kvm                   860160  1 kvm_**intel** 
snd_pcm               167936  8 snd_hda_codec_hdmi,snd_hda_**intel**,snd_hda_codec,snd_compress,snd_soc_core,snd_soc_skl,snd_hda_core,snd_pcm_dmaengine 
snd                   114688  20 snd_hda_codec_generic,snd_seq,snd_seq_device,snd_hda_codec_hdmi,snd_hwdep,snd_hda_**intel**,snd_hda_codec,snd_hda_codec_realt
ek,snd_timer,snd_compress,snd_soc_core,snd_pcm 
**intel**_lpss_pci         24576  0 
**intel**_lpss             16384  1 **intel**_lpss_pci 
**intel**_xhci_usb_role_switch    16384  0 
roles                  16384  1 **intel**_xhci_usb_role_switch 
**intel**_pch_thermal      16384  0 
**intel**_rapl_common      32768  2 **intel**_rapl_msr,processor_thermal_device 
**intel**_soc_dts_iosf     20480  1 processor_thermal_device 
**intel**_ishtp_loader     24576  0 
**intel**_ishtp_hid        28672  0 
ghash_clmulni_**intel**    16384  0 
aesni_**intel**           372736  8 
glue_helper            16384  1 aesni_**intel** 
crypto_simd            16384  1 aesni_**intel** 
cryptd                 28672  3 crypto_simd,ghash_clmulni_**intel** 
**intel**_ish_ipc          28672  0 
**intel**_ishtp            61440  4 cros_ec_ishtp,**intel**_ishtp_hid,**intel**_ish_ipc,**intel**_ishtp_loader 
crc32c_**intel**           24576  5

I’ll try the following right now and report right away.


intel_pstate=disable

I’m sorry to report that disable intel pstate didn’t change anything to the issue. The computer is still stuck at 400Mhz, even when the CPU is used:


**cat** /sys/devices/system/cpu/cpu*/cpufreq/scaling_cur_freq

399999 
400000 
400000 
399999

Hi
OK, remove that setting, reboot into the older kernel from grub - 5.8.12 is it working? If it is then a bug report is needed… openSUSE:Submitting bug reports - openSUSE

Hi
As a side note, if the earlier kernel does work, then you might want to look at updating the multi kernel version to keep that one as the next update pulls in kernel 5.8.15 which may or maynot work…

I don’t have the older kernel since I reinstalled the system since the issue. But if I get this right, you wanted to check this wasn’t a kernel bug, is that it? To check that, I booted a newly downloaded USB live of Leap (so, different, more “robust” kernel), and my CPU is still stuck at 400Mhz (checked with the command in my previous message, while the CPU was busy, directly from the live USB). I guess this means it’s coming from the hardware, right?

Hi
Sounds like it :frowning: Does the system have a fan? Do you hear it running? If so, first thing I would look at is cleaning out the fan… likely dust bunnies and overheating (check temps via sensors command).

Yes, that’s what I thought… Thanks for your help anyway, greatly appreciated! I guess we pushed this as far as we could from the software side. As I said, sensors are showing very reasonable temperatures (around 30-40°C) and the fan isn’t running like crazy. I might try to open the laptop once again, but I did it so many times over the previous week-end and couldn’t find anything wrong (with my very limited skills, of course…).

Hi
So do you hear the fan running? How old is the system? Might need to look at new thermal paste… The main issue is blocked vent, dust builds up against the cooling fins. Sometime blowing air into the vent will blow dust out the bottom (air intake) of the laptop. Or shine a light into the air intake and look in the vent to see if can see any light.

I’ll look into those things, thanks!

EDIT: Sorry, and yes, I can hear the fan. The laptop is 4 years old.