Unexpected system error at every boot, and keyboard disabled after suspend

Ask Fedora
1

Hello,

I got a brand new 2-in-1 laptop with a touchscreen (Teclast F5 R) running Fedora Workstation. Everything works fine except a few things:

The screen orientation is tilted to the right (I could easily fix that by locking the screen orientation, but the sensors are not recognised correctly, so if I ever unlock the screen orientation, it reverts back to its default, tilted to the right, state. I don’t really care about it much because I don’t use this laptop as a tablet anyway, but I’m still mentioning it here because it’s maybe linked to the other issues I encounter.

Every boot, I get a hardware error. Here is an example of my logs after a reboot.

If I suspend my computer, sometimes (most of the times), the keyboard and the touchpad are disabled. I can use the touchscreen and the virtual keyboard just fine, however. Suspending the computer again, or opening/closing the screen does not change anything. Here is an example of the logs when the issue happens (the important part is at the end, as I rebooted the machine when it locked the keyboard/touchpad).

Here is the output of dmesg:

[silejonu@F5-R ~]$ dmesg | grep microcode
[ 0.000000] microcode: microcode updated early to revision 0x38, date = 2019-01-15
[ 0.670470] mce: [Hardware Error]: PROCESSOR 0:506c9 TIME 1576269600 SOCKET 0 APIC 0 microcode 38
[ 2.813157] microcode: sig=0x506c9, pf=0x1, revision=0x38
[ 2.813215] microcode: Microcode Update Driver: v2.2.

And the output of journalctl -b:

[silejonu@F5-R ~]$ journalctl -b | grep mce
déc. 13 21:40:04 localhost.localdomain kernel: mce: CPU0: Thermal monitoring enabled (TM1)
déc. 13 21:40:04 localhost.localdomain kernel: mce: [Hardware Error]: Machine check events logged
déc. 13 21:40:04 localhost.localdomain kernel: mce: [Hardware Error]: CPU 0: Machine Check: 0 Bank 4: a600000000020408
déc. 13 21:40:04 localhost.localdomain kernel: mce: [Hardware Error]: TSC 0 ADDR fef48540
déc. 13 21:40:04 localhost.localdomain kernel: mce: [Hardware Error]: PROCESSOR 0:506c9 TIME 1576269600 SOCKET 0 APIC 0 microcode 38
déc. 13 21:40:10 localhost.localdomain audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg=‘unit=mcelog comm=“systemd” exe=“/usr/lib/systemd/systemd” hostname=? addr=? terminal=? res=success’
déc. 13 21:40:10 localhost.localdomain mcelog[692]: mcelog: Family 6 Model 92 CPU: only decoding architectural errors
déc. 13 21:40:10 localhost.localdomain mcelog[692]: mcelog: warning: 8 bytes ignored in each record
déc. 13 21:40:10 localhost.localdomain mcelog[692]: mcelog: consider an update

I spent hours searching for a hint of a solution, but couldn’t find anything. Does anyone have any idea of what the issue may be?

Thanks in advance for any help. =)

1 Like
2

@silejonu Have a nice day!

Based on the data I found, it seems to be a hardware part error.

Machine Check Exception (MCE) is an error that occurs when a hardware problem is detected by the computer’s CPU. In general, if a hardware failure is imminent, the kernel panics to prevent data corruption.

Machine tests may indicate hardware failures, system overheating, bad DIMMs, or other problems. Some serious MCEs can usually only be resolved by rebooting and hardware replacement.

mcelog handles machine checks on newer x86 Linux systems, especially memory and CPU hardware errors.

This is a hardware error and may require a system check, but there are some opinions that the Intel part is not an operating system problem but a hardware part. Note please.

https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=1467040
3

Hi, thanks for your response!
I highly doubt this is a hardware issue, as the laptop is brand new, and I didn’t encounter any issue with the preinstalled Windows 10 (though I didn’t use it for very long).
I suspect an option with my processor (Intel Celeron N3450) that I have to disable in the UEFI, but I can’t put my finger on which one it may be.

1 Like
4

@silejonu did you ever figure out what was causing this for you? I’ve had the same issue since June with both Fedora 32 and now 33. Not a huge deal but an annoyance that happens on every reboot.

5

I don’t have this machine anymore, so I’ll answer from memory.
A kernel update on Fedora 32 fixed the touch-pad issue for me (though I can’t recommend which one), but the keyboard remained buggy after suspend (and maybe the touchpad, I don’t remember).
I used a script that I launched (with sudo) at boot if my touchpad didn’t respond:
#/bin/bash
echo ‘on’ > /sys/bus/i2c/devices/i2c-SYNA3602:00/power/control
rmmod i2c_hid
modprobe i2c_hid
I launched it until it fixed my touchpad, usually 2-3 times, but up to 10.

1 Like