I recently got a new laptop (an Acer Aspire A515-57) and naturally moved my current configuration
(making sure to change things like stateVersion and file system options) from my old one.
Whenever I come back from the machine sleeping or reopening the lid i get 15-30 seconds before it stops responding. And its very unusual as it freezes gradually: on my browser clicking links stops working but i can still move my mouse, i can run a command on the terminal but it doesn’t execute. After some time everything stops responding completely and the only way to reboot is by holding the power button.
I looked it up a bit and this thread seemed very similar to my problem, which would make sense since I do have an Intel Alder Lake-P. The ArchWiki tells me to run intel_vbt_decode to identify duplicate VBT entries, but in my case I saw no duplicates. I tried the patch anyways but it did not fix the problem.
I’m not sure how I could get useful logs to add to this issue.
I’ve enabled the kernel parameters: boot.kernelParams = [ "i915.force_probe=46a3" "initcall_debug" "pm_debug" "no_console_suspend" ]; and got this really big log from journalctl -b -1 -k --priority 4 -o cat.
I have no clue how to read this and really need some help. I’ve also noticed my /var/lib/systemd/pstore got filled up so if that’s need for further context I can provide it.
hello
so what is the output of this:
intel_vbt_decode /sys/kernel/debug/dri/1/i915_vbt
Also could you provide more infos of your system?
Could you also try
https://wiki.archlinux.org/title/Intel_graphics#Crash/freeze_on_low_power_Intel_CPUs
Could you also post the demsg log?
Also I could not find in your configuration where you applied the kernel Patch mentioned in the arch wiki.
Ok, so intel_vbt_decode gives me this. I gave up on using the VBT patch provided on the other thread as it didn’t seem to fix my problem, it conflicted with newer kernel versions, and the ArchWiki states that the problem stems from duplicate eDP entries which is not my case I think.
I don’t know how to get the dmesg output from a previous boot that crashed (outside of journalctl -b -1 as I provided above) but the output from this boot is this.
I’m now trying the low power fix from the ArchWiki:
services.xserver.config = lib.mkAfter ''
Section "Device"
Identifier "Intel Graphics"
Driver "modesetting"
Option "NoAccel" "True"
EndSection
'';
And will see if this works today. I had forgotten to push my changes to remote but all of it is in the affected loamy-sand system.
so I don’t really understand what you want to do with this:
Also can you post the output of this command with the /sys/kernel/debug…:
intel_vbt_decode /sys/kernel/debug/dri/1/i915_vbt
You said it stops responding. Does this also mean you can’t go to a different tty?
Oh, my mistake sorry, I was looking at the wrong section of the wiki. I’ve added the kernel parameters, and will see if it fixes it.
For a more involved description of the hang, last night it hanged on me while playing a lightweight game. My inputs (keybord, touchpad) suddenly stopped registering while the game kept running, this included changing tty. I had the idea to plug in another keyboard and for a brief moment was able to alt-tab out of the game before it too stopped responding. At this point I held the power button and rebooted.
Output of intel_vbt_decode: VBT header: VBT signature: "$VBT ALDERLAKE-P " VBT version: 0x0064 (1.0 - Pastebin.com
So it crashes even when it wasn’t in sleep?
If that is the case than the range of problems increases.
Maybe take a look at your temp and see if they are unreasonably high.
Also does this happen with X or wayland?
I have this problem sometimes with my laptop too (core i7 7500u). But there the problem is that I am in powersaver mode and if I wait a minute I can use it as normal. Have you got any powersaving settings on?
Before it crashed last night it had been asleep before. It’s only after I wake it that it freezes, sometimes right away, sometimes after a few minutes. I know that that’s the problem as I’ve tried leaving the laptop on for hours while manually blocking idle detection and there were no crashes.
I’m on Plasma 6 wayland. I haven’t changed any powersaving settings through my NixOS configuration, but I do have the power profile from Plasma’s power settings set to power save. Could that be it?
you could try to change the powersetting in kde to balance or even performance and see if it gets better.
If not try running your session in Xorg and see if that “resolves” the problem.
Also can you provide the demsg logs right after it suspends form sleep (maybe something is not really “waken” up)