@bogdancovaciu I read through your post about choosing an ACPI argument but I’m still unsure which one is right for me.
My Laptop was running FreeDOS when I bought it, not Windows, not any Linux distro.
@bogdancovaciu I read through your post about choosing an ACPI argument but I’m still unsure which one is right for me.
My Laptop was running FreeDOS when I bought it, not Windows, not any Linux distro.
Not sure if ACPI kernel boot parameters would be required for your laptop model, but AFAIK the acpi_osi=! acpi_osi='Windows 2012'
works in most cases. A second guess would be acpi_osi='Windows 2018'
Is hard to tell, so probably a bit of tinkering and experimenting/testing is required from your part.
What i want to make sure is that you have installed only the video-hybrid-intel-nvidia-440xx-prime
driver, and no other. Not sure when the video-hybrid-intel-nvidia-450xx-prime
will arrive to stable branch, but that might be best choice and hassle free option. (based on rumors, not tested by myself, so don’t quote me on that)
mhwd -li
will display what drivers you have installed. Remove any other driver if is installed except the mentioned one. Then, i would add this kernel boot parameter nouveau.modeset=0
Once this is done, you should have a functional laptop and be able to make use of the prime-offload
capability, aka run prime-run glxinfo
as example, so prime-run application
would launch that application using the dGPU (Nvidia).
To have even more options and control over this, the optimus-manager
would be a good choice. I had no time to bring the old tutorial here from the archived forum, so i apologize to send you over there to check it out. Let me know how it goes.
Installed PCI configs:video-hybrid-intel-nvidia-440xx-prime 2019.10.25 false PCI
Warning: No installed USB configs!
As far as I can tell I only have the video-hybrid-intel-nvidia-440xx-prime
driver. and prime-run glxinfo
works just fun, as do other apps and games.
My only issues (besides the NVENC deal, already reported) is how my laptop boots, or rather doesn’t.
Great, we render out the possibility of a faulty driver install, so now we move our focus on GDM
and/or check if there is another DM installed?
What is from TTY the status of GDM?
systemctl status gdm
Once you log in with startx
(i suppose) or is that the login screen only gets accesible once you cicle trough multiple TTYs, what is the session type, X11 or Wayland?
To find out run this command from terminal:
echo $XDG_SESSION_TYPE
Now, if the login screen comes to “life” after you cicle trough TTY, then installing and enabling haveged
might be helpful.
sudo pacman -Syu haveged
systemctl enable haveged --now
then reboot the system.
Sorry for the crappy picture, I don’t know how to copy from tty.
Regarding echo $XDG_SESSION_TYPE
it’s X11
I’ve done
sudo pacman -Syu haveged
systemctl enable haveged --now
but that didn’t seem to do anything, I still need to cycle through ttys a few times before going back to tty and being able to log in.
Thank you very much for your ongoing assistance.
Possibly unrelated, but I got this error Possibly missing firmware for module: xhci_pci
when I ran sudo mkinitcpio -P && sudo update-grub
This is preventing the login functionality:
gkr-pam: unable to locate daemon control file
That can happen because of a couple of reasons, so we can try to fix without entering in a lot of details, just run from terminal:
dbus-launch gsettings reset org.gnome.clocks world-clocks
in case time settings is an issue.
More extreme (you will lose all custom settings but have them in that backup file):
mv ~/.config/dconf/radu ~/.config/dconf/radu.bak
More direct and with no backup, run this:
dconf reset -f /
Reinstalling some packages might also help:
sudo pacman -Syu gdm gnome-shell gnome-keyring libgnome-keyring
dbus-launch gsettings reset org.gnome.clocks world-clocks
gives
No such schema “org.gnome.clocks”
I tried dconf reset -f /
I tried sudo pacman -Syu gdm gnome-shell gnome-keyring libgnome-keyring
to no avail. Issue remains
You mean after reboot?
Yes, @bogdancovaciu after reboot I am greeted with a black screen and a blinking cursor. This time, not even switching ttys allowed me to log in. I had to force shutdown by long-pressing the power button, boot again to the same black screen and blinking cursor, cycle through a few ttys and only then was I able to get to the login screen.
There is something else going on there, please see this and follow accordingly and see the linked topics to it too:
Have you tried a different Kernel?
I’ve tried 5.6, 5.7 and 5.8. they all do the same.
Very strange indeed, could you check your logs after a successful boot if there is any indication of what causes it?
Run bmenu
if it’s already installed, press D
in the main menu and check options 5,6,7
Yeah… well… I wouldn’t know how to read those logs. All that programmer talk means nothing to me. I have just started using Linux this spring, after having been on Windows exclussively for the last 20 years.
We can try to take a look on the logs if you post them here . Let’s start with:
journalctl -p 3 -x
sudo dmesg -T --level=err,warn
And these only after a succesful reboot?
That shouldn’t really matter as the logs would also show us previous boots. Of course if you can somehow copy/paste from tty after a failed boot it would be even better.
How do you even copy from tty?
But thing is, when it fails to boot I usually can’t even enter tty, it just freezes on a black screen even the cursor doesn’t blink. I can only long press power to force a shutdown.
Via remote connection for example, but we should be fine if you post the logs from a graphical session. May i ask if this problem does only happen on Manjaro for you, or did you also try other distro’s?