I’m just start using arch for a while so i’m pretty noob in everything.
Currently using a laptop with amd igpu and nvidia 1650.
Since i finish download driver for nvidia, my screen after login 80% went on black screen with nothing even cursor for a while and auto comeback to login again and work properly, just wayland got this bug.
Did try someway like setting my nvidia drm and not working, boot properly for once when i reinstall nvidia and the bug just occur everytime else.
kwin 6.7.4-5
7.1.8-arch1-3
Thanks for reading all this, hope anyone can give me some clue for this.
likely a driver issue (either the wrong one for the GPU or it was installed incorrectly).
i would back out whatever driver you installed and stick to the kernel support until you figure it out.
according to the nvidia website you should be using the 595 or 580 driver
NVIDIA-SMI 580.178.04 Driver Version: 580.178.04 CUDA Version: 13.0
Tks but i dont sense any different after that
how did you install it?
using apt?
did you uninstall the other one?
sudo apt update
sudo apt purge *nvidia*
sudo apt autoremove
reboot
did you make any changes to grub to accommodate the nvidia drivers?
I used this package aur .archlinux.org/packages/nvidia-580xx-dkms
Quite know it done cuz some orther probs went good but the boot problem still as the same
oh, arch… using the aur packages.
as the old joke goes…
well there’s your problem.
unless you want to become an expert at packaging your own drivers, then i would suggest a different distro.
something like kubuntu where the driver install is point and click.
So i suggest those aur packages needed to be do orther ways so that it may work properly?
Don’t think it is necessarily a nivida driver issue.
If that is set up according to the archwiki, keywords nvidia_drm/modeset/fbdev and stuff, even if the wiki says that should be enabled by default if the nvidia-utils package is installed, checking and/or enabling it manually also should not hurt. With 580 there is even a NVreg_EnableGpuFirmware=0 module parameter that (maybe) has to be set/used.
That aside, why I think it is not necessarily on that driver, normally at that stage the igpu should be prioritised and used, the dedicated GPU only as a “on demand prime-run” thing as it needs more (battery) power.
At least that is usually how it should work on Laptops for power management reasons (longer battery lifetime).
But I don’t have a amd igpu, mine is a intel igpu, a nvidia 4070M and I use SDDM instead of the Plasma login manager.
