You are not logged in.
Pages: 1
Hello,
I have a desktop PC with an Intel CPU and a NVIDIA GPU. In Linux, I am using the proprietary NVIDIA driver for best performance. However, I'm dual booting and the other OS does not support my NVIDIA card - but it does support the iGPU from my Intel CPU.
So I simply connected one input from my monitor to the iGPU and the other one to the discrete GPU. I've enabled automatic source detection for my monitor so that it jumps to the active channel. I was forced to configure the primary GPU to be the iGPU in BIOS so that the other OS recognizes the iGPU.
Now the problem resides with Linux. The NVIDIA driver is loaded correctly and the Xorg-session runs without problems. However, at boot time, the iGPU is enabled (since I configured it as the primary GPU in BIOS) and it does not get disabled when the NVIDIA driver is loaded. I.e. i'm seeing a black screen with that little "terminal-underscore-like-cursor" on the top left. I.e. the iGPU is still connected though the Xorg-session runs on the discrete GPU. I wouldn't care about this but unfortunately this prevents my monitor to automatically switch the channel and I always have to do that manually. Controversly, xrandr does only see one screen and only one port is connected - the one connected to the discrete GPU.
So what I would need is to disable the iGPU when booting Linux.
I tried the following without success:
-Blacklisting nouveau and i915 via /etc/modprobe.d and via GRUB
-Tried "nvidia-drm.modeset=1" kernel parameter (https://wiki.archlinux.org/index.php/NV … de_setting)
-https://wiki.archlinux.org/index.php/NVIDIA_Optimus#Use_NVIDIA_graphics_only
Thanks a lot for any help
EDIT: By the way I noticed also that the TTY-sessions (CTRL+ALT+3 or so) are all displayed using the iGPU
Last edited by 1700 (2020-06-05 14:03:36)
Offline
Pages: 1