You are not logged in.
Pages: 1
Whenever I run startx, the terminal shows the default startx output that appears whenever I use the command, but the GUI doesn’t show up. To some extent, I can tell that xorg is somewhat working because I am still able to use my keyboard shortcuts (I have keyboard shortcuts set up in my i3 config that allow me to exit I3 or reboot my laptop and they still work despite nothing showing up). The only problem is that i still see the terminal on the screen. How should I go about debugging this?
Last edited by umi (2022-09-23 00:29:25)
Offline
Start by posting the xorg log which should be under ~/.local/share/xorg as well as the content of your ~/.xinitrc. Also describe your graphics card / driver. What is the GPU? What (if any) xorg drivers have you installed?
"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman
Offline
I don’t know how to post logs that I get from the terminal. If it helps, I could post pictures of it. As for the my GPU, it is a RTX3050ti and I have nvidia installed.
Offline
I don’t know how to post logs that I get from the terminal.
See the examples from pastebin to post files from the terminal.
Offline
Here is my Xorg.0.log:
https://0x0.st/oVC7.txt
And here is my .xinitrc:
https://0x0.st/-eNp.sh
Offline
The system has two GPUs a GA107M [GeForce RTX 3050 Ti Mobile] and and an integrated TigerLake-LP GT2 [Iris Xe Graphics] and it appears to me that the display is connected to the integrated GPU?
Offline
Is that something that can be changed? I honestly had no clue that it had another integrated GPU, I’m just reading specs off my laptop.
Offline
See whether you can disable the IGP in the BIOS, but you're running on the intel chip where the internal display is connected - the nvidia chip isn't even considered and probably fails because of IBT, https://wiki.archlinux.org/title/NVIDIA#Installation
Disable that, reboot and see what happens.
If the issue remains, please post an updated xorg log and a complete system journal (sudo journalctl -b | curl -F 'file=@-' 0x0.st)
Offline
How can I tell if I disabled it? I took a few steps but I’m not entirely sure if I did it correctly. startx still doesn’t work.
Here is my new Xorg.0.log:
https://0x0.st/oWs3.txt
Here is my system journal:
https://0x0.st/oWsE.txt
Offline
If you disable the IGP in the BIOS, it won't show up in lspci - this is however neither required to deal w/ the situation nor necessarily a good idea.
Presently, both GPUs show up in your xorg log and /etc/X11/xorg.conf enforces the use of the nvidia driver but it struggles with
(EE) NVIDIA: Failed to load module "glxserver_nvidia" (module does not exist, 0)because /usr/share/X11/xorg.conf.d/10-nvidia-drm-outputclass.conf doesn't apply.
1. remove that xorg.conf
2. get rid of bumblebee
3. see https://wiki.archlinux.org/title/NVIDIA_Optimus - decide how you want to use the HW and explain why it's not PRIME.
Edit:
Presently the nvidia kernel modules are only loaded after your session starts…
Sep 22 17:41:28 umi systemd[1]: Started Session 1 of User umi.
Sep 22 17:41:29 umi kernel: nvidia: module license 'NVIDIA' taints kernel.Last edited by seth (2022-09-22 22:13:42)
Offline
I’m not sure I understand your 3rd point.
Offline
Did you read the wiki within the last 8 minutes?
There're various ways to use an optimus system, the default and typically sane choice is prime offloading (ie. run on the IGP and invoke the GPU on demand, eg. for a game)
Offline
Yes, that was originally how I was using my system. It only changed after I started tampering with my system because of this post. I was using I3 just fine yesterday, but something happened when I rebooted my laptop today.
Offline
I finally got it to work, thanks everyone! I did as seth instructed and removed bumblebee and deleted my xorg.conf. After that, startx would instead show me a black screen, so I did a bit of research and found this . I followed their solution and I added
xrandr --setprovideroutputsource modesetting NVIDIA-0
xrandr --autoto the end of my .xinitrc file.
Offline
That would suggest that for some reason you're currently running the server on the nvidia chip and use the intel one as VGA hub only, ie. kinda this scenario: https://wiki.archlinux.org/title/PRIME# … rimary_GPU
Offline
Pages: 1