You are not logged in.
No log shows an extension of the ModulePath nor nvidia-drm.modeset=1 being used, so there's probably also no early KMS and the GPU is simply late.
https://wiki.archlinux.org/title/NVIDIA … de_setting
Semi-OT:
Section "InputDevice"
Identifier "Mouse0"
Driver "mouse"
EndSection
Section "InputDevice"
Identifier "Keyboard0"
Driver "kbd"
EndSectionThis is cruft, remove it, even iff you desire to maintain a static server layout.
Edit: the GPU still responds to nvidia-smi so I don't think there're another issue w/ the kernel module.
Last edited by seth (2021-08-31 14:43:18)
Offline
ExellentCoin wrote:Lone_Wolf wrote:Creating a config from scratch tends to be best method to get everything right.
Would this be good to start with?
I should rephrase that :
Configuring X is easiest when using autodetection and only adding *.conf files for stuff X autodetection gets wrong.
back to issue :
X still doesn't load nvidia, we may need to investigate what happens before X starts.post lspci -k output .
As root , full dmesg and SYSTEMD_LESS=FRXMK journalctl -b output .dmesg and journal are probably large, consider using a pastebin service
lspci -k:
https://pastebin.com/2M2U1Tbq
dmesg:
https://pastebin.com/B2iHhJvJ
journalctl -b:
https://pastebin.com/MiX1cn0L
Offline
Semi-OT:
Section "InputDevice" Identifier "Mouse0" Driver "mouse" EndSection Section "InputDevice" Identifier "Keyboard0" Driver "kbd" EndSectionThis is cruft, remove it, even iff you desire to maintain a static server layout.
when I don't add this, I have no keyboard or mouse input.
Offline
Something you can do if you really do not get the nvidia X modules to load without your custom config which should not be normal if you've indeed properly enabled early KMS for all participants is adding
Section "Files"
ModulePath "/usr/lib/nvidia/xorg"
ModulePath "/usr/lib/xorg/modules"
EndSectionto your xorg.conf
Offline
Something you can do if you really do not get the nvidia X modules to load without your custom config which should not be normal if you've indeed properly enabled early KMS for all participants is adding
Section "Files" ModulePath "/usr/lib/nvidia/xorg" ModulePath "/usr/lib/xorg/modules" EndSectionto your xorg.conf
It did't change anything weirdly enough. Is there an output I should post with it?
Offline
when I don't add this, I have no keyboard or mouse input.
[ 391.634] (WW) Hotplugging is on, devices using drivers 'kbd', 'mouse' or 'vmmouse' will be disabled.
[ 391.634] (WW) Disabling Keyboard0
[ 391.634] (WW) Disabling Mouse0
This is cruft
No log shows an extension of the ModulePath nor nvidia-drm.modeset=1 being used, so there's probably also no early KMS and the GPU is simply late.
https://wiki.archlinux.org/title/NVIDIA … de_setting
Offline
New xorg log. But also look at seths post to enable early KMS properly.
Offline
New xorg log. But also look at seths post to enable early KMS properly.
I think I did that. I added the modules to mkinitcpio.conf and recompiled it using mkinitcpio -p linux and rebooted. It did change the start resolution tho. This laptop is running with luks encryption and normally, the resolution is all fddd up, by doing what he told me this isn't the case anymore. I have normal prompts with normal input. Not just typing on feels anymore.
Offline
Updated xorg log and "glxinfo -B"?
Offline
[ 391.634] (WW) Hotplugging is on, devices using drivers 'kbd', 'mouse' or 'vmmouse' will be disabled.
[ 391.634] (WW) Disabling Keyboard0
[ 391.634] (WW) Disabling Mouse0
This is cruft
I removed them and indeed you were right, I am sorry. I have a working mouse and keyboard.
Offline
Updated xorg log and "glxinfo -B"?
Updated Xorg.0.log:
https://pastebin.com/TbzjziWb
glxinfo -B:
https://pastebin.com/Ms9F3Tph
Offline
You're running on the AMDGPU, but "prime-run glxinfo -B" should™ get you the nvidia chip (everything is properly loaded) and "prime-run superturboturkeypuncher3" should™ allow you to smash turkeys on the nvidia chip.
Offline
You're running on the AMDGPU, but "prime-run glxinfo -B" should™ get you the nvidia chip (everything is properly loaded) and "prime-run superturboturkeypuncher3" should™ allow you to smash turkeys on the nvidia chip.
I enabled prime (I think, just changed the device used by the screen). Now I have the external monitors, but not the internal one it just freezes until I close X.
glxinfo -B:
https://pastebin.com/TcZKsQ7g
And UNFORTUNATELY the superturboturkeypuncher3 isn't found and I have no idea how to install it ![]()
Offline
I enabled prime (I think, just changed the device used by the screen). Now I have the external monitors, but not the internal one it just freezes until I close X.
What exactly did you do?
https://wiki.archlinux.org/title/PRIME#Reverse_PRIME - but you'll have to use the modesetting driver (drop xf86-video-amdgpu)
I have no idea how to install it
Offline
I have no idea how to install it
https://doom.fandom.com/wiki/Super_Turb … _Puncher_3
I am propably blind, but I can't find it.
Offline
It's a running gag in Doom…
Offline
It's a running gag in Doom…
Yeah that I know of, but I thought that it was also some kind of PRIME tester like fire from mese-demos ![]()
Offline
I enabled prime (I think, just changed the device used by the screen). Now I have the external monitors, but not the internal one it just freezes until I close X.
What exactly did you do?
https://wiki.archlinux.org/title/PRIME#Reverse_PRIME - but you'll have to use the modesetting driver (drop xf86-video-amdgpu)
I don't know what you mean with drop amdgpu.
I now folowed the article and use their xorg.conf (see below) and used xrandr command (also see below), but nothing has changed. It does display all the outputs, but only enables the internal one. When I try to enable the external ones, I get the crtc error again.
xorg.conf:
Section "ServerLayout"
Identifier "layout"
Screen 0 "amdgpu"
Inactive "nvidia"
Option "AllowNVIDIAGPUScreens"
EndSection
Section "Device"
Identifier "nvidia"
Driver "nvidia"
EndSection
Section "Screen"
Identifier "nvidia"
Device "nvidia"
EndSection
Section "Device"
Identifier "amdgpu"
Driver "amdgpu"
BusID "PCI:7:0:0"
EndSection
Section "Screen"
Identifier "amdgpu"
Device "amdgpu"
EndSectionxrandr --listproviders and the other one:
Providers: number : 2
Provider 0: id: 0x55 cap: 0xf, Source Output, Sink Output, Source Offload, Sink Offload crtcs: 4 outputs: 2 associated providers: 1 name:Unknown AMD Radeon GPU @ pci:0000:07:00.0
Provider 1: id: 0x1fb cap: 0x2, Sink Output crtcs: 4 outputs: 4 associated providers: 1 name:NVIDIA-G0xrandr --setprovideroutputsource NVIDIA-G0 "Unknown AMD Radeon GPU @ pci:0000:07:00.0"Offline
Nahh… I've a flash version, though ![]()
Edit:
I don't know what you mean with drop amdgpu.
pacman -R xf86-video-amdgpuYou'll be using the modesetting driver for the amd chip (on top of the amdgpu kernel module) since amdgpu can't do reverse prime (as by the wiki)
Last edited by seth (2021-08-31 15:39:40)
Offline
I don't know what you mean with drop amdgpu.
pacman -R xf86-video-amdgpuYou'll be using the modesetting driver for the amd chip (on top of the amdgpu kernel module) since amdgpu can't do reverse prime (as by the wiki)
Again the crtc X error.
Offline
Because you still have a static config referencing amdgpu?
Offline
Because you still have a static config referencing amdgpu?
No the modesetting as the driver.
Offline
seth wrote:Because you still have a static config referencing amdgpu?
No the modesetting as the driver.
I can get the xorg log in an hour if you want.
Offline
Log and present config files, yes (that includes anything in /{etc,usr/share}/X11/xorg.conf.d )
Offline
Log and present config files, yes (that includes anything in /{etc,usr/share}/X11/xorg.conf.d )
Ok I am not home, but i can get them in an hour
Offline