Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Nothing happens, libvirtd locks up #44

Open
FelicitusNeko opened this issue May 29, 2024 · 4 comments
Open

Nothing happens, libvirtd locks up #44

FelicitusNeko opened this issue May 29, 2024 · 4 comments

Comments

@FelicitusNeko
Copy link

Hi there,

I've gone through your guide to set up a VM. However, after all the passthrough setup is done, I go to start the VM, and nothing happens. Hooks don't run, and libvirtd locks up indefinitely until I force the computer to reboot.

Running Garuda Linux (Arch variant) host, Windows 11 guest (or trying to), systemd, NVIDIA RTX 3060.

@ccalhoun1999
Copy link

ccalhoun1999 commented Jun 4, 2024

Exact same issue here. Had the problem on AuroraOS and now have the exact same issue on EndeavorOS. Sending SIGKILL isn't enough to end the libvirtd process when it gets stuck.

@DragonSWDev
Copy link

Same issue here on Fedora 41 and NVIDIA RTX 3060 GPU. Nothing happens when VM is started and libvirtd process become stuck and can't be even killed. After attempting to start VM there is following message in dmesg:
NVRM: Attempting to remove device 0000:07:00.0 with non-zero usage count!

this message is shortly followed with this error in log:
virtqemud[4620]: End of file while reading data: Input/output error

I wonder if this is something related to RTX 3060 or other NVIDIA GPUs in general as some time ago I was able to successfully do single GPU passthrough with AMD GPU.

@ccalhoun1999 what GPU do you have?

@ccalhoun1999
Copy link

ccalhoun1999 commented Nov 16, 2024

@ccalhoun1999 what GPU do you have?

I have a 3070 so same generation. About two years ago I had it working. But something's changed and I can't get it working anymore.

@DragonSWDev
Copy link

DragonSWDev commented Nov 16, 2024

Ok I've managed to make it working for me. For me the issue was caused by invalid PCIe address in start and stop scripts. There are two lines in each script:
start.sh

virsh nodedev-detach pci_0000_01_00_0
virsh nodedev-detach pci_0000_01_00_1

stop.sh

virsh nodedev-reattach pci_0000_01_00_0
virsh nodedev-reattach pci_0000_01_00_1

In my case address was invalid and after I changed it to valid it started working. You can easily find this address in virtual machine configuration after you assign GPU to it.

However for some reason I can't start VM when I'm on the desktop (KDE in my case). When I start VM something happens as screen turns black but after a while screen shows message that there is no input and turns off and I'm unable to do anything and I need to force reset my PC. To start VM properly I need to log out and start VM from another TTY. This is the command to start VM from CLI:
virsh start vm_name

It is probably caused by #50

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants