[SOLVED] [HELP] GPU passthrough VM can't boot.

[Karol45]

Hello, I am migrating my GPU passthrough from 1 vm to another, I commented out hostpci in the former VM's config. I keep getting this error when I start my vm

WARN: no efidisk configured! Using temporary efivars disk.
error writing '1' to '/sys/bus/pci/devices/0000:01:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:01:00.0', but trying to continue as not all devices need a reset
TASK WARNINGS: 1

my vm's config

bios: ovmf
boot: order=scsi0;ide2;net0
cores: 3
cpu: host,hidden=1,flags=+pcid
args: -cpu 'host,+kvm_pv_unhalt,+kvm_pv_eoi,hv_vendor_id=proxmox,kvm=off'
hostpci0: 0000:01:00.0,pcie=1,x-vga=1
ide2: local:iso/debian-13.4.0-amd64-netinst.iso,media=cdrom,size=754M
machine: q35
memory: 6144
meta: creation-qemu=10.1.2,ctime=1777629160
name: docker
net0: virtio=BC:24:11:ED:C0:A8,bridge=vmbr0,firewall=1
numa: 0
onboot: 1
ostype: l26
scsi0: local-lvm:vm-102-disk-0,iothread=1,size=50G
scsihw: virtio-scsi-single
smbios1: uuid=47c3601c-77c8-42f3-83e3-009fe7cfd773
sockets: 1
vga: none
vmgenid: 8aaffa31-e385-4767-aafa-92d2b7be0755

my GPU is gtx 1050 ti 4GB.

 

[uzumo]

Why don't you try fixing this first?
It looks like it's saying, “No EFI disk.”
Since the BIOS is OVMF (UEFI), isn't this required?

WARN: no efidisk configured! Using temporary efivars disk.

The following is unrelated to the issue where the system won't start, and you would have likely seen it on your previous virtual machines as well, even when passthrough was working properly.

error writing '1' to '/sys/bus/pci/devices/0000:01:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:01:00.0', but trying to continue as not all devices need a reset

 

[Karol45]

Why don't you try fixing this first?
It looks like it's saying, “No EFI disk.”
Since the BIOS is OVMF (UEFI), isn't this required?

WARN: no efidisk configured! Using temporary efivars disk.

The following is unrelated to the issue where the system won't start, and you would have likely seen it on your previous virtual machines as well, even when passthrough was working properly.

error writing '1' to '/sys/bus/pci/devices/0000:01:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:01:00.0', but trying to continue as not all devices need a reset

Fixed this and removed my PCIE passthrough to check what's going on, and I ecountered
BdsDxe: failed to load Boot0003 "UEFI QEMU QEMU HARDDISK " from PciRoot(0x0)/Pci(0x1E,0x0)/Pci(0x4,0x0)/Pci(0x1,0x0)/Scsi(0x0,0x0): Not Found
(or something along that, didn't take a screenshot)
so I unchecked pre-enroll keys in efi disk but now I am getting
BdsDxe: failed to load Boot0002 "UEFI QEMU QEMU HARDDISK " from PciRoot(0x0)/Pci(0x1E,0x0)/Pci(0x4,0x0)/Pci(0x1,0x0)/Scsi(0x0,0x0): Not Found

 

[Karol45]

Fixed this and removed my PCIE passthrough to check what's going on, and I ecountered
BdsDxe: failed to load Boot0003 "UEFI QEMU QEMU HARDDISK " from PciRoot(0x0)/Pci(0x1E,0x0)/Pci(0x4,0x0)/Pci(0x1,0x0)/Scsi(0x0,0x0): Not Found
(or something along that, didn't take a screenshot)
so I unchecked pre-enroll keys in efi disk but now I am getting
BdsDxe: failed to load Boot0002 "UEFI QEMU QEMU HARDDISK " from PciRoot(0x0)/Pci(0x1E,0x0)/Pci(0x4,0x0)/Pci(0x1,0x0)/Scsi(0x0,0x0): Not Found

It seems to be an issue with debian being installed with legacy boot instead of uefi, as I can boot normally with SeaBIOS, I will be reinstalling the system, with a different distro as I wanted to move to RHEL based before.

If someone finds this in the future and doesn't want to reinstall debian, I found a guide to switch from legacy to UEFI boot mode: https://blog.getreu.net/projects/legacy-to-uefi-boot/