Tesla k20xm PROBLEM (GPU passthrough) The device could not start. (Code 10) Not enough system resources to complete the API.

busoo530

New Member
Mar 1, 2024
1
0
1
Hi, unfortunately I have this problem with my Tesla K20Xm, I ran the passthrough on my HP DL380p Gen 8 server with the latest Proxmox 8.1.4 patches on board, once passed the guest device detects it and once the drivers are installed it gives me this problem appears on win 10 device manager:


[The device could not start. (Code 10)

Insufficient system resources to complete the API.],




I don't understand why this thing happens, I've tried everything, I attach the configuration of my vm:



agent: 1
balloon: 4096
boot: order=scsi0;net0
cores: 6
cpu: host,flags=+hv-evmcs;+aes
hostpci0: mapping=gpu-1-NVIDIA-k20xm,pcie=1
machine: pc-q35-8.1
memory: 8292
meta: creation-qemu=8.1.5,ctime=1708805356
name: gpu
net0: e1000=BC:24:11:88:EB:13,bridge=vmbr0,firewall=1
numa: 1
ostype: win10
parent: v1
scsi0: local-lvm:vm-7000-disk-0,cache=writeback,discard=on,iothread=1,size=48G,ssd=1
scsihw: virtio-scsi-single
smbios1: uuid=95f3c921-e1e2-4ca1-9c90-6912f5ddf357
sockets: 2
vmgenid: 719e7cbc-36cb-4c02-9c1b-971d57912fb8

[v1]
agent: 1
balloon: 4096
boot: order=scsi0;net0
cores: 6
cpu: host,flags=+hv-evmcs;+aes
hostpci0: mapping=gpu-1-NVIDIA-k20xm,pcie=1
machine: pc-q35-8.1
memory: 8292
meta: creation-qemu=8.1.5,ctime=1708805356
name: gpu
net0: e1000=BC:24:11:88:EB:13,bridge=vmbr0,firewall=1
numa: 1
ostype: win10
scsi0: local-lvm:vm-7000-disk-0,cache=writeback,discard=on,iothread=1,size=48G,ssd=1
scsihw: virtio-scsi-single
smbios1: uuid=95f3c921-e1e2-4ca1-9c90-6912f5ddf357
snaptime: 1709307113
sockets: 2
vmgenid: 719e7cbc-36cb-4c02-9c1b-971d57912fb8








this is my grub configuration:





GRUB_CMDLINE_LINUX_DEFAULT="quiet intel_iommu=on vfio-pci.ids=10de:1021,10de:097d iommu=pt nox2apic intremap=no_x2apic_optout video=efifb:off,vesafb:off,simplefb:off"


I added it here later vfio-pci.ids=10de:1021,10de:097d and also other modules I tried taking them off and putting them on.









this is the configuration of vfio.conf:



options vfio-pci ids=10de:1021,10de:097d disable_vga=1










these are the blacklisted modules:



blacklist nouveau
blacklist nvidiafb
blacklist nvidia*
blacklist i915







this is the configuration of "/etc/modprobe.d/iommu_unsafe_interrupts.conf":



options vfio_iommu_type1 allow_unsafe_interrupts=1





questa è la confiurazione di "nano /etc/modprobe.d/kvm.conf":



options kvm ignore_msrs=1 report_ignored_msrs=0


If you can help me I would be truly grateful

Good Day
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!