AMD Ryzen AI 9 HX 370 Drivers

dCipher · Aug 10, 2025

Hi

I have been unable to get Proxmox 9 to properly use the 890M gpu in my AMD Ryzen AI 9 HX370. The maximum resolution I can get is 1920x1080 but my monitor is 2560x1440. Does anyone know where to get the appropriate Linux drivers for this chip?

Thanks,
Simon

dCipher · Aug 16, 2025

Hi

I reinstalled the Windows virtio drivers and then was able to select virtGPU for the display and everything is working correctly as I can select many more resolution options.

Simon

TonyHawkins · Aug 17, 2025

Hi. Were you able to pass your gpu to the virtual machine? I have exactly the same config: AMD Ryzen AI 9 Radeon 890 m. But when I try to make a gpu passthrough in Windows 11, I get error 43 in the device manager. Can you please explain how to fix this?

dCipher · Aug 18, 2025

TonyHawkins said:
Hi. Were you able to pass your gpu to the virtual machine? I have exactly the same config: AMD Ryzen AI 9 Radeon 890 m. But when I try to make a gpu passthrough in Windows 11, I get error 43 in the device manager. Can you please explain how to fix this?

Hi

When you say "to make a gpu passthrough in Windows 11" are you referring to and external gpu because I do not have an external gpu I am just using the built in 890M?

Simon

uzumo · Aug 18, 2025

Here is what those who achieved their objectives have implemented

Please include the results for future viewers of the forum.

https://t.co/Ft08TR9nYF

https://github.com/isc30/ryzen-gpu-passthrough-proxmox

DerekG · Aug 31, 2025

Just in case people reach this threat trying to overcome the Windows error 43 with the AMD iGPU, I have just completed the setup and got the 780m working correctly, including overcoming the AMD reset bug, so now I can start / stop the VM correctly from the host and from within the Guest VM.
I followed this really extensive & detailed guide to get the system working, and the only aspect I have not tested is getting the HDMI video output (as I am using RDP)

This is the guide I followed and note: 2 separate workarounds required for the AMDResetBug Fix.
https://github.com/xiongyw/docs/blo...rum-um880pro/pve-8.4-1_minisforum-um880pro.md

Kelebro · Oct 8, 2025

Hi!
Could you, please, help me?
I have Minisforum N5 Pro with the same AMD Ryzen AI 9 HX370 and 890M, but i cant passthrough 890M to my Windows 10 VM properly.
I tried many guides include https://github.com/isc30/ryzen-gpu-passthrough-proxmox but unfortunately iGPU doesn't work in Windows 10 VM.

First of all here is output of lspci -k -s c6:00 :

Code:

root@n5:/sys/kernel/iommu_groups/27/devices# lspci -k -s c6:00
c6:00.0 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] Strix [Radeon 880M / 890M] (rev d1)
        Subsystem: Device 1f4c:b020
        Kernel driver in use: vfio-pci
        Kernel modules: amdgpu
c6:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Radeon High Definition Audio Controller [Rembrandt/Strix]
        Subsystem: Device 1f4c:b020
        Kernel driver in use: vfio-pci
        Kernel modules: snd_hda_intel
c6:00.2 Encryption controller: Advanced Micro Devices, Inc. [AMD] Strix/Krackan/Strix Halo CCP/ASP
        Subsystem: Advanced Micro Devices, Inc. [AMD] Strix/Krackan/Strix Halo CCP/ASP
        Kernel driver in use: vfio-pci
        Kernel modules: ccp
c6:00.4 USB controller: Advanced Micro Devices, Inc. [AMD] Device 151e
        Subsystem: Advanced Micro Devices, Inc. [AMD] Device 15b9
        Kernel driver in use: vfio-pci
        Kernel modules: xhci_pci
c6:00.6 Audio device: Advanced Micro Devices, Inc. [AMD] Family 17h/19h/1ah HD Audio Controller
        DeviceName: Realtek ALC245
        Subsystem: Advanced Micro Devices, Inc. [AMD] Device d808
        Kernel driver in use: vfio-pci
        Kernel modules: snd_hda_intel
c6:00.7 Signal processing controller: Advanced Micro Devices, Inc. [AMD] Sensor Fusion Hub
        Subsystem: Advanced Micro Devices, Inc. [AMD] Sensor Fusion Hub
        Kernel driver in use: vfio-pci
        Kernel modules: amd_sfh

So, my Display controller and Audio device uses vfio-pci on pve host.

Secondly, here is my VM config:

Code:

agent: 1
args: -cpu host,hv_passthrough,level=35,-hypervisor,+svm
balloon: 16384
bios: ovmf
boot: order=scsi0;ide2;ide0;net0
cores: 12
cpu: host
efidisk0: local-lvm:vm-103-disk-0,efitype=4m,pre-enrolled-keys=1,size=4M
hostpci0: 0000:c6:00.0,pcie=1,romfile=vbios_7xxx.bin
hostpci1: 0000:c6:00.1,pcie=1,rombar=0,romfile=AMDGopDriver_7950x.rom
ide0: local:iso/virtio-win.iso,media=cdrom,size=771138K
ide2: local:iso/ru-ru_windows_10_business_editions_version_22h2_updated_sep_2025_x64_dvd_8ff5dabe.iso,media=cdrom,size=6498902K
machine: pc-q35-10.0+pve1
memory: 49152
meta: creation-qemu=10.0.2,ctime=1758824270
name: Windows
net0: virtio=BC:24:11:CB:7A:21,bridge=vmbr0,firewall=1
numa: 0
onboot: 1
ostype: win10
scsi0: local-lvm:vm-103-disk-1,aio=native,backup=0,discard=on,iothread=1,size=1000G
scsihw: virtio-scsi-single
smbios1: uuid=ea7113a3-af66-4143-9b66-530a5b6184a6
sockets: 1
vga: none
vmgenid: 3f66f763-d063-4726-9865-09d4e0df651d

File vbios_7xxx.bin i did with script from readme of https://github.com/isc30/ryzen-gpu-passthrough-proxmox/tree/main
There is no BIOS file on minisforum official site for now, so file AMDGopDriver_7950x.rom i got from repo as is.

On start of VM i get warning message:

Code:

kvm: -device vfio-pci,host=0000:c6:00.0,id=hostpci0,bus=ich9-pcie-port-1,addr=0x0,romfile=/usr/share/kvm/vbios_7xxx.bin: Failed to mmap 0000:c6:00.0 BAR 0. Performance may be slow
TASK OK

it can be fixed with

Code:

echo 1 > /sys/bus/pci/devices/0000\:c6\:00.0/remove
echo 1 > /sys/bus/pci/rescan

But i don't know is it critical for my case.

In VM i can see AMD Radeon(TM) 890M Graphics in devices list and i can install drivers.
But after reboot VM i am getting error with code 43 and there is no GPU in Task manager.

I tried many files of Audio roms, nothing was changed, tried don't pass audio to VM, tried couple of args parameters in VM config.

I rebooted host machine with pve and autostart of windows machine, there was no successful iGPU passthrought, so i think this is not RadeonResetBug

What am doing wrong?

uzumo · Oct 8, 2025

Does the screen display after installing the driver during the first startup?

If it does display, I think you can use a hookscript to reset/rescan it each time it starts.

Also, I saw that 0000:c6:00.6 also needs to be passed through.

I think you can create the gop from the BIOS of another equivalent CPU, but if it works, it's fine to leave it as is.

*I've heard that phenomenon is occurring on the Minis Forum N5 Pro.

adminsblog · Oct 8, 2025

Hello.
I was having the same problem.
Some people have been able to pass through the N5 Pro with simple settings,
but I was able to do so successfully with the following settings.
Shutting down and restarting the computer also worked without any problems.
https://wiki.adminsblog.com/Knowledge/PC/minisforum_n5_pro/iGPU_Passthrough（成功編）
I hope this helps a little.

Kelebro · Oct 8, 2025

uzumo said:
Does the screen display after installing the driver during the first startup?

Thank you for your reply!

Not sure that i understand properly.
On first startup of VM (or VM + pve host together) there is always code 43 and there is no GPU in task manager, so i think GPU don't work in VM.
If i disable and enable device AMD Radeon(TM) 890M Graphics in Windows code 43 error is gone, but there is no changes in task manager.

If you ask about physical screen through HDMI, it wasn't show anything in all tries

uzumo said:
Also, I saw that 0000:c6:00.6 also needs to be passed through.

I have a problem to add this device.
If i do that, i have error on start:

Code:

kvm: vfio: Cannot reset device 0000:c6:00.6, depends on group 24 which is not owned.

So, i think i need to add 0000:c6:00.2

Code:

root@n5:~# find /sys/kernel/iommu_groups/ -type l | grep -i 24
/sys/kernel/iommu_groups/24/devices/0000:c6:00.2

After that i got:

Code:

kvm: vfio: Cannot reset device 0000:c6:00.6, depends on group 27 which is not owned.

And added:

Code:

root@n5:~# find /sys/kernel/iommu_groups/ -type l | grep -i 27
/sys/kernel/iommu_groups/27/devices/0000:c6:00.7

And after that i got error:

Code:

root@n5:~# qm stop 103 && qm start 103
kvm: -device vfio-pci,host=0000:c6:00.0,id=hostpci0,bus=ich9-pcie-port-1,addr=0x0,romfile=/usr/share/kvm/vbios_7xxx.bin: Failed to mmap 0000:c6:00.0 BAR 0. Performance may be slow
kvm: -device vfio-pci,host=0000:c6:00.7,id=hostpci5,bus=ich9-pcie-port-6,addr=0x0,rombar=0: vfio 0000:c6:00.7: failed to setup container for group 27: Failed to set group container: Invalid argument
start failed: QEMU exited with code 1

uzumo said:
I think you can create the gop from the BIOS of another equivalent CPU, but if it works, it's fine to leave it as is.

I tried to do this from Minisforum AI X1 Pro BIOS file, but there was nothing to extract.
How can i check that it works?
There is no error on this device in Windows and there is no errors in VM start log.

adminsblog said:
Hello.
I was having the same problem.
Some people have been able to pass through the N5 Pro with simple settings,
but I was able to do so successfully with the following settings.
Shutting down and restarting the computer also worked without any problems.
https://wiki.adminsblog.com/Knowledge/PC/minisforum_n5_pro/iGPU_Passthrough（成功編）
I hope this helps a little.

Thank you!
Didn't find this guide before, i'll try to do this and reply.

Kelebro · Oct 8, 2025

adminsblog said:
Hello.
I was having the same problem.
Some people have been able to pass through the N5 Pro with simple settings,
but I was able to do so successfully with the following settings.
Shutting down and restarting the computer also worked without any problems.
https://wiki.adminsblog.com/Knowledge/PC/minisforum_n5_pro/iGPU_Passthrough（成功編）
I hope this helps a little.

It worked, thank you!
Changed GRUB settings to yours:

Code:

GRUB_CMDLINE_LINUX_DEFAULT="quiet iommu=pt initcall_blacklist=sysfb_init"

Added drivers to blacklist:

Code:

blacklist amdgpu
blacklist radeon
blacklist snd_sof_amd_acp63
blacklist snd_sof_amd_vangogh
blacklist snd_sof_amd_rembrandt
blacklist snd_sof_amd_renoir
blacklist snd_sof_amd_acp
blacklist snd_sof_pci
blacklist snd_sof_xtensa_dsp
blacklist snd_sof
blacklist snd_sof_utils
blacklist snd_soc_core
blacklist snd_hda_intel
blacklist snd_compress
blacklist snd_intel_dspcfg
blacklist snd_pcm_dmaengine
blacklist snd_intel_sdw_acpi
blacklist snd_pci_ps
blacklist snd_rpl_pci_acp6x
blacklist snd_hda_codec
blacklist snd_acp_pci
blacklist snd_acp_legacy_common
blacklist snd_hda_core
blacklist snd_pci_acp6x
blacklist snd_hwdep
blacklist snd_pci_acp5x
blacklist snd_pcm
blacklist snd_rn_pci_acp3x
blacklist snd_timer
blacklist snd_acp_config
blacklist snd_soc_acpi
blacklist snd_pci_acp3x

added device id to vfio-pci

Code:

options vfio-pci ids=1002:150e,1002:1640,1022:15e3

and used your rom files.

Not sure what exactly help, but that was i done.

For now, after reboot the host i have my iGPU into VM and can see display through HDMI.

There is something wrong with returning 0000:c6:00.0 to host, so pre-stop of hookscript is stucking, i will try to investigate.

Kelebro · Oct 8, 2025

After modification hook script it works perfectly:

Bash:

#!/usr/bin/bash

# ref: https://gist.github.com/kiler129/215e2c8de853209ca429ad5ed40ce128

set -e -o errexit -o pipefail -o nounset


# Do not modify these variables (set by Proxmox when calling the script)
vmId="$1"
runPhase="$2"
echo ">>> Entering $runPhase on VM=$vmId" > /dev/kmsg

case "$runPhase" in
    pre-start)
      echo "### rescan pcie bus..." > /dev/kmsg
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.0/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.1/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.6/remove || true
      echo 1 > /sys/bus/pci/rescan
      sleep 2
      ;;
    post-start)
      ;;
    pre-stop)
      echo "### remove gpu/snd devices..." > /dev/kmsg
      GFX_PATH=$(dirname $(grep -il "PCI_ID=1002:150e" /sys/bus/pci/devices/*/uevent))
      SND_PATH=$(dirname $(grep -il "PCI_ID=1002:1640" /sys/bus/pci/devices/*/uevent))
      echo 1 > ${GFX_PATH}/remove || true
      sleep 1
      echo 1 > ${SND_PATH}/remove || true
      sleep 1
      ;;
    post-stop)
      echo "### remove pcie bus..." > /dev/kmsg
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.0/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.1/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.6/remove || true
      sleep 2
      echo "### rescan pcie bus..." > /dev/kmsg
      echo 1 > /sys/bus/pci/rescan
      sleep 2
      ;;
    *)
      echo "### Unknown run phase \"$runPhase\"!" > /dev/kmsg
      ;;

But when VM works there are so much errors like this:

Code:

Oct 08 20:13:00 n5 kernel: amd_iommu_report_page_fault: 6175 callbacks suppressed
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030ff000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030c8000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030c9000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030ca000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030cb000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030cc000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030cd000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec58000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec59000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec5a000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: amd_iommu_report_page_fault: 38265 callbacks suppressed
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec50000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec51000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec52000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec53000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec54000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec55000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec56000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec57000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec58000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeaa44000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: amd_iommu_report_page_fault: 50036 callbacks suppressed
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab30000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab31000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab32000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab33000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab34000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab35000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab36000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab37000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab38000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeaa44000 flags=0x0000]
Oct 08 20:13:15 n5 kernel: amd_iommu_report_page_fault: 51134 callbacks suppressed

Maybe someone knows how to avoid it?

I think that because not all devices of 0000:c6:00 was passed to VM, but regarding to error i can't do it

Code:

kvm: -device vfio-pci,host=0000:c6:00.7,id=hostpci5,bus=ich9-pcie-port-6,addr=0x0,rombar=0: vfio 0000:c6:00.7: failed to setup container for group 27: Failed to set group container: Invalid argument

adminsblog · Oct 9, 2025

Kelebro said:

After modification hook script it works perfectly:

Bash:

#!/usr/bin/bash

# ref: https://gist.github.com/kiler129/215e2c8de853209ca429ad5ed40ce128

set -e -o errexit -o pipefail -o nounset


# Do not modify these variables (set by Proxmox when calling the script)
vmId="$1"
runPhase="$2"
echo ">>> Entering $runPhase on VM=$vmId" > /dev/kmsg

case "$runPhase" in
    pre-start)
      echo "### rescan pcie bus..." > /dev/kmsg
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.0/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.1/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.6/remove || true
      echo 1 > /sys/bus/pci/rescan
      sleep 2
      ;;
    post-start)
      ;;
    pre-stop)
      echo "### remove gpu/snd devices..." > /dev/kmsg
      GFX_PATH=$(dirname $(grep -il "PCI_ID=1002:150e" /sys/bus/pci/devices/*/uevent))
      SND_PATH=$(dirname $(grep -il "PCI_ID=1002:1640" /sys/bus/pci/devices/*/uevent))
      echo 1 > ${GFX_PATH}/remove || true
      sleep 1
      echo 1 > ${SND_PATH}/remove || true
      sleep 1
      ;;
    post-stop)
      echo "### remove pcie bus..." > /dev/kmsg
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.0/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.1/remove || true
      echo 1 > /sys/bus/pci/devices/0000\:c6\:00.6/remove || true
      sleep 2
      echo "### rescan pcie bus..." > /dev/kmsg
      echo 1 > /sys/bus/pci/rescan
      sleep 2
      ;;
    *)
      echo "### Unknown run phase \"$runPhase\"!" > /dev/kmsg
      ;;

But when VM works there are so much errors like this:

Code:

Oct 08 20:13:00 n5 kernel: amd_iommu_report_page_fault: 6175 callbacks suppressed
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030ff000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030c8000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030c9000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030ca000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030cb000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030cc000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0x1030cd000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec58000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec59000 flags=0x0000]
Oct 08 20:13:00 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec5a000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: amd_iommu_report_page_fault: 38265 callbacks suppressed
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec50000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec51000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec52000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec53000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec54000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec55000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec56000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec57000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeec58000 flags=0x0000]
Oct 08 20:13:05 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeaa44000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: amd_iommu_report_page_fault: 50036 callbacks suppressed
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab30000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab31000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab32000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab33000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab34000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab35000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab36000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab37000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab38000 flags=0x0000]
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeaa44000 flags=0x0000]
Oct 08 20:13:15 n5 kernel: amd_iommu_report_page_fault: 51134 callbacks suppressed

Maybe someone knows how to avoid it?

I think that because not all devices of 0000:c6:00 was passed to VM, but regarding to error i can't do it

Code:

kvm: -device vfio-pci,host=0000:c6:00.7,id=hostpci5,bus=ich9-pcie-port-6,addr=0x0,rombar=0: vfio 0000:c6:00.7: failed to setup container for group 27: Failed to set group container: Invalid argument

I think only .00, .01, and .06 need to be passed through, so why is .7 passed through?
Here are my Proxmox settings.

IO_PAGE_FAULT was the last error I saw. The link is to my settings, so if it's missing, it might be in the BIOS.
Are the settings on the iGPU configuration page different?

https://wiki.adminsblog.com/Knowledge/PC/minisforum_n5_pro/iGPU_Passthrough_BIOS

Kelebro · Oct 9, 2025

adminsblog said:
I think only .00, .01, and .06 need to be passed through, so why is .7 passed through?
Here are my Proxmox settings.
View attachment 91484
IO_PAGE_FAULT was the last error I saw. The link is to my settings, so if it's missing, it might be in the BIOS.
Are the settings on the iGPU configuration page different?
View attachment 91485
https://wiki.adminsblog.com/Knowledge/PC/minisforum_n5_pro/iGPU_Passthrough_BIOS

I searched threads and there was written that error:
Oct 08 20:13:10 n5 kernel: vfio-pci 0000:c6:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0002 address=0xbeab32000 flags=0x0000]

can be because of splitting groups of pci devices, so i tried to pass throught all 0000:c6:0 and faced this error

About BIOS:

SVM Lock and SVM Enable was on Auto
iGPU was on UMA_SPECIFIED

After setting same as yours error is gone thank you very much!

dockyr · Nov 10, 2025

In my case afteur "blacklist + modules + grub options + adding pci BIN + hook script"

its working on my N5 pro but:
* If i shutdown my VM from proxmox et start it again --> No Problem
* If i shutdown or reboot from VM --> Error 43 again and the only option to solve the issue is to reboot proxmox host again.

When i shutdown VM from windows inside, some errors on restart from proxmox:
START:
swtpm_setup: Not overwriting existing state file.
kvm: vfio: Cannot reset device 0000:c8:00.6, depends on group 28 which is not owned.
kvm: vfio: Cannot reset device 0000:c8:00.6, depends on group 28 which is not owned.
TASK OK

jor0 · Nov 19, 2025

adminsblog said:
I think only .00, .01, and .06 need to be passed through, so why is .7 passed through?
Here are my Proxmox settings.
View attachment 91484
IO_PAGE_FAULT was the last error I saw. The link is to my settings, so if it's missing, it might be in the BIOS.
Are the settings on the iGPU configuration page different?
View attachment 91485
https://wiki.adminsblog.com/Knowledge/PC/minisforum_n5_pro/iGPU_Passthrough_BIOS

Hey, looks like the wiki site is down not sure if that's just temporary. Anyway you can post the guide here? I'm trying to passthrough the IGPU in my N5 Pro as well.

Search

Search

AMD Ryzen AI 9 HX 370 Drivers

dCipher

Member

dCipher

Member

TonyHawkins

New Member

dCipher

Member

uzumo

Active Member

DerekG

Well-Known Member

Kelebro

New Member

uzumo

Active Member

adminsblog

New Member

Kelebro

New Member

Kelebro

New Member

Kelebro

New Member

adminsblog

New Member

Kelebro

New Member

dockyr

New Member

jor0

New Member

We value your privacy