[SOLVED] Configuring Proxmox for PCIE passthrough NVIDIA GeForce 9600

[pookieman]

Hiya

Before I get too into the weeds.. I've been following various guides how to pass through graphics cards to windows VM .. some success but still getting this Code error 43 in Windows Device Manager.

So before I go any further I wanted to check if anyone knew .. can I even use this card for video pass through?

GeForce 9600 GSO 512 MB

https://www.gpuzoo.com/GPU-NVIDIA/GeForce_9600_GSO_512.html

It's a little old .. but for my needs it's fine .. if it's not compatible then I'll give up. How can I tell?

Thanks in advance

 

[leesteken]

If you see the GPU in Windows, then at least it is being passed through and your IOMMU groups etc. are working fine.
Please show your VM configuration file (from the /etc/pve/qemu-server/ directory). Have you tried enabling the Primary GPU option?
Is this the only GPU in your system? Does the BIOS/POST-screen of your computer show on a monitor connected to this GPU? Have you tried dumping the GPU ROM and passing it?

 

[pookieman]

This is the config for the VM, primary GPU is enabled

root@homeserver:~# cat  /etc/pve/qemu-server/105.conf
bios: ovmf
boot: order=ide0;ide2;net0
cores: 4
efidisk0: local-zfs:vm-105-disk-1,size=1M
hostpci0: 0000:82:00,x-vga=1,pcie=1
ide0: local-zfs:vm-105-disk-0,size=50G
ide2: local:iso/Windows10.iso,media=cdrom
machine: pc-q35-6.0
memory: 8096
name: Gamer1
net0: e1000=4A:E5:BD:6F:94:29,bridge=vmbr0,firewall=1
numa: 0
ostype: win10
scsihw: virtio-scsi-pci
smbios1: uuid=19db7dc5-8c12-47f3-a2a2-6bcd713e77ed
sockets: 2
vga: virtio,memory=256
vmgenid: c6829e45-d06a-44f6-8317-02c37e9a4689

This is the only GPU in the system.

Does the BIOS/POST-screen of your computer show on a monitor connected to this GPU?

POST for the server? the proxmox instance?

Have you tried dumping the GPU ROM and passing it?

How would I do this? Is there a guide?

thanks

 

[pookieman]

This is what I see in proxmox for the PCI device

 

[pookieman]

On the proxmox loading screen I see this

 

[pookieman]

In the VM I see this

 

[leesteken]

Because it is the only GPU, you might need to add video=efifb:off video=vesafb:off to your kernel parameters. I don't know whether you use /etc/default/grub or /etc/kernel/cmdline. I found this suggestion here.

EDIT: Can you show lspci -nnvvks '82:00' (after a reboot before starting the VM)? Maybe you need a vfio_pci.ids= parameter as well.

 

[pookieman]

Here's the output

root@homeserver:~# lspci -nnvvks '82:00'
82:00.0 VGA compatible controller [0300]: NVIDIA Corporation G92 [GeForce 9600 GSO] [10de:0610] (rev a2) (prog-if 00 [VGA controller])
    Subsystem: XFX Pine Group Inc. G92 [GeForce 9600 GSO] [1682:239a]
    Physical Slot: 2-1
    Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
    Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
    Interrupt: pin A routed to IRQ 11
    NUMA node: 1
    IOMMU group: 76
    Region 0: Memory at fa000000 (32-bit, non-prefetchable) [disabled] [size=16M]
    Region 1: Memory at e0000000 (64-bit, prefetchable) [disabled] [size=256M]
    Region 3: Memory at f8000000 (64-bit, non-prefetchable) [disabled] [size=32M]
    Region 5: I/O ports at e000 [disabled] [size=128]
    Expansion ROM at fb000000 [disabled] [size=128K]
    Capabilities: [60] Power Management version 3
        Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
        Status: D3 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
    Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
        Address: 0000000000000000  Data: 0000
    Capabilities: [78] Express (v1) Endpoint, MSI 00
        DevCap:    MaxPayload 128 bytes, PhantFunc 0, Latency L0s <512ns, L1 <4us
            ExtTag+ AttnBtn- AttnInd- PwrInd- RBE+ FLReset- SlotPowerLimit 25.000W
        DevCtl:    CorrErr- NonFatalErr- FatalErr- UnsupReq-
            RlxdOrd- ExtTag+ PhantFunc- AuxPwr- NoSnoop+
            MaxPayload 128 bytes, MaxReadReq 512 bytes
        DevSta:    CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr- TransPend-
        LnkCap:    Port #0, Speed 2.5GT/s, Width x16, ASPM L0s L1, Exit Latency L0s <512ns, L1 <1us
            ClockPM- Surprise- LLActRep- BwNot- ASPMOptComp-
        LnkCtl:    ASPM Disabled; RCB 128 bytes, Disabled- CommClk+
            ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
        LnkSta:    Speed 2.5GT/s (ok), Width x16 (ok)
            TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
    Capabilities: [100 v1] Virtual Channel
        Caps:    LPEVC=0 RefClk=100ns PATEntryBits=1
        Arb:    Fixed- WRR32- WRR64- WRR128-
        Ctrl:    ArbSelect=Fixed
        Status:    InProgress-
        VC0:    Caps:    PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
            Arb:    Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256-
            Ctrl:    Enable+ ID=0 ArbSelect=Fixed TC/VC=01
            Status:    NegoPending- InProgress-
    Capabilities: [128 v1] Power Budgeting <?>
    Capabilities: [600 v1] Vendor Specific Information: ID=0001 Rev=1 Len=024 <?>
    Kernel driver in use: vfio-pci
    Kernel modules: nvidiafb, nouveau

 

[leesteken]

Is that the whole output? Does it not have an audio device?
As the driver in use is already vfio-pci (this is just after a system reboot before starting the VM, right?), you are already binding the GPU to vfio-pci early (so no change needed there).
Did adding video=efifb:off video=vesafb:off help in any way?

 

[pookieman]

Eeeek I've ran into another issue. I did update the /etc/kernel/cmdline then I rebooted .. now I'm stuck here

 

[pookieman]

wondering if I should start another thread .. I did try to import like this and reboot .. but I still get back to this same prompt

 

[pookieman]

 

[pookieman]

OK so I moved on from this by replacing the graphics card and fixed this issue here - https://forum.proxmox.com/threads/h...ol-import-n-not-found-error.96842/post-419456