Adding a second GPU causes VM not to boot

[droctakan]

 

I'm using unraid on a dual Titan X (Pascal) machine, and have gotten a windows 11 VM up and running with one of the GPUs successfully. That involved

• unraid is booting in UEFI mode (and I can't quite get the bios to boot in legacy mode, but this works now anyway)
• I added efifb=off to the unraid system loader to avoid having it grab the GPU
• I don't have any vbios set in the VM as I understand that's no longer necessary

 

I then added a second GPU and the nvidia audio device, set them up as multifunction devices on the same slot in the xml, and so far the VM won't boot in that configuration. One CPU gets pegged to 100% and nothing is displayed on the monitor connected to GPU1.

 

I've found a lot of resources on GPU passthrough in general, but never seen anything specific to this kind of issue. Any ideas?

 

full XML below

 

<?xml version='1.0' encoding='UTF-8'?>
<domain type='kvm'>
  <name>Windows 11</name>
  <uuid>908eb4a1-7f40-3056-ccf3-bb41c1563575</uuid>
  <metadata>
    <vmtemplate xmlns="unraid" name="Windows 11" icon="windows11.png" os="windowstpm"/>
  </metadata>
  <memory unit='KiB'>102760448</memory>
  <currentMemory unit='KiB'>102760448</currentMemory>
  <memoryBacking>
    <nosharepages/>
  </memoryBacking>
  <vcpu placement='static'>10</vcpu>
  <cputune>
    <vcpupin vcpu='0' cpuset='1'/>
    <vcpupin vcpu='1' cpuset='7'/>
    <vcpupin vcpu='2' cpuset='2'/>
    <vcpupin vcpu='3' cpuset='8'/>
    <vcpupin vcpu='4' cpuset='3'/>
    <vcpupin vcpu='5' cpuset='9'/>
    <vcpupin vcpu='6' cpuset='4'/>
    <vcpupin vcpu='7' cpuset='10'/>
    <vcpupin vcpu='8' cpuset='5'/>
    <vcpupin vcpu='9' cpuset='11'/>
  </cputune>
  <os>
    <type arch='x86_64' machine='pc-i440fx-7.1'>hvm</type>
    <loader readonly='yes' type='pflash'>/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi-tpm.fd</loader>
    <nvram>/etc/libvirt/qemu/nvram/908eb4a1-7f40-3056-ccf3-bb41c1563575_VARS-pure-efi-tpm.fd</nvram>
  </os>
  <features>
    <acpi/>
    <apic/>
  </features>
  <cpu mode='host-passthrough' check='none' migratable='on'>
    <topology sockets='1' dies='1' cores='5' threads='2'/>
    <cache mode='passthrough'/>
  </cpu>
  <clock offset='localtime'>
    <timer name='rtc' tickpolicy='catchup'/>
    <timer name='pit' tickpolicy='delay'/>
    <timer name='hpet' present='no'/>
  </clock>
  <on_poweroff>destroy</on_poweroff>
  <on_reboot>restart</on_reboot>
  <on_crash>restart</on_crash>
  <devices>
    <emulator>/usr/local/sbin/qemu</emulator>
    <disk type='file' device='disk'>
      <driver name='qemu' type='raw' cache='writeback'/>
      <source file='/mnt/user/domains/Windows 11/vdisk1.img'/>
      <target dev='hdc' bus='virtio'/>
      <boot order='1'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'/>
    </disk>
    <controller type='usb' index='0' model='qemu-xhci' ports='15'>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x07' function='0x0'/>
    </controller>
    <controller type='virtio-serial' index='0'>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
    </controller>
    <controller type='pci' index='0' model='pci-root'/>
    <interface type='bridge'>
      <mac address='52:54:00:1e:24:33'/>
      <source bridge='br0'/>
      <model type='virtio-net'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
    </interface>
    <serial type='pty'>
      <target type='isa-serial' port='0'>
        <model name='isa-serial'/>
      </target>
    </serial>
    <console type='pty'>
      <target type='serial' port='0'/>
    </console>
    <channel type='unix'>
      <target type='virtio' name='org.qemu.guest_agent.0'/>
      <address type='virtio-serial' controller='0' bus='0' port='1'/>
    </channel>
    <input type='tablet' bus='usb'>
      <address type='usb' bus='0' port='1'/>
    </input>
    <input type='mouse' bus='ps2'/>
    <input type='keyboard' bus='ps2'/>
    <tpm model='tpm-tis'>
      <backend type='emulator' version='2.0' persistent_state='yes'/>
    </tpm>
    <audio id='1' type='none'/>
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0' multifunction='on'/>
    </hostdev>
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x02' slot='0x00' function='0x0'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x1'/>
    </hostdev>
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x01' slot='0x00' function='0x1'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0' multifunction='on'/>
    </hostdev>
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x02' slot='0x00' function='0x1'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x1'/>
    </hostdev>
    <hostdev mode='subsystem' type='usb' managed='no'>
      <source startupPolicy='optional'>
        <vendor id='0x046d'/>
        <product id='0xc52b'/>
      </source>
      <address type='usb' bus='0' port='2'/>
    </hostdev>
    <hostdev mode='subsystem' type='usb' managed='no'>
      <source startupPolicy='optional'>
        <vendor id='0x0b05'/>
        <product id='0x180a'/>
      </source>
      <address type='usb' bus='0' port='3'/>
    </hostdev>
    <hostdev mode='subsystem' type='usb' managed='no'>
      <source startupPolicy='optional'>
        <vendor id='0x093a'/>
        <product id='0x2510'/>
      </source>
      <address type='usb' bus='0' port='4'/>
    </hostdev>
    <memballoon model='none'/>
  </devices>
</domain>

 

[hot22shot]

I assume that your Titan X has 12GB of vRAM each, so 24GB for both of them.

There's a software limit on how much memory the OVMF UEFI can address by default.

I bump on this limit with a RTX 3090 that has 24GB of RAM and had the same issue you had : black screen at boot with a CPU at 100%.

 

Anyway you can extend this limit by tweaking your XML, see this topic for guidance : 

 

 

Edited November 5, 2022 by hot22shot

[ghost82]

In addition to hot22shot suggestion, which I think it's necessary, otherwise you could get a code 12 error in windows, pay attention to the layout in the guest os; you can't have the audio of the gpu 2 in the same bus and slot of the video of gpu 1. Moreover addresses and multifunction are in the wrong place

So change with this:

    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0' multifunction='on'/>
    </hostdev>
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x02' slot='0x00' function='0x0'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0' multifunction='on'/>
    </hostdev>
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x01' slot='0x00' function='0x1'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x1'/>
    </hostdev>
    <hostdev mode='subsystem' type='pci' managed='yes'>
      <driver name='vfio'/>
      <source>
        <address domain='0x0000' bus='0x02' slot='0x00' function='0x1'/>
      </source>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x1'/>
    </hostdev>

 

Edited November 6, 2022 by ghost82