proxmox 上 win10 VM 中的 GPU 直通代码 12

proxmox 上 win10 VM 中的 GPU 直通代码 12

设备管理器中出现代码 12 错误。已经尝试了 538 数据中心特斯拉和 442 特斯拉桌面驱动程序。我无法打开 nvidia 控制中心,但第一次尝试时我得到了 EULA 提示,我接受了,然后就没有其他了。

设备管理器+gpu-z

dmesg | grep -e IOMMU

[    0.128824] DMAR: IOMMU enabled
[    0.332016] DMAR-IR: IOAPIC id 0 under DRHD base  0xfbffc000 IOMMU 0
[    0.332020] DMAR-IR: IOAPIC id 2 under DRHD base  0xfbffc000 IOMMU 0

dmesg | grep 重新映射

    [    0.332025] DMAR-IR: Queued invalidation will be enabled to support x2apic and Intr-remapping.
    [    0.332366] DMAR-IR: Enabled IRQ remapping in x2apic mode

pvesh get /nodes/{nodename}/hardware/pci --pci-class-blacklist ""
    class    │ device │ id           │ iommugroup │ vendor │ device_name
    0x030200 │ 0x1b38 │ 0000:04:00.0 │         27 │ 0x10de │ GP102GL [Tesla P40]

/etc/默认/grub

    GRUB_CMDLINE_LINUX_DEFAULT="quiet intel_iommu=on intel_iommu=pt"

试过:

    pci=realloc
    initcall_blacklist=sysfb_init
    nomodeset
    vfio_iommu_type1.allow_unsafe_interrupts=1 pcie_acs_override=downstream video=efifb:off video=vesafb:off

/etc/模块

    vfio
    vfio_iommu_type1
    vfio_pci
    #vfio_virqfd

/etc/modprobe.d/pve-blacklist.conf

    blacklist nvidiafb
    blacklist nouveau

/etc/modprobe.d/vfio.conf

    options vfio-pci ids=10de:1b38 disable_vga=1 disable_idle_d3=1

试过:

    disable_denylist=1
    pci=realloc

lspci -n -s 04:00 -v

04:00.0 0302: 10de:1b38 (rev a1)
        Subsystem: 10de:11d9
        Flags: fast devsel, IRQ 24, IOMMU group 27
        Memory at fa000000 (32-bit, non-prefetchable) [disabled] [size=16M]
        Memory at 3ff000000000 (64-bit, prefetchable) [disabled] [size=32G]
        Memory at 3ff800000000 (64-bit, prefetchable) [disabled] [size=32M]
        Capabilities: [60] Power Management version 3
        Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
        Capabilities: [78] Express Endpoint, MSI 00
        Capabilities: [100] Virtual Channel
        Capabilities: [250] Latency Tolerance Reporting
        Capabilities: [128] Power Budgeting <?>
        Capabilities: [420] Advanced Error Reporting
        Capabilities: [600] Vendor Specific Information: ID=0001 Rev=1 Len=024 <?>
        Capabilities: [900] Secondary PCI Express
        Kernel driver in use: vfio-pci
        Kernel modules: nvidiafb, nouveau, nvidia_drm, nvidia

虚拟机.conf:

    agent: 1
    balloon: 0
    bios: ovmf
    boot: order=scsi0;ide0;ide2
    cores: 4
    cpu: x86-64-v2-AES
    efidisk0: ssd-1tb:333/vm-333-disk-0.qcow2,efitype=4m,pre-enrolled-keys=1,size=528K
    hostpci0: 0000:04:00,pcie=1,x-vga=1
    machine: pc-q35-8.1
    memory: 4096
    meta: creation-qemu=8.1.2,ctime=1706034719
    name: passthrough
    net0: e1000=BC:24:11:84:4B:ED,bridge=vmbr0,firewall=1
    numa: 0
    ostype: win10
    scsi0: ssd-1tb:333/vm-333-disk-1.qcow2,iothread=1,size=32G
    scsihw: virtio-scsi-single
    smbios1: uuid=26873cc1-edc2-4493-b09d-82a93e37bb77
    sockets: 1
    tpmstate0: ssd-1tb:333/vm-333-disk-2.raw,size=4M,version=v2.0
    vga: virtio
    vmgenid: 957dfd44-c238-4720-811f-75f72b833405

试过:

    args: qm set VMID -args '-global q35-pci host.pci-hole64-size=2048G'
    args: -device vfio-pci,host=1a:00.0,multifunction=on,x-vga=on,bus=pcie.0,addr=0x10.0 -device vfio-pci,host=1a:00.1,bus=pcie.0,addr=0x10.1
    args: -cpu host,hv_time,kvm=off,hv_vendor_id=TheNeverhood,-hypervisor -device vfio-pci,host=02:00.0,id=hostpci0.0,bus=pcie.0,addr=0x10.0,multifunction=on -device vfio-pci,host=02:00.1,id=hostpci0.1,bus=pcie.0,addr=0x10.1

也尝试过设置显示:无,然后使用远程桌面。

在虚拟机硬件、PCI 设备中,尝试过使用和不使用主 GPU 和 ROM-Bar,尝试提供 ROM 文件、原始文件和使用 NVIDIA-vBIOS-VFIO-Patcher 修补的文件

相关内容