lib: params_size for v19 driver compatibility: NVA081_CTRL_CMD_VGPU_CONFIG_GET_VGPU_TYPE_INFO

[gentoorax]

I've updated the v19 drivers and I'm seeing the following. I suspect it's maybe something similar to the bug just fixed on main lib: params_size for v18 driver compatibility, but I'm not close enough to it all to be sure. Unless I'm doing something wrong after updating to this version. I did git pull on vgpu_unlock-rs and rebuilt cargo clean, cargo build --release hopefully that's enough.

Help is much appreciated.

Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: Parameters size for NVA081_CTRL_CMD_VGPU_CONFIG_GET_VGPU_TYPE_INFO was 5424 bytes, expected 4960 bytes
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: cmd: 0x2080014b failed.
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: Parameters size for NVA081_CTRL_CMD_VGPU_CONFIG_GET_VGPU_TYPE_INFO was 5424 bytes, expected 4960 bytes

Driver: NVIDIA-Linux-x86_64-580.65.05-vgpu-kvm.run

Cargo.toml

[package]
name = "vgpu_unlock-rs"
version = "2.5.0"
edition = "2018"

[lib]
crate-type = ["cdylib"]

[dependencies]
ctor = "0.2.7"
libc = "0.2.102"
parking_lot = "0.12.1"
serde = { version = "1.0.130", features = ["derive"] }
toml = "0.8.11"

[features]
# Feature flag to enable syntactic sugar for proxmox users
default = ["proxmox"]
proxmox = []

nvidia-smi

Wed Aug 27 22:27:58 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 580.65.05              Driver Version: 580.65.05      CUDA Version: N/A      |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA RTX A5000               On  |   00000000:3B:00.0 Off |                    0 |
| 33%   62C    P8             35W /  230W |    1857MiB /  23028MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A          118156    C+G   vgpu                                   1856MiB |
+-----------------------------------------------------------------------------------------+

journalctl -u nvidia-vgpu-mgr.service

notice: vmiop_env_log: (0x0): Received start call from nvidia-vgpu-vfio module: mdev uuid 00000000-0000-0000-0000-000000000116 GPU PCI id 00:3b:00.7 config params vgpu_type_id=660
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_env_log: (0x0): pluginconfig: vgpu_type_id=660
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_env_log: Successfully updated env symbols!
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): detected a VF at 0:3b:0.7
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: Parameters size for NVA081_CTRL_CMD_VGPU_CONFIG_GET_VGPU_TYPE_INFO was 5424 bytes, expected 4960 bytes
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: cmd: 0x2080014b failed.
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: Parameters size for NVA081_CTRL_CMD_VGPU_CONFIG_GET_VGPU_TYPE_INFO was 5424 bytes, expected 4960 bytes
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): gpu-pci-id : 0x3b00
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vgpu_type : Quadro
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): Framebuffer: 0x74000000
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): Virtual Device Id: 0x2231:0x1565
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): FRL Value: 60 FPS
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: ######## vGPU Manager Information: ########
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: Driver Version: 580.65.05
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vGPU BAR1 size 256 MB
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): Detected ECC enabled on physical GPU.
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): Guest usable FB size is reduced due to ECC.
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vGPU supported range: (0x120001, 0x1a0001)
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vGPU manager is running in SRIOV mode.
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): placement Id: 0xffff
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vGPU UUID : 4e410c41-838b-11f0-9aae-bcd7ccdde9ed
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): Init frame copy engine: syncing...
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vGPU migration enabled
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: cmd: 0x20801322 failed.
Aug 27 22:17:58 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: display_init inst: 0 successful
Aug 27 22:18:24 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): Detected ECC enabled by guest.
Aug 27 22:18:24 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: ######## Guest NVIDIA Driver Information: ########
Aug 27 22:18:24 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: Driver Version: 580.65.06
Aug 27 22:18:24 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: vGPU version: 0x1a0001
Aug 27 22:18:24 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): MBP is enabled
Aug 27 22:18:24 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vGPU license state: Unlicensed (Unrestricted)
Aug 27 22:18:27 gamma nvidia-vgpu-mgr[118156]: notice: vmiop_log: (0x0): vGPU license state: Licensed

lspci -nn | grep nvidia

3b:00.0 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:00.4 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:00.5 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:00.6 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:00.7 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.0 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.1 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.2 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.3 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.4 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.5 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.6 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:01.7 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.0 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.1 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.2 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.3 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.4 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.5 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.6 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:02.7 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:03.0 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:03.1 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:03.2 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)
3b:03.3 3D controller [0302]: NVIDIA Corporation GA102GL [RTX A5000] [10de:2231] (rev a1)

[gentoorax]

OK, so I think the fix is as follows...

This removes the error I was seeing but it's still not setting the correct VRAM. e.g. framebuffer overrided by VMID still doesn't work. This is a real shame, as this is the main reason I use this :(

Perhaps I missed something else?

Created a PR for the below.

+                    // 19.0 driver sends a larger struct with size 5424 bytes, 18.0 driver sends a larger struct with size 5232 bytes, and 17.0 driver sends a larger struct with size 5096 bytes. Only extra members are added at the end,
@@ NVA081_CTRL_CMD_VGPU_CONFIG_GET_VGPU_TYPE_INFO => {
-                    if io_data.params_size == 5232
-                        || io_data.params_size == 5096
+                    if io_data.params_size == 5424
+                        || io_data.params_size == 5232
+                        || io_data.params_size == 5096

logs

Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Applying profile nvidia-660 overrides
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/num_heads: 0 -> 1
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/max_resolution_x: 0 -> 5120
Aug 27 23:08:18 gamma nvidia-vgpu-mgr[146470]: notice: vmiop_env_log: nvidia-vgpu-mgr daemon started
Aug 27 23:08:37 gamma nvidia-vgpu-mgr[132108]: notice: vmiop_log: (0x0): vGPU license state: Unlicensed (Unrestricted)
Aug 27 23:08:37 gamma nvidia-vgpu-mgr[132108]: notice: vmiop_log: (0x0): vGPU license state: Unlicensed (Restricted)
Aug 27 23:08:38 gamma nvidia-vgpu-mgr[132108]: notice: vmiop_log: (0x0): Guest driver unloaded!
Aug 27 23:08:41 gamma nvidia-vgpu-mgr[132108]: notice: vmiop_env_log: (0x0): QEMU migration state: STOPNCOPY_ACTIVE
Aug 27 23:08:41 gamma nvidia-vgpu-mgr[132108]: notice: vmiop_env_log: (0x0): Plugin migration stage change none -> stop_and_copy.
Aug 27 23:08:42 gamma nvidia-vgpu-mgr[132108]: notice: vmiop_log: Stopping all vGPU migration threads
Aug 27 23:08:55 gamma nvidia-vgpu-mgr[146470]: Nv0000CtrlVgpuCreateDeviceParams {
                                                   vgpu_name: {00000000-0000-0000-0000-000000000116},
                                                   gpu_pci_id: 0x3b00,
                                                   gpu_pci_bdf: 15111,
                                                   vgpu_type_id: 660,
                                                   vgpu_id: 1,
                                               }
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_env_log: vmiop-env: guest_max_gpfn:0x0
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_env_log: (0x0): Received start call from nvidia-vgpu-vfio module: mdev uuid 00000000-0000-0000-0000-000000000116 GPU PCI id 00:3b:00.7 config params vgpu_type_id=660
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_env_log: (0x0): pluginconfig: vgpu_type_id=660
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_env_log: Successfully updated env symbols!
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): detected a VF at 0:3b:0.7
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: NvA081CtrlVgpuConfigGetVgpuTypeInfoParams {
                                                   vgpu_type: 660,
                                                   vgpu_type_info: NvA081CtrlVgpuInfo {
                                                       vgpu_type: 660,
                                                       vgpu_name: "NVIDIA RTXA5000-2Q",
                                                       vgpu_class: "",
                                                       vgpu_signature: 0x51756164726f000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000>
                                                       license: "",
                                                       max_instance: 12334,
                                                       num_heads: 0,
                                                       max_resolution_x: 0,
                                                       max_resolution_y: 0,
                                                       max_pixels: 0,
                                                       frl_config: 0,
                                                       cuda_enabled: 0,
                                                       ecc_supported: 0,
                                                       gpu_instance_size: 0,
                                                       multi_vgpu_supported: 0,
                                                       vdev_id: 0x0,
                                                       pdev_id: 0x0,
                                                       profile_size: 0xc00000000,
                                                       fb_length: 0x1e0000000004,
                                                       gsp_heap_size: 0x2328000000010e0,
                                                       fb_reservation: 0x10000003c,
                                                       mappable_video_size: 0x1,
                                                       encoder_capacity: 0x1,
                                                       bar1_length: 0x22311565,
                                                       frl_enable: 8753,
                                                       adapter_name: "",
                                                       adapter_name_unicode: "噎䑉䅉删塔㕁〰ⴰ儲",
                                                       short_gpu_name_string: "",
                                                       licensed_product_name: "GA102GL-A",
                                                       vgpu_extra_params: [],
                                                       ftrace_enable: 0,
                                                       gpu_direct_supported: 0,
                                                       nvlink_p2p_supported: 0,
                                                       multi_vgpu_exclusive: 0,
                                                       exclusive_type: 0,
                                                       exclusive_size: 0,
                                                       gpu_instance_profile_id: 0,
                                                   },
                                               }
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Applying profile nvidia-660 overrides
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/num_heads: 0 -> 1
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/max_resolution_x: 0 -> 5120
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/max_resolution_y: 0 -> 1440
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/max_pixels: 0 -> 7372800
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/cuda_enabled: 0 -> 1
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/frl_enable: 8753 -> 1
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Applying mdev UUID 00000000-0000-0000-0000-000000000116 profile overrides
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/fb_length: 32985348833284 -> 3959422976
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: cmd: 0x2080014b failed.
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: NvA081CtrlVgpuConfigGetVgpuTypeInfoParams {
                                                   vgpu_type: 660,
                                                   vgpu_type_info: NvA081CtrlVgpuInfo {
                                                       vgpu_type: 660,
                                                       vgpu_name: "NVIDIA RTXA5000-2Q",
                                                       vgpu_class: "",
                                                       vgpu_signature: 0x51756164726f000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000>
                                                       license: "",
                                                       max_instance: 12334,
                                                       num_heads: 0,
                                                       max_resolution_x: 0,
                                                       max_resolution_y: 0,
                                                       max_pixels: 0,
                                                       frl_config: 0,
                                                       cuda_enabled: 0,
                                                       ecc_supported: 0,
                                                       gpu_instance_size: 0,
                                                       multi_vgpu_supported: 0,
                                                       vdev_id: 0x0,
                                                       pdev_id: 0x0,
                                                       profile_size: 0xc00000000,
                                                       fb_length: 0x1e0000000004,
                                                       gsp_heap_size: 0x2328000000010e0,
                                                       fb_reservation: 0x10000003c,
                                                       mappable_video_size: 0x1,
                                                       encoder_capacity: 0x1,
                                                       bar1_length: 0x22311565,
                                                       frl_enable: 8753,
                                                       adapter_name: "",
                                                       adapter_name_unicode: "噎䑉䅉删塔㕁〰ⴰ儲",
                                                       short_gpu_name_string: "",
                                                       licensed_product_name: "GA102GL-A",
                                                       vgpu_extra_params: [],
                                                       ftrace_enable: 0,
                                                       gpu_direct_supported: 0,
                                                       nvlink_p2p_supported: 0,
                                                       multi_vgpu_exclusive: 0,
                                                       exclusive_type: 0,
                                                       exclusive_size: 0,
                                                       gpu_instance_profile_id: 0,
                                                   },
                                               }
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Applying profile nvidia-660 overrides
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/num_heads: 0 -> 1
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/max_resolution_x: 0 -> 5120
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/max_resolution_y: 0 -> 1440
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/max_pixels: 0 -> 7372800
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/cuda_enabled: 0 -> 1
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/frl_enable: 8753 -> 1
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Applying mdev UUID 00000000-0000-0000-0000-000000000116 profile overrides
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: Patching nvidia-660/fb_length: 32985348833284 -> 3959422976
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): gpu-pci-id : 0x3b00
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vgpu_type : Quadro
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): Framebuffer: 0x74000000
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): Virtual Device Id: 0x2231:0x1565
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): FRL Value: 60 FPS
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: ######## vGPU Manager Information: ########
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: Driver Version: 580.65.05
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vGPU BAR1 size 256 MB
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): Detected ECC enabled on physical GPU.
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): Guest usable FB size is reduced due to ECC.
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vGPU supported range: (0x120001, 0x1a0001)
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vGPU manager is running in SRIOV mode.
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): placement Id: 0xffff
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vGPU UUID : 7e21bdc0-8392-11f0-b610-48bcd7ccdde9
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): Init frame copy engine: syncing...
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vGPU migration enabled
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: cmd: 0x20801322 failed.
Aug 27 23:09:25 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: display_init inst: 0 successful
Aug 27 23:09:50 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): Detected ECC enabled by guest.
Aug 27 23:09:50 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: ######## Guest NVIDIA Driver Information: ########
Aug 27 23:09:50 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: Driver Version: 580.65.06
Aug 27 23:09:50 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: vGPU version: 0x1a0001
Aug 27 23:09:50 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): MBP is enabled
Aug 27 23:09:50 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vGPU license state: Unlicensed (Unrestricted)
Aug 27 23:09:53 gamma nvidia-vgpu-mgr[147369]: notice: vmiop_log: (0x0): vGPU license state: Licensed

[gentoorax]

OK had a chance to look a this a bit further...

The struct layout changed in v19, not just its length. The pretty-printed fields are garbage (e.g. max_instance: 759449938, fb_length: 0x1e0000000004, Unicode junk). That’s the smoking gun: the repr(C) layout for NvA081CtrlVgpuInfo no longer matches what the v19 driver returns.

I think I might be able to fix this with a bit of AI help, I'm not a rust expert, I'm a C# platform engineer type guy. I'll give it an attempt tomorrow after work, but feel free to put me out of my misery anyone else that wants to help.

Would you believe this all started from trying to deploy vLLM Qwen models in my k3s cluster that required CUDA v.12.8 or better, that resulted in another worker node and a whole lot of driver upgrade pain, and here we are; talk about a side quest!

[rbqvq]

I think you are not read my PR.
NOT only length are extended.
ALL Array size also extended.

I think you should see what I changed first.

[rbqvq]

I wouldn't have thrown out backward compatibility if a simple change to the check would have made it pass.

The fundamental reason is that the whole structure has changed dramatically, unless we make two structures for the old version and the new version respectively, and then operate dynamically through the size.