Skip to content

Commit ab5c618

Browse files
committed
docs(gpu-vgpu): fix driver container build instructions and NLS config
Replace simplified Containerfile with NVIDIA's Makefile-based build system from gitlab.com/nvidia/container-images/driver. The GPU Operator expects pre-compiled kernel modules, not a raw .run file. Add EULA notice about public redistribution of vGPU driver images. Add note about NLS ServerPort being deployment-dependent. Assisted-By: Claude <[email protected]> Signed-off-by: Aleksei Sviridkin <[email protected]>
1 parent 2e1f349 commit ab5c618

1 file changed

Lines changed: 18 additions & 13 deletions

File tree

‎docs/gpu-vgpu.md‎

Lines changed: 18 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -18,28 +18,32 @@ The GPU Operator package supports two variants:
1818

1919
## Building the vGPU Manager Image
2020

21-
The vGPU Manager driver is proprietary and must be obtained from NVIDIA:
21+
The vGPU Manager driver is proprietary and must be obtained from NVIDIA. The GPU Operator expects a pre-built driver container image — it does not install the driver from a raw `.run` file at runtime.
2222

2323
1. Log in to the [NVIDIA Licensing Portal](https://ui.licensing.nvidia.com)
24-
2. Download the NVIDIA vGPU Software package for your GPU and target Linux version
25-
3. Build a container image containing the vGPU Manager driver:
24+
2. Navigate to **Software Downloads** and download the NVIDIA vGPU Software package for your GPU
25+
3. Build the driver container image using NVIDIA's Makefile-based build system:
2626

2727
```bash
28-
# Example Containerfile
29-
FROM ubuntu:22.04
30-
ARG DRIVER_VERSION
31-
COPY NVIDIA-Linux-x86_64-${DRIVER_VERSION}-vgpu-kvm.run /opt/
32-
RUN chmod +x /opt/NVIDIA-Linux-x86_64-${DRIVER_VERSION}-vgpu-kvm.run
33-
```
28+
# Clone the NVIDIA driver container repository
29+
git clone https://gitlab.com/nvidia/container-images/driver.git
30+
cd driver
3431

35-
4. Push the image to your private registry:
32+
# Place the downloaded .run file in the appropriate directory
33+
cp NVIDIA-Linux-x86_64-550.90.05-vgpu-kvm.run vgpu/
3634

37-
```bash
38-
docker build --build-arg DRIVER_VERSION=550.90.05 --tag registry.example.com/nvidia/vgpu-manager:550.90.05 .
35+
# Build using the provided Makefile
36+
make OS_TAG=ubuntu22.04 \
37+
VGPU_DRIVER_VERSION=550.90.05 \
38+
PRIVATE_REGISTRY=registry.example.com/nvidia
39+
40+
# Push to your private registry
3941
docker push registry.example.com/nvidia/vgpu-manager:550.90.05
4042
```
4143

42-
> Refer to the [NVIDIA GPU Operator documentation](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/install-gpu-operator-vgpu.html) for detailed instructions on building the vGPU Manager image.
44+
> **Important:** The build process compiles kernel modules against the host kernel version. Refer to the [NVIDIA GPU Operator vGPU documentation](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/install-gpu-operator-vgpu.html) for the complete build procedure and supported OS/kernel combinations.
45+
>
46+
> Uploading the vGPU driver to a publicly available registry is a violation of the NVIDIA vGPU EULA.
4347
4448
## Deploying with vGPU Variant
4549

@@ -89,6 +93,7 @@ data:
8993
ServerAddress=nls.example.com
9094
ServerPort=443
9195
FeatureType=1
96+
# ServerPort depends on your NLS deployment (commonly 443 for DLS or 7070 for legacy NLS)
9297
```
9398

9499
2. Reference the ConfigMap in the Package values:

0 commit comments

Comments
 (0)