Kernel Modules#

The NVIDIA Linux GPU Driver contains several kernel modules:

  • nvidia.ko

  • nvidia-modeset.ko

  • nvidia-uvm.ko

  • nvidia-drm.ko

  • nvidia-peermem.ko

Starting in the 515 driver release series, two “flavors” of these kernel modules are provided:

  • Proprietary - This is the flavor that NVIDIA has historically shipped. For older GPUs from the Maxwell, Pascal, or Volta architectures. The open-source GPU kernel modules are not compatible with your platform, so the proprietary modules is what you are required to use.

  • Open-source - Published kernel modules that are dual licensed MIT/GPLv2. With every driver release, the source code to the open kernel modules will be published at NVIDIA/open-gpu-kernel-modules and a tarball will be provided at https://download.nvidia.com/XFree86/NVIDIA-kernel-module-source/. These are only for Turing and newer architectures, and this is what you should use if you have one of those architectures.

Starting in the 560 driver release series, the open kernel module flavor is the default installation.

Open GPU kernel modules are supported only on Turing and newer generations. To verify that your NVIDIA GPU is at least Turing or newer:

$ lspci | grep VGA

Open GPU Kernel Modules Installation#

For simplification, we’ve condensed the package manager recommendations in table format. All releases beyond driver version 570 will use these packaging conventions.

The following commands will install a full driver, which will include all desktop components as well as CUDA libraries and tools for computational workloads.

Table 5 Package manager installation recommendations#

Distribution

Install the Latest

Install a Specific Release

Red Hat Enterprise Linux 8/9, Rocky Linux 8/9

Oracle Linux 8/9, KylinOS 10, Amazon Linux 2023

# dnf module enable nvidia-driver:open-dkms

# dnf install nvidia-open

# dnf module enable nvidia-driver:580-open

# dnf install nvidia-open-580

Red Hat Enterprise Linux 10, Rocky Linux 10,

Fedora 42

# dnf install nvidia-open

# dnf install nvidia-open-580

Azure Linux 2 / Azure Linux 3

# tdnf install nvidia-open

# tdnf install nvidia-open-580

openSUSE Leap 15

# zypper install nvidia-open

# zypper install nvidia-open-580

SUSE Linux Enterprise Server 15 (x86_64)

# zypper install nvidia-open

# zypper install nvidia-open-580

SUSE Linux Enterprise Server 15 (aarch64)

# zypper install nvidia-open

# zypper install nvidia-open-580

Debian 12

# apt install nvidia-open

# apt install nvidia-open-580

Ubuntu 22.04/24.04

# apt install nvidia-open

# apt install nvidia-open-580

Proprietary GPU Kernel Modules Installation#

For simplification, we’ve condensed the package manager recommendations in table format. All releases beyond driver version 570 will use these packaging conventions.

The following commands will install a full driver, which will include all desktop components as well as CUDA libraries and tools for computational workloads.

Table 6 Proprietary GPU Kernel Module Installation#

Distribution

Install the Latest

Install a Specific Release

Red Hat Enterprise Linux 8/9, Rocky Linux 8/9

Oracle Linux 8/9, KylinOS 10, Amazon Linux 2023

# dnf module enable nvidia-driver:latest-dkms

# dnf install cuda-drivers

# dnf module enable nvidia-driver:580-dkms

# dnf install cuda-drivers-580

Red Hat Enterprise Linux 10, Rocky Linux 10,

Fedora 42

# dnf install cuda-drivers

# dnf install cuda-drivers-580

Azure Linux 2 / Azure Linux 3

Only the open kernel modules are supported.

Only the open kernel modules are supported.

openSUSE Leap 15

# zypper install cuda-drivers

# zypper install cuda-drivers-580

SUSE Linux Enterprise Server 15

# zypper install cuda-drivers

# zypper install cuda-drivers-580

Debian 12

# apt install cuda-drivers

# apt install cuda-drivers-580

Ubuntu 22.04/24.04

# apt install cuda-drivers

# apt install cuda-drivers-580