Open Source GPU Drivers for Linux – The New Stack


Chipmaker Nvidia is finally taking steps to improve its GPUs’ compatibility with Linux after years of largely ignoring users of the open-source operating system.

Nvidia has open-source GPU kernel modules for Linux so programmers can write applications to run more efficiently on the company’s GPUs.

“Canonical and SUSE can immediately package the open kernel modules with Ubuntu and SUSE Linux enterprise distributions,” Nvidia said in a blog post.

The release is Nvidia’s latest attempt to appeal to the open source community. The company is also involved in bringing native support for parallel code execution across CPUs, GPUs and AI accelerators to C++23, which will supersede the current C++20 standard.


Nvidia has received a lot of criticism for not making its GPUs Linux-friendly. Earlier this year, hackers from Lapsus$ broke into Nvidia’s systems, and one of the group’s demands was to open source their GPU drivers for the Linux operating system.

Software developers can also integrate the drivers into custom Linux builds using the open source drivers. Nvidia also welcomes input from the Linux community to improve drivers, which the company will review. Nvidia’s goal is to eventually push the drivers upstream.

The GPU kernel modules are open source with a dual GPL/MIT license. The R515 release drivers are an alpha release and will support some newer Nvidia GPUs that include consumer hardware based on the Turing architecture released in 2018 and Ampere architecture for consumer and professional GPUs released in 2020. However, the drivers do not apply to these latest Hopper architecture-based GPUs, which have special modifications that include lower-level floating-point execution for smaller datasets.

Nvidia GPUs, originally famous among gamers, paved the way for AI adoption in enterprises. The driver pack is aimed at companies that want to use neural networks, but also at engineers and scientific computers that need to visualize information.

driver pack

The driver pack also supports game features based on OpenGL and Vulkan, standards on which most games for the Linux operating system are written. Nvidia GPUs still work best with games on Windows, but the driver set allows Nvidia GPUs to work better with Linux games. The driver set supports multiple displays and ray tracing specifications from Vulkan.

Open-sourcing the drivers could be a way to encourage adoption of CUDA, Nvidia’s parallel programming framework. Nvidia’s rivals AMD and Intel have been releasing open source GPU drivers to the open source community in the gaming and parallel programming framework for many years. But Nvidia dominates the supercomputing and AI hardware space, while Intel’s OneAPI and AMD’s ROCm parallel programming frameworks have not found widespread adoption.

Nvidia also gave some hints that future releases of open-source drivers would include hooks for features like on-chip functionality to secure data on its latest Hopper GPUs.

Nvidia last month hired open-source advocate Guy Martin as its director of open source and standards, to lead the company’s open-source efforts. Previously, he was executive director at OASIS, an organization promoting the use of open source in both the public and private spheres.

Featured image via Pixabay.


Comments are closed.