NVIDIA Announces Full Transition to Open Source GPU Kernel Modules
NVIDIA has announced that, beginning with the upcoming R560 driver release, they will fully transition to open source GPU kernel modules across all platforms.
In the technical blog post, the company notes that since the initial open source release of Linux GPU kernel modules in May 2022 under dual GPL and MIT licensing, they’ve “achieved equivalent or better application performance with our open source GPU kernel modules and added substantial new capabilities,” such as:
- Heterogeneous memory management support
- Confidential computing
- Coherent memory architectures of Grace platforms
“Switching to open source GPU kernel modules helps NVIDIA improve integration with the Linux operating system. It also allows developers to debug, integrate, and contribute to enhancing the drivers,” the blog post says.
Learn more about specific support and installation requirements at NVIDIA.
Related content
- NVIDIA Finally Embraces Open Source with GPU Kernel Modules
- Linux Kernel 5.11 has been Released
- NVIDIA Announces a Specification for Hybrid Quantum-Classical Computing
-
Parallel Programming with OpenCL
We take a close look at how to integrate graphics processors into your parallel programs with OpenCL.
- NVidia Announces New Kepler-Based GPUs