After years of insinuations, Nvidia announced yesterday that it would open source a piece of its Linux GPU driver code, as both Intel and AMD have done for years. Previously, Linux users who wanted to avoid Nvidia's proprietary driver had to rely on reverse-engineered software like the Nouveau project, which worked better on older hardware and offered limited support for all of Nvidia's GPU features.
“This release is a major step towards improving the experience of using NVIDIA GPUs on Linux, for tighter integration with the operating system and for developers to find bugs, integrate, and generally contribute,” reads a post attributed to several Nvidia employees. “For Linux distribution providers, open source modules increase ease of use. They also improve the “out of the box” user experience for distributing the Nvidia GPU driver. Canonical and SUSE can immediately merge the open kernel modules with Ubuntu and SUSE Linux Enterprise distributions.”
See also: Instant NeRF from NVIDIA: Converts 2D photos into 3D scenes
Nvidia specifically releases an open source kernel driver under a dual MIT/GPL license, and currently does not open source the driver code that runs in user space. This includes drivers for OpenGL, Vulkan, OpenCL, and CUDA, which are still closed source, along with the firmware for the GPU system processor (GSP). Nvidia says that these drivers will remain closed source and will be released with pre-built binaries, so there do not appear to be any immediate plans to release open source versions.
Nvidia still lags behind Intel and AMD when it comes to open source GPU drivers – both companies maintain open source for kernel and user-space drivers, along with closed-source firmware. But this is a first step towards open source parity for Nvidia's Linux driver packages.
Nvidia's open-source drivers only support the company's Turing-based GPUs and newer, including the GeForce GTX 1600 series, RTX 2000 and 3000 series, and Quadro workstation GPUs based on the same architecture. The company's proprietary drivers feature the 2012 Kepler architecture.
Open source drivers will not be integrated into the Linux kernel yet, as Phoronix reports, because the API, ABI, and interface between the kernel driver and the GSP firmware interface have not been finalized. This means that everything from the firmware to the kernel driver to the user space drivers must match their versions to work properly (currently, version R515.43.04). Once these interfaces are finalized, the driver can be integrated into the kernel, and different versions of the firmware, kernel driver, and user space drivers should be able to interoperate.
See also: NVIDIA Unveils Grace CPU Superchip
Nvidia says that support for its data center GPUs in the current drivers is production-ready in this initial release, but that support for GeForce, Quadro, and other consumer GPUs is first-rate—this isn't software you'll want to rush into installing unless you want to test it out on a test system.
Hector Martin, one of the developers behind the Asahi Linux distribution, criticized Nvidia for moving many functions to closed-source firmware, which is then called by the open-source driver. Martin calls the open-source driver “a clear win for practical purposes” as the mass of proprietary code can be more easily sandboxed.
Source: arstechnica.com
