The open source Rust CUDA project is “restarted” to enable the NVIDIA CUDA COMPUTE kernel to be able to code in the Rust programming language.
The Rust CUDA project enables the CUDA kernel of the NVIDIA GPU via the Rust code, and as a result, the NVIDIA NVVM IR is focused on the GPU calculation kernel and translated into the NVIDIA GPU executable PTX code. We target the target compiler.
Since the V0.3 Mile Stone in February 2022, there has been no new release of the open source Rust CUDA project, but the developers have returned to work three years later. They also want to find more open source developers to work on this project. Developers want to move at least some Rust CUDA shared assets in the Rust GPU project focusing on the target of SPIR-V / Vulkan. There is also a larger collaboration with CUDARC as abstraction of CUDA programming language between Rustc’s PTX backend and CUDARC.
Rust CUDA developers want to merge their outstanding pull demands, update their dependencies, launch a new project website, and take care of other items on the TODO list. Mid -term work includes interactions of Rust and C ++, PTX backend collaboration, and Rust GPU project collaboration. Long -term, they plan to explore the Rust language, which is a unified GPU API for integration and rust, and evolve the Rust language to support GPU programming better.
If you want to know more about this Rust CUDA project, you can do so through the Rust-GPU GitHub blog of Christian Legnitto.
