VectorWare maps Rust portable SIMD onto NVIDIA GPU warps

Christian Legnitto's compiler team is extending ordinary Rust abstractions across GPU threads, lanes and async execution.

By · Published

Primary source: VectorWare

Why it matters

VectorWare is moving GPU programming toward familiar systems code. Portable SIMD fills the lane-level gap, though Rust's API and VectorWare's compiler remain experimental.

The translation and efficient execution of Rust's portable SIMD abstractions onto NVIDIA GPU warps (Flat modernist vector illustration in the spirit of mid-century corporate annual reports — bold simple shapes, textured overprint)

VectorWare on August 10th demonstrated Rust's portable SIMD API running on an NVIDIA GPU, giving founder Christian Legnitto (@legneato) another piece of the programming model he wants developers to use for GPU-native applications. In a technical post, VectorWare showed ordinary core::simd code compiling into operations spread across the 32 lanes of a GPU warp.…

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

Reader comments

Conversation for this story loads after sign-in.