Rust's Portable SIMD Now Runs on GPUs

Rust SIMD on the GPU

VectorWare announces support for Rust's portable SIMD (`core::simd`) on GPUs, mapping SIMD vectors directly to warp lanes. This enables the same SIMD code to run on CPUs and GPUs, with reductions and shuffles using warp primitives. The implementation leverages Rust's type system to encode a warp-level IR, offering zero-cost abstraction when vector width matches warp size. Challenges include unstable SIMD features and lane-width mismatches.