← All resources

Towards GPU Parallelism Abstractions in Rust: A Case Study with Linear Pipelines

Auto-discovered

Paper Advanced

Leonardo Gibrowski Faé, Dalvan Griebler

Visit resource ↗

Description

Towards GPU Parallelism Abstractions in Rust: A Case Study with Linear Pipelines presents a 2025 SBLP paper by Leonardo Gibrowski Fae and Dalvan Griebler (PUCRS) on high-level Rust abstractions for GPU linear pipelines.

It notes that GPGPU work still often needs CUDA or OpenCL expertise while Rust GPU use is early. The authors use procedural macros to generate GPU code from constrained Rust functions marked source, stage, or sink. Case studies sobel and latbol are compared with manual OpenCL for effort and throughput.

It is for intermediate to advanced readers comfortable with Rust and basic GPU pipeline ideas. Prior OpenCL or CUDA experience helps when reading the results.

This URL is a free hgpu.org abstract page with SBLP 2025 bibliographic metadata (pages 75-83).

It is useful as a compact reference on constrained Rust macros for simpler GPU pipeline authoring. Link last verified accessible on 2026-09-29. Link last verified accessible on 2026-09-28.