Modular GPU Programming with Typed Perspectives

TL;DR

Prism is a new GPU programming language that uses typed perspectives to enable modular, safe, and high-performance code for collective operations on modern GPUs.

Contribution

It introduces typed perspectives in Prism, a GPU language that balances modularity and low-level control, supported by a formal calculus and compiler implementation.

Findings

Prism enables safe modular GPU programming with competitive performance.

Typed perspectives improve correctness and safety in collective GPU operations.

Theoretical foundations support practical implementation of Prism.

Abstract

To achieve peak performance on modern GPUs, one must balance two frames of mind: issuing instructions to individual threads to control their behavior, while simultaneously tracking the convergence of many threads acting in concert to perform collective operations like Tensor Core instructions. The tension between these two mindsets makes modular programming error prone. Functions that encapsulate collective operations, despite being called per-thread, must be executed cooperatively by groups of threads. In this work, we introduce Prism, a new GPU language that restores modularity while still giving programmers the low-level control over collective operations necessary for high performance. Our core idea is typed perspectives, which materialize, at the type level, the granularity at which the programmer is controlling the behavior of threads. We describe the design of Prism, implement a compiler for it, and lay its theoretical foundations in a core calculus called Bundl. We implement state-of-the-art GPU kernels in Prism and find that it offers programmers the safety guarantees needed to confidently write modular code without sacrificing performance.