Tail call elimination and data representation for functional languages on the Java virtual machine

Abstract

The Java Virtual Machine (JVM) offers an attractive runtime environment for programming language implementers. The JVM has a simple bytecode format, excellent performance, multiple state-of-the art garbage collectors, robust backwards compatibility, and it runs on almost all platforms. Further, the Java ecosystem grants access to a plethora of libraries and tooling, including debuggers and profilers. Yet, the JVM was originally designed for Java, and its representation of code and data may cause difficulties for other languages. In this paper, we discuss how to efficiently implement functional languages on the JVM. We focus on two overall challenges: (a) how to efficiently represent algebraic data types in the presence of tags and tuples, option types, newtypes, and parametric polymorphism, and (b) how to support full tail call elimination on the JVM. We present two technical contributions: A fused representation of tags and tuples and a full tail call elimination strategy that is thread-safe. We implement these techniques in the Flix language and evaluate their performance.