Applying Optimizations for Dynamically-typed Languages to Java

Abstract

While Java is a statically-typed language, some of its features make it behave like a dynamically-typed language at run time. This includes Java's boxing of primitive values as well as generics, which rely on type erasure. This paper investigates how runtime technology for dynamically-typed languages such as JavaScript and Python can be used for Java bytecode. Using optimistic optimizations, we specialize bytecode instructions that access references in such a way, that they can handle primitive data directly and also specialize data structures in order to avoid boxing for primitive types. Our evaluation shows that these optimizations can be successfully applied to a statically-typed language such as Java and can also improve performance signiicantly. With this approach, we get an eicient implementation of Java's generics, avoid changes to the Java language, and maintain backwards compatibility, allowing existing code to beneit from our optimization transparently.

Supporting on-stack replacement in unstructured languages by loop reconstruction and extraction

“…Specifically, our evaluation with a set of wellknown benchmarks shows significant reductions by up to 9x in program warm-up given that OSR is applicable. Note that our approach can be used by other Truffle bytecodebased implementations lacking high-level loop information, including GraalSqueak [19], Truffle Java [11] and Truffle CIL. Furthermore, it is applicable in any compiler with a background system which provides means for establishing mappings between extracted control flow.…”