|
Funded by the Vicerreitoría de Investigación of the University of A Coruña and the Ministry of Education and Science of Spain 
|
Improving the Java Virtual Machine using Type-Separated Bytecode
Philipp Adler and Wolfram Amme
Friedrich-Schiller-Universität, Jena, Germany
Java Bytecode is currently the most used mobile code representation, although it contains some well-known major flaws. In the paper we introduce the principle operation of type-separated bytecode. Type-separated bytecode is a new intermediate representation that compensates some of the drawbacks introduced by Java Bytecode. In particular the use of type-spearated bytecode can considerably speed-up the verification process and supports optimizations performed on the producer side of a mobile code generation system. This is especially important when using bytecode on constrained devices like Java cards or in embedded systems.
