Java Virtual Machine Optimizations for Java and Dynamic Languages

Abstract

Java virtual machine (JVM) has been introduced as the machine-independent run- time environment to run a Java program. As a 32-bit stack machine, JVM can execute bytecode instructions generated through compilation of a Java program on any ma- chine if the JVM runtime was correctly ported on it. The machine-independence of JVM brought about the huge success of both the Java programming language and the Java virtual machine itself on various systems encompassing from cloud servers to embedded systems including handsets and smart cards. Since a bytecode instruction should be interpreted by the JVM runtime for execu- tion on top of a specific underlying system, a Java program runs innately slower due to the interpretation overhead than a C/C++ program that is compiled directly for the sys- tem. Java just-in-time (JIT) compilers, the de facto performance add-on modules, are employed to improve the performance of a Java virtual machine (JVM) by translating Java bytecode into native machine code on demand. One important problem in Java JIT compilation is how to map stack entries and local variables of the JVM runtime to physical registers efficiently and quickly, since register-based computations are much faster than memory-based ones, while JIT com- pilation overhead is part of the whole running time. This paper introduces LaTTe, an open-source Java JIT compiler that performs fast generation of efficiently register- mapped RISC code. LaTTe first maps all local variables and stack entries into pseudo registers, followed by real register allocation which also coalesces copies correspond- ing to pushes and pops between local variables and stack entries aggressively. In ad- dition to the efficient register allocation, LaTTe is equipped with various traditional and object-oriented optimizations such as CSE, dynamic method inlining, and special- ization. We also devised new mechanisms for Java exception handling and monitor handling in LaTTe, named on-demand exception handling and lightweight monitor, respectively, to boost up the JVM performance more. Our experimental results indicate that LaTTes sophisticated register mapping and allocation really pay off, achieving twice the performance of a naive JIT compiler that maps all local variables and stack entries to memory. It is also shown that LaTTe makes a reasonable trade-off between quality and speed of register mapping and allocation for the bytecode. We expect these results will also be beneficial to parallel and distributed Java computing 1) by enhancing single-thread Java performance and 2) by significantly reducing the number of memory accesses which the rest of the system must properly order to maintain coherence and keep threads synchronized. Furthermore, Java virtual machine (JVM) has recently evolved into a general- purpose language runtime environment to execute popular programming languages such as JavaScript, Ruby, Python, or Scala. These languages have complex non-Java features including dynamic typing and first-class function, so additional language run- times (engines) are provided on top of the JVM to support them with bytecode ex- tensions. Although there are high-performance JVMs with powerful just-in-time (JIT) compilers, running these languages efficiently on the JVM is still a challenge. This paper introduces a simple and novel technique for the JVM JIT compiler called exceptionization to improve the performance of JVM-based language runtimes. We observed that the JVM executing some non-Java languages encounters at least 2 times more branch bytecodes than Java, most of which are highly biased to take only one target. Exceptionization treats such a highly-biased branch as some implicit exception-throwing instruction. This allows the JVM JIT compiler to prune the infre- quent target of the branch from the frequent control flow, thus compiling the frequent control flow more aggressively with better optimization. If a pruned path was taken, it would run like a Java exception handler, i.e., a catch block. We also devised de- exceptionization, a mechanism to cope with the case when a pruned path is actually executed more often than expected. Since exceptionization is a generic JVM optimization, independent of any specific language runtime, it would be generally applicable to any language runtime on the JVM. Our experimental result shows that exceptionization accelerates the performance of several non-Java languages. The JavaScript-on-JVM runs faster by as much as 60%, and by 6% on average, when running the Octane benchmark suite on Oracles latest Nashorn JavaScript engine and HotSpot 1.9 JVM. Additionally, the Ruby-on-JVM experiences the performance improvement by as much as 60% and by 6% on average, while the Python-on-JVM by as much as 6%. We found that exceptionization is most effectively applicable to the branch bytecode of the language runtime itself, rather than the bytecode corresponding to the application code or the bytecode of the Java class libraries. This implies that the performance benefit of exceptionization comes from better JIT compilation of the non-Java language runtime.