Java annotation-aware just-in-time (AJIT) complilation system

Abstract

The Java Bytecode language lacks expressiveness for traditional compiler optimizations, making this portable, secure software distribution format inefficient as a program representation for high performance. This inefficiency results from the underlying stack model, as well as the fact that many bytecode operations intrinsically include sub-operations (e.g., iaXoad includes the address computation, a,rray bounds checks and the actual load of the array element). The stack model, with no operand registers and limiting access to the top of the stack, prevents the reuse of values and bytecode reordering. In addition, the bytecodes have no mechanism to indicate which sub-operations in the bytecode stream are redundant or subsumed by previous ones. As a consequence, the Java Bytecode language inhibits the expression of important compiler optimizations, including common sub-expression elimination, register allocation and instruction scheduling.The bytecode stream generated by the Java frontend is a significantly under-optimized program representation. The most common solution to overcome this aspect of the language is the use of a Just-in-Time (JIT) compiler to not only generate native code, but perform optimization as well. However, the latter is a time consuming operation in an already time-constrained translation process. In this paper we present an alternative to an optimizing JIT compiler that makes use of code annotations generated by the Java front-end. These annotations carry information concerning compiler optimization. During the translation process, an annotationaware JIT (AJIT) system then uses this information to produce high-performance native code without per-*This work supported K part by CAPES.Permission to make digital or hard copies of all or part ofthis work fix personal or classroom use is granted without fee provided that topics are not made or distributed for profit or commercial advantage and that copies hear this notice and the full citation on the tirst page. To copy otherwise, to republish, to post on servers or to redistribute to lists. requires prior specific permission and/or a fee. JAVA'99 San Francisco California USA Copyright ACM 1999 I-581 13-161-5/99/06...$5.00 forming much of the necessary analyses or transformations. We describe the implementation of the first prototype of our annotation-aware JIT system and show performance results comparing our system with other Java Virtual Machines (JVMs) running on SPARC architecture.

Split Register Allocation: Linear Complexity Without the Performance Penalty

“…Interestingly, annotation-driven JIT compilation was first directed to register allocation, with the pioneering work of Azevedo et al [11]. This work demonstrated how to achieve performance competitive with native priority-based graph coloring allocation.…”

Using annotations to reduce dynamic optimization time

“…Annotations have been widely used on program source code in various languages to exploit parallelization and optimization opportunities in parallel and distributed codes. More recently, annotation-based techniques have focused on communicating information that aids optimization, but is too time-consuming to collect on-line [4,13,21,10]. The goal of these efforts has been to make costly optimizations feasible in dynamic compilation settings.…”

“…However, this is not always the case and as such, we present results on the effect of annotating only non-local, or application, class files. Most of the prior work [4,13] on bytecode annotations does not address the effect of limiting optimization to remote class files yet it is vital to the empirical evaluation of annotation-based techniques. Figure 7 shows the reduction in compile time due to annotated-execution of only non-local class files.…”

“…Existing annotation-based techniques annotate Java bytecode with analysis information that is time-consuming to collect to guide dynamic compilation [4,13,21,10]. The goal of this prior work was to make costly optimizations feasible in dynamic compilation settings.…”

Coupling on-line and off-line profile information to improve program performance

“…More recently, off-line profile information has been communicated as part of class files via annotation to reduce dynamic compilation overhead [9]. Annotation has also been used to communicate analysis information that is collected off-line to enable optimized execution times [14,8,4]. Our system combines the use of annotation with adaptive optimization to enable optimized execution times with very little on-line overhead.…”