Fast, effective code generation in a just-in-time Java compiler

Adl-Tabatabai, Ali-Reza; Cierniak, Michał; Lueh, Guei-Yuan; Parikh, Vishesh M.; Stichnoth, James M.

doi:10.1145/277650.277740

Cited by 112 publications

(49 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The Intel JIT compiler uses lazy code selection in the context of on-the-fly register allocation to speed up compilation [1]. Previously encountered bytecode sequences that generated the current value in a register are cached, as well as possible bytecode aUases, so that future equivalent sequences can simply be replaced by an instantiation of the register.…”

Section: Related Workmentioning

confidence: 99%

“…The inliner is also responsible for inlining fast, common case handlers for the checkcast and instanceof bytecodes, which must execute a costly class hierarchy search if the object class is not equivalent to the requested class. [2] cmp [Lll [1] [2] cmp [LU [1] 2 iaload [3] cbranch ult [2] ,, [31 cbranch ult [2] 3 iconst_l ' C -> [51 c"^ -> [51 4 iadd [4] call bad array_idx [4] call bad array_idx 5 aload 0 (5) target [5] target 6 iload_2 [6] const 2 [6] const 2 7 iastore [7] sll [Lll [6] [7] sll [Lll [6] Equivalent C [8] Figure 4. Array accesses bytecodes are decomposed so that optimizations can be performed on their components.…”

Section: Inlining and Specializationmentioning

confidence: 99%

“…The first approach uses techniques that focus solely on generating code quickly [1] [9]. This usually involves focusing only on simple optimizations, or relying on imprecise analysis or heuristics to speed up compilation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Targeting Dynamic Compilation for Embedded Environments

Chen¹,

Olukotun²

2001

View full text Add to dashboard Cite

A generally held notion is that high quality code comes with high compilation cost. As a result, previous efforts at minimizing dynamic compilation costs have focused on designing fast, lightweight compilers that sacrifice code quality for compilation speed, and resource intensive approaches that combine multiple engines to limit expensive optimizations to critical sections. In this paper, we show one possible way fast compilers can be constructed to generate high quality code. We have implemented microJIT, a small and portable just-in-time (JIT) compiler for Java that can produce high quality code 2.5x faster than a comparable dataflow-based compiler and 30% faster than a compiler that performs only limited optimizations. We use dataflow techniques, but speed up compilation by minimizing the number of major compiler passes given the number of optimizations performed. Architectural features of our compiler also allow it to perform instruction set dependent optimizations efficiently. microJIT achieves these high compilation rates while still maintaining small static and dynamic memory requirements. This compiler can be highly effective in an embedded system where computing and memory resources are highly constrained and where multiple target platforms must be supported.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Inlining and Specializationmentioning

confidence: 99%

See 1 more Smart Citation

Targeting Dynamic Compilation for Embedded Environments

Chen¹,

Olukotun²

2001

View full text Add to dashboard Cite

show abstract

“…The execution engine, the Common Language Runtime (CLR), is a stack-based Virtual Machine (VM) which executes a portable bytecode: the Common Intermediate Language (CIL) [10]. The CLR enforces type safety through its bytecode verifier (BCV), it supports polymorphism, the memory is garbage collected and the bytecode is Just-In-Time [1,21] compiled to native code.…”

Section: Introductionmentioning

confidence: 99%

Bigloo.NET: compiling Scheme to .NET CLR.

Bres¹,

Serpette²,

Serrano³

2004

JOT

View full text Add to dashboard Cite

This paper presents the compilation of the Scheme programming language to .NET. This platform provides a virtual machine, the Common Language Runtime (CLR), that executes bytecode, the Common Intermediate Language (CIL). Since CIL was designed with language agnosticism in mind, it provides a rich set of language constructs and functionalities. As such, the CLR is the first execution environment that offers type safety, managed memory, tail recursion support and several flavors of pointers to functions. Therefore, the CLR presents an interesting ground for functional language implementations. We discuss how to map Scheme constructs to CIL. We present performance analyses on a large set of real-life and standard Scheme benchmarks. In particular, we compare the speed of these programs when compiled to C, JVM and .NET. We show that in term of speed performance of the Mono implementation of .NET, the best implementing running on both Windows and Linux, still lags behind C and fast JVMs such as the Sun's implementations.

show abstract

“…JVM is designed so that it can be efficiently implemented on a large variety of platforms. JVM can be implemented directly in hardware as a JVM processor [2], by means of a software interpreter, or by means of a just-in-time compiler [3,4] which compiles bytecodes to native machine code just before the code is executed and caches it for future execution.…”

Section: Introductionmentioning

confidence: 99%

Pipelined Java Virtual Machine Interpreters

Hoogerbrugge

Augusteijn

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The performance of a Java Virtual Machine (JVM) interpreter running on a very long instruction word (VLIW) processor can be improved by means of pipelining. While one bytecode is in its execute stage, the next bytecode is in its decode stage, and the next bytecode is in its fetch stage. The paper describes how we implemented threading and pipelining by rewriting the source code of the interpreter and several modifications in the compiler. Experiments for evaluating the effectiveness of pipelining are described. Pipelining improves the execution speed of a threaded interpreter by 19.4% in terms of instruction count and 14.4% in terms of cycle count. Most of the simple bytecodes, like additions and multiplications, execute in four cycles. This number corresponds to the branch latency of our target VLIW processor. Thus most of the code of the interpreter is executed in branch delay slots.

show abstract

Fast, effective code generation in a just-in-time Java compiler

Cited by 112 publications

References 15 publications

Targeting Dynamic Compilation for Embedded Environments

Targeting Dynamic Compilation for Embedded Environments

Bigloo.NET: compiling Scheme to .NET CLR.

Pipelined Java Virtual Machine Interpreters

Contact Info

Product

Resources

About