ABSTRACT
This paper evaluates the code efficiency of the ARM, Java, and x86 instruction sets by compiling the SPEC CPU95/ CPU2000/JVM98 and CaffeineMark benchmarks, in terms of code sizes, basic block sizes, instruction distributions, and average instruction lengths.
As a result, mainly because (i) the Java architecture is a stack machine, (ii) there are only four local variables which can be accessed by a 1-byte instruction, and (iii) additional instructions are provided for the network security, the code efficiency of Java turns out to be inferior to that of ARM Thumb. Moreover, through this efficiency analysis it should be claimed that a more efficient code architecture can be constructed by taking minute account of the customization of an instruction set as well as the number of registers.
- 1.Y.Yoshida,B.Song,H.Okuhata,T.Onoye,and I.Shirakawa.A low power consumption architecture for embedded processors.IEICE Trans.Fundamentals , J80-A(5):765 -771,May 1997.Google Scholar
- 2.N.Ishiura,M.Yamaguchi,and N.Nitta.Field partitioning algorithms for compression of instruction codes of application speci .c vliw processors.In Proc.International Technical Conference on Circuits/Systems,Computers and Communications , pages 1387 -1390,July 1998.Google Scholar
- 3.S.Segars,K.Clarke,and L.Goudge.Embedded control problems,thumb,and the arm7tdmi.IEEE MICRO ,15(7):22 -30,October 1995. Google ScholarDigital Library
- 4.T.Lindholm and F.Yellin.Java Virtual Machine Speci .cation Second Edition .Addison Wesley,1999. Google ScholarDigital Library
- 5.A.Hasegawa,I.Kawasaki,K.Yamada,S.Yoshioka, S.Kawasaki,and P.Biswas.Sh3:High code density, low power.IEEE MICRO ,15(6):11 -19,December 1995. Google ScholarDigital Library
- 6.Y.Nunomura,T.Shimizu,and O.Tomisawa. M32r/d-integrating dram and microprocessor.IEEE MICRO ,17(6):40 -48,November/December 1997. Google ScholarDigital Library
- 7.K.Kissell.MIPS16:High-density MIPS for the Embedded Market .Silicon Graphics MIPS Group, 1997.Google Scholar
- 8.R.E.Gonzalez.Xtensa:A con .gurable and extensible processor.IEEE MICRO ,20(2):60 -70,March/April 2000. Google ScholarDigital Library
- 9.N.Sugino,H.Miyazaki,and A.Nishihara.Dsp code optimization methods utilizing addressing operations at the codes without memory accesses.IEICE Trans. Fundamentals ,E80-A(12):2562 -2571,December 1997.Google Scholar
- 10.M.Game and A.Booker.CodePack:Code Compression for PowerPC Processors .MicroNews 5(1),IBM,1999.Google Scholar
- 11.S.Nam,I.Park,and C.Kyung.Improving dictionary-based code compression in vliw architectures.IEICE Trans.Fundamentals , E82-A(11):2318 -2324,November 1999.Google Scholar
- 12.SPEC CPU95 VERSION 1.00 .the Standard Performance Evaluation Corporation,1996.Google Scholar
- 13.SPEC CPU2000 VERSION 1.00 .the Standard Performance Evaluation Corporation,2000.Google Scholar
- 14.SPEC JVM98 VERSION 1.03 .the Standard Performance Evaluation Corporation,1998.Google Scholar
- 15.Java Ca .eineMark 3.0 .Pentagon Software Corporation,1999.Google Scholar
- 16.J.M.O 'connor and M.Tremblay.picojava-i:The java virtual machine in harware.IEEE MICRO , 17(2):45 -53,March/April 1997. Google ScholarDigital Library
- 17.picoJava-II Microarchitecture Guide .Sun Microsystems Inc.,1999.Google Scholar
- 18.EEMBC 's Version 1.0 .EDN Embedded Microprocessor Benchmark Consortium,(To appear).Google Scholar
Index Terms
- Evaluation of processor code efficiency for embedded systems
Recommendations
A continuation-based noninterruptible multithreading processor architecture
Current trend of research on multithreading processors is toward the chip multithreading (CMT), which exploits thread level parallelism (TLP) and improves performance of softwares built on traditional threading components, e.g., Pthread. There exist ...
Compiler Processor Tradeoffs for DISVLIW Architecture
ISPAN '02: Proceedings of the 2002 International Symposium on Parallel Architectures, Algorithms and NetworksThe Dynamically Instruction Scheduled VLIW (DISVLIW) processor architecture is designed for balancing scheduling effort more evenly between the compiler and the processor. The DISVLIW instruction format is augmented to allow dependency bit vectors to be ...
Retargetable code generation for application-specific processors
Special issue: Parallel computing technologiesAn approach of intelligent retargetable compiler is introduced to overcome the gap between hardware and software development and to increase performance of embedded systems. It focuses on knowledgeable treatment of code generation where knowledge about ...
Comments