Code compression by register operand dependency

K. Lin, Jyh-Jiun Shann, Chung-Ping Chung

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

This paper proposes a dictionary-based code compression technique that maps the source register operands to the nearest occurrence of a destination register in the predecessor instructions. The key idea is that most destination registers have great potential to be used as source registers in the following instructions. The dependent registers can be removed from the dictionary if this information can be specified otherwise. As a result, the compression ratio benefits from the decreased dictionary size. A set of programs has been compressed using this feature. The compression results show that the average compression ratio is reduced to 38.6% on average for MediaBench benchmarks compiled for MIPS R2000 processor.

Original languageEnglish
Title of host publicationProceedings - 6th Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT 2002
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages91-101
Number of pages11
ISBN (Electronic)0769515347
DOIs
StatePublished - 1 Jan 2002
Event6th Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT 2002 - Cambridge, United States
Duration: 3 Feb 2002 → …

Publication series

NameProceedings - Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT
Volume2002-January
ISSN (Print)1550-6207

Conference

Conference6th Annual Workshop on Interaction between Compilers and Computer Architectures, INTERACT 2002
CountryUnited States
CityCambridge
Period3/02/02 → …

Keywords

  • Application specific integrated circuits
  • Costs
  • Dictionaries
  • Environmental economics
  • High level languages
  • Kelvin
  • Power generation economics
  • Read only memory
  • Registers
  • Silicon

Fingerprint Dive into the research topics of 'Code compression by register operand dependency'. Together they form a unique fingerprint.

Cite this