SIMD-efficient loop unrolling design for embedded multimedia applications

Yunyang Dai, Qing Li, Qi Zhang, C. C. Jay Kuo

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

4 Citations (Scopus)

Abstract

Due to the rising complexity of modern embedded media applications (EMAs), compilers must have the capability to exploit the superword level parallelism (SLP). This work analyzes the memory access patterns found in EMAs and presents a scheme to calculate the loop unrolling factor to fully utilize these patterns to generate efficient Single Instruction Multiple Data (SIMD) instructions. The loop nest is also considered for actual memory access patterns, which can be used to improve the efficiency of the compiler. We observe a performance improvement by an average factor of 12 times for manual experiments conducted on the TriMedia TM-1300 processor for the H.264 encoding application.

Original languageEnglish
Title of host publication2004 IEEE International Conference on Multimedia and Expo (ICME)
Pages1851-1854
Number of pages4
Publication statusPublished - 1 Dec 2004
Externally publishedYes
Event2004 IEEE International Conference on Multimedia and Expo (ICME) - Taipei, Taiwan
Duration: 27 Jun 200430 Jun 2004

Publication series

Name2004 IEEE International Conference on Multimedia and Expo (ICME)
Volume3

Conference

Conference2004 IEEE International Conference on Multimedia and Expo (ICME)
Country/TerritoryTaiwan
CityTaipei
Period27/06/0430/06/04

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'SIMD-efficient loop unrolling design for embedded multimedia applications'. Together they form a unique fingerprint.

Cite this