Timing Analysis for Instruction Caches

  • Authors:
  • Frank Mueller

  • Affiliations:
  • Humboldt University Berlin, Institut f. Informatik, 10099 Berlin (Germany)

  • Venue:
  • Real-Time Systems - Special issue on worst-case execution-time analysis
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper contributes a comprehensivestudy of a framework to bound worst-case instruction cache performancefor caches with arbitrary levels of associativity. The frameworkis formally introduced, operationally described and its correctnessis shown. Results of incorporating instruction cache predictionswithin pipeline simulation show that timing predictions for set-associativecaches remain just as tight as predictions for direct-mappedcaches. The low cache simulation overhead allows interactiveuse of the analysis tool and scales well with increasing associativity. The approach taken is based on a data-flow specificationof the problem and provides another step toward worst-case executiontime prediction of contemporary architectures and its use inschedulability analysis for hard real-time systems.