Abstract/Details

Design and implementation of instruction set extension identification for a multiprocessor system-on-chip hardware/software co-design toolchain

Shapiro, Daniel.   University of Ottawa (Canada) ProQuest Dissertations Publishing,  2009. MR51858.

Abstract (summary)

Designers of embedded systems are faced with the challenge of creating systems containing customized processors. Efficient algorithms for the automated identification of extensions to processor instruction sets are needed, as well as flexible tools for the specification and generation of customized embedded systems.

This work on instruction set extension identification is part of a larger effort to design a multiprocessor system design tool. An algorithm for the identification of instruction set extensions is presented along with four approaches for improving the state of the art. In this research we addressed two important aspects of instruction set extension identification: reducing compiler execution time and identifying more instructions. Using the approaches described here, we achieved a compiler execution time speedup over the baseline compiler configuration as high as 3.8 times, and showed that compared to another popular approach our method identifies between 9.5 and 20 times the number of instruction set extensions.

Indexing (details)


Subject
Electrical engineering
Classification
0544: Electrical engineering
Identifier / keyword
Applied sciences
Title
Design and implementation of instruction set extension identification for a multiprocessor system-on-chip hardware/software co-design toolchain
Author
Shapiro, Daniel
Number of pages
121
Degree date
2009
School code
0918
Source
MAI 48/01M, Masters Abstracts International
Place of publication
Ann Arbor
Country of publication
United States
ISBN
978-0-494-51858-8
University/institution
University of Ottawa (Canada)
University location
Canada -- Ontario, CA
Degree
M.A.Sc.
Source type
Dissertation or Thesis
Language
English
Document type
Dissertation/Thesis
Dissertation/thesis number
MR51858
ProQuest document ID
305138347
Copyright
Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.
Document URL
https://www.proquest.com/docview/305138347