article · 2000
Iterative Compilation in Program Optimization
T. Kisuki, P.M.W. Knijnenburg, M.F.P. O'Boyle, H. A. G. Wijshoff
Cited by 0Abstract and details only
Abstract
In this paper we describe a novel approach to program optimization, namely, iterative compilation. This technique enables compilers to deliver efficient code by searching for the best sequence of optimizations using profile-directed feedback. We have implemented this iterative approach in an existing compiler system which is designed for embedded systems and compared its performance to two of the best known static techniques. Initial experimental results show that this approach delivers an average improvement of 35% over existing techniques and delivers efficient code with reasonable compilation times. 1. Introduction Modern compilers make extensive use of optimization to improve program performance. The use of a particular optimization largely depends on static program analysis based on simplified machine models. Optimizations include high-level loop transformations, such as loop unrolling and tiling. These techniques are extensively studied for over 30 years and have produced, in m...
Cite this paper
Kisuki, T., Knijnenburg, P., O'Boyle, M., & Wijshoff, H. A. G. (2000). Iterative Compilation in Program Optimization.
Kisuki, T., et al. “Iterative Compilation in Program Optimization.” 2000.
Kisuki, T., P.M.W. Knijnenburg, M.F.P. O'Boyle, and H. A. G. Wijshoff. 2000. “Iterative Compilation in Program Optimization.”
Kisuki, T. et al. (2000) ‘Iterative Compilation in Program Optimization’.
@article{kisuki2000iterative,
title = {{Iterative Compilation in Program Optimization}},
author = {Kisuki, T. and Knijnenburg, P.M.W. and O'Boyle, M.F.P. and Wijshoff, H. A. G.},
year = {2000}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.