@inproceedings{57421a0d5f024b8d99bfed14015f284c,
title = "Using profile information to assist advanced compiler optimization and scheduling",
abstract = "Compilers for superscalar and VLIW processors must expose sufficient instruction-level parallelism in order to achieve high performance. Compile-time code transformations which expose instruction-level parallelism typically take into account the constraints imposed by all execution scenarios in the program. However, there are additional opportunities to increase instruction-level parallelism along the frequent execution scenario at the expense of the less frequent execution sequences. Profile information identifies these important execution sequences in a program. In this paper, two major categories of profile information are studied: control-flow and memory-dependence. Profile-based transformations have been incorporated into the IMPACT compiler. These transformations include global optimization, acyclic global scheduling, and software pipelining. The effectiveness of these profile-based techniques is evaluated for a range of superscalar and VLIW processors.",
author = "W. Chen and R. Bringmann and S. Mahlke and S. Anik and T. Kiyohara and N. Warter and D. Lavery and Hwu, {W. M.} and R. Hank and J. Gyllenhaal",
year = "1993",
month = jan
day = "1",
language = "English (US)",
isbn = "9783540575023",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "31--48",
editor = "David Padua and Alex Nicolau and David Gelernter and Utpal Banerjee",
booktitle = "Languages and Compilers for Parallel Computing - 5th International Workshop, Proceedings",
note = "IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2017 ; Conference date: 03-09-2017 Through 07-09-2017",
}