Characterizing function inlining with genetic programming
Author(s)
Yu, Chris, 1981-
DownloadFull printable version (3.648Mb)
Other Contributors
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Advisor
Saman P. Amarasinghe.
Terms of use
Metadata
Show full item recordAbstract
Function inlining is a compiler optimization where the function call is replaced by the code from the function itself. Using a form of machine learning called genetic programming, this thesis examines which factors are important in determining which function calls to inline to maximize performance. A number of different heuristics are generated for inlining decisions in the Trimaran compiler, which improve on performance from the current default inlining heuristic. Also, trends in function inlining are examined over the thousands of compilation runs that are completed.
Description
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2004. Includes bibliographical references (leaves 74-75).
Date issued
2004Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer SciencePublisher
Massachusetts Institute of Technology
Keywords
Electrical Engineering and Computer Science.