Characterizing function inlining with genetic programming

Abstract

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.