Procedure cloning

Author

Cooper, Keith D. ; Hall, Mary W. ; Kennedy, Ken

Author_Institution

Dept. of Comput. Sci., Rice Univ., Houston, TX, USA

fYear

1992

fDate

20-23 Apr 1992

Firstpage

96

Lastpage

105

Abstract

Procedure cloning is an interprocedural optimization where the compiler creates specialized copies of procedure bodies. The authors examine the problem of procedure cloning and describe an experiment where cloning was required to make other transformations possible. They present a three-phase algorithm for deciding how to clone a program and analyze the algorithm´s complexity. The algorithm finds potential improvements in forward interprocedural data-flow solutions and clones those procedures that lead to sharper information. A set of assumptions that bound both the running time of the algorithm and the expansion in code size is presented

Keywords

computational complexity; parallel programming; program compilers; software reusability; subroutines; code size; compiler; complexity; forward interprocedural data-flow solutions; interprocedural optimization; procedure bodies; procedure cloning; running time; specialized copies; three-phase algorithm; Algorithm design and analysis; Cloning; Computer science; Data analysis; Degradation; Optimizing compilers; Program processors; Runtime;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Languages, 1992., Proceedings of the 1992 International Conference on

Conference_Location

Oakland, CA

Print_ISBN

0-8186-2585-6

Type

conf

DOI

10.1109/ICCL.1992.185472

Filename

185472