A Symbolic Approach to Bernstein Expansion for Program Analysis and Optimization

Several mathematical frameworks for static analysis of programs have been developed in the past decades. Although these tools are quite useful, they have still many limitations. In particular, integer multi-variate polynomials arise in many situations while analyzing programs, and analysis systems are unable to handle such expressions. Although some dedicated methods have already been proposed, they only handle some subsets of such expressions. This paper presents an original and general approach to Bernstein expansion which is symbolic. Bernstein expansion allows bounding the range of a multivariate polynomial over a box and is generally more accurate than classic interval methods.

[1]  D. K. Arvind,et al.  Detection of Concurrency-Related Errors in Joyce , 1992, CONPAR.

[2]  Rida T. Farouki,et al.  On the numerical condition of polynomials in Bernstein form , 1987, Comput. Aided Geom. Des..

[3]  Adrian Bowyer,et al.  Robust arithmetic for multivariate Bernstein-form polynomials , 2000, Comput. Aided Des..

[4]  J. Garloff,et al.  Solving strict polynomial inequalities by Bernstein expansion , 1999 .

[5]  Fujio Yamaguchi,et al.  Curves and Surfaces in Computer Aided Geometric Design , 1988, Springer Berlin Heidelberg.

[6]  Rudolf Eigenmann,et al.  Nonlinear and Symbolic Data Dependence Testing , 1998, IEEE Trans. Parallel Distributed Syst..

[7]  Benoît Meister,et al.  Automatic memory layout transformations to optimize spatial locality in parameterized loop nests , 2000, CARN.

[8]  Sebastian Pop.  Analysis of induction variables using chains of recurrences: exten-sions , 2003 .

[9]  Ralph R. Martin,et al.  Comparison of interval methods for plotting algebraic curves , 2002, Comput. Aided Geom. Des..

[10]  J. Delgado,et al.  A linear complexity algorithm for the Bernstein basis , 2003, 2003 International Conference on Geometric Modeling and Graphics, 2003. Proceedings.

[11]  Kyle A. Gallivan,et al.  Tight Timing Estimation With the Newton-Gregory Formulae∗ , 2002 .

[12]  William Pugh,et al.  Simplifying Polynominal Constraints Over Integers to Make Dependence Analysis More Precise , 1994, CONPAR.

[13]  Rudolf Eigenmann,et al.  Symbolic range propagation , 1995, Proceedings of 9th International Parallel Processing Symposium.

[14]  Jürgen Garloff,et al.  Application of Bernstein Expansion to the Solution of Control Problems , 2000, Reliab. Comput..