Automatic loop parallelization: an abstract interpretation approach

Abstract

This paper presents an abstract interpretation to support the automatic parallelization of iterative constructs. The analysis approximates the range of values of any array subscript by integrating an interval analysis with one returning a set of simple linear restraints among program variables. To reduce the complexity of the analysis, a general methodology has been adopted that defines two different abstract domains, i.e. the interval domain and the bisector domain. The domains are then combined through the reduced product operator A static tool implementing the analysis has been automatically generated through PAG. The results of the analysis of kernel loops of the Perfect Club Benchmark are shown.