Transformation of meta-information by abstract Co- interpretation

Abstract

In this paper we present an approximation method based on abstract interpretation to transform meta-information in parallel with the transformation of concrete data. The meta-information is assumed to describe further properties of the specific data. The construction of a correct transformation function for the meta-information can be quite complicated in case of complex data transformations or data structures. A special approximation method is presented that works with data abstraction.Performing worst-case execution time (WCET) analysis for optimized code is described as a concrete example for the application of this approach. A transformation framework is constructed to correctly update the flow information in case of code transformations.