The focus of this research is on interprocedural program analysis; the specific goal of this project is to quantify the tradeoffs between precision and compilation speed for a number of interprocedural analyses. The initial work on this project has focused on implementing a flow-sensitive interprocedural pointer alias analysis. The current activities on this project are aimed in the following three directions. (1) A flow-insensitive interprocedural pointer alias algorithm will be implemented. The tradeoffs in precision and efficiency between this algorithm and the flow- sensitive variant will be quantified. (2) A sparse representation of alias information will be developed that avoids storing redundant information at the cost of retrieving it during an alias query. (3) Experiments will be conducted to demonstrate the effectiveness of this representation. A natural extension of a flow-sensitive pointer alias analysis for C++ type analysis will be implemented. Experiments will be conducted to demonstrate its effectiveness and compare it to previous work. ***
