Conference paper
A Scalable Inclusion Constraint Solver Using Unification
We describe a parameterized framework with which users can take advantage of unification over analysis variables to implement efficient or precise analyses, or even both. To be illustrative we instantiate the framework with reaching definition analysis and conduct a systematic evaluation of performance and precision of the analysis.
We compare our result with that of a state-of-the-art solver, the Succinct Solver and show our solver is at least 10-times faster than the Succinct Solver. On some benchmarks linearity is reached by the use of unification. Although the result of unification is often imprecise, a heuristic study is conducted to detect where the loss of precision may happen.
We apply the heuristics on benchmarks and achieve not only efficient but also precise analysis.
Language: | English |
---|---|
Publisher: | Springer Verlag |
Year: | 2007 |
Pages: | 121-137 |
Proceedings: | A Scalable Inclusion Constraint Solver Using Unification |
Series: | Lecture Notes in Computer Science |
ISBN: | 3540787682 , 3540787690 , 9783540787686 and 9783540787693 |
Types: | Conference paper |
DOI: | 10.1007/978-3-540-78769-3_9 |
ORCIDs: | Nielson, Flemming |