Chaff: engineering an efficient SAT solver
Proceedings of the 38th annual Design Automation Conference
DATE '03 Proceedings of the conference on Design, Automation and Test in Europe - Volume 1
Predicate learning and selective theory deduction for a difference logic solver
Proceedings of the 43rd annual Design Automation Conference
Engineering DPLL(T) + Saturation
IJCAR '08 Proceedings of the 4th international joint conference on Automated Reasoning
Generalizing DPLL to Richer Logics
CAV '09 Proceedings of the 21st International Conference on Computer Aided Verification
CP'09 Proceedings of the 15th international conference on Principles and practice of constraint programming
Cutting to the Chase solving linear integer arithmetic
CADE'11 Proceedings of the 23rd international conference on Automated deduction
Numeric bounds analysis with conflict-driven learning
TACAS'12 Proceedings of the 18th international conference on Tools and Algorithms for the Construction and Analysis of Systems
Abstract conflict driven learning
POPL '13 Proceedings of the 40th annual ACM SIGPLAN-SIGACT symposium on Principles of programming languages
CAV'13 Proceedings of the 25th international conference on Computer Aided Verification
Proceedings of the 41st ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages
| Hi-index | 0.00 |
SMT solvers have traditionally been based on the DPLL(T) algorithm, where the driving force behind the procedure is a DPLL search over truth valuations. This traditional framework allows for a degree of modularity in the treatment of theory solvers. Over time, theory solvers have become more and more closely integrated into the DPLL process, and consequently less and less modular. In this paper, we present a DPLL-like algorithm for SMT solving in which the search takes place over the natural domain of the variables in the problem. As a case study, we analyze its application to continuous domain linear arithmetic, present implementation techniques and some experimentation with difference logic. Results indicate the method can sometimes outperform leading SMT solvers but that the method is not yet robust.