Simplifying Boolean constraint solving for random simulation-vector generation

Authors:
Jun Yuan;A. Aziz;C. Pixley;K. Albin
Affiliations:
Jasper Design Autom., Mountain View, CA, USA;-;-;-
Venue:
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Year:
2006

Citing 0
Cited 7

Stimulus generation for constrained random simulation

Proceedings of the 2007 IEEE/ACM international conference on Computer-aided design
Self-adjusting constrained random stimulus generation using splitting evenness evaluation and XOR constraints

Proceedings of the 2009 Asia and South Pacific Design Automation Conference
Automatic constraint generation for guided random simulation

Proceedings of the 2010 Asia and South Pacific Design Automation Conference
Toward an extremely-high-throughput and even-distribution pattern generator for the constrained random simulation techniques

Proceedings of the International Conference on Computer-Aided Design
A robust general constrained random pattern generator for constraints with variable ordering

Proceedings of the International Conference on Computer-Aided Design
A robust constraint solving framework for multiple constraint sets in constrained random verification

Proceedings of the 50th Annual Design Automation Conference
A scalable and nearly uniform generator of SAT witnesses

CAV'13 Proceedings of the 25th international conference on Computer Aided Verification

Quantified Score

Hi-index	0.03

Visualization

Abstract

Simulation by random vectors is meaningful only if the vectors meet certain requirements on the environment that drives the design under verification. When that environment is modeled by constraints, we face the problem of solving constraints efficiently. We present an efficient algorithm for simplifying conjunctive Boolean constraints defined over state and input variables, and apply it to constrained random simulation vector generation using binary decision diagrams (BDDs). The method works by extracting "hold-constraints" from the system of constraints. Hold-constraints are deterministic and trivially resolvable. They can be used to simplify the original constraints as well as refine the conjunctive partition. Experiments demonstrate significant reductions in the time and space required for constructing the conjunction BDDs, and the time spent in vector generation during simulation.