Constraint-Based Local Search for the Automatic Generation of Architectural Tests

Abstract

Abstract. This paper considers the automatic generation of architectural tests (ATGP), a fundamental problem in processor validation. ATGPs are complex conditional constraint satisfaction problems which typically feature both hard and soft constraints and very large domains (e.g., all memory addresses). Moreover, the goal is to generate a large number of diverse solutions under tight runtime constraints. To improve solution diversity, this paper proposes a novel approach to ATGPs by modeling them as MaxDiversekSet problems and solving them with constraintbased local search over conditional variables. The paper presents the semantics and implementation of conditional variables in this context and demonstrates the computational benefits of the approach.

Enriching Solutions to Combinatorial Problems via Solution Engineering

“…A number of other studies (Petit and Trapp 2015, Trapp and Konrad 2015, Camm 2014, Eiter et al 2013, Hebrard et al 2011, Danna and Woodruff 2009, Schittekat and Sörensen 2009, Hentenryck et al 2009a, Glover et al 2000, Kuo et al 1993 have demonstrated interest in exploration beyond the single solution returned by the solver. These studies discuss generating multiple solutions to combinatorial problems, and some further integrate diversity into their discussions.…”

Kangaroo: An Efficient Constraint-Based Local Search System Using Lazy Propagation

“…Problems instances were generated using the graph generator programs written by Joseph Culberson 4 with parameter values p = 0.5, k ∈ [3,10], n ∈ {25, 50, 75, 100} and 5 instances per setting. 5 The problem model uses a not-equal constraint for each pair of adjacent vertexes. It uses max/min search.…”

“…Constraint-based local search (CBLS) has been quite successful in solving problems that prove difficult for constraint solvers based on constructive search. Unfortunately, there is little published work on existing implementations that led to this success -most work has addressed language features [12,7,8] and applications [5,2]. In this paper we present Kangaroo, a new constraint-based local search system and expose key details of its implementation.…”

“…social golfer: Given w weeks, g groups, and p persons per group, the problem is to find a golf playing schedule such that each person plays every week in a group with other persons, but no two persons fall in the same group more than once. Problems instances were generated for w ∈ [3,9], g ∈ [3,5], and p ∈ [2,5]. The problem model uses atmost and exactly constraints.…”

Soft Constraints of Difference and Equality

“…The first approaches to this problem relied on heuristic methods (Hebrard et al, 2005;Hentenryck, Coffrin, & Gutkovich, 2009), It was also shown that when the problem P allows it, knowledge compilation methods could efficiently solve this problem (Hadzic, Holland, & O'Sullivan, 2009). …”