Theory and Applications of Satisfiability Testing -- SAT 2015: 18th International Conference, Austin, TX, USA, September 24-27, 2015, Proceedings

457 Pages·2015·14.299 MB·English

by  Marijn Heule, Sean Weaver

Marijn Heule
Sean Weaver (Eds.)
Theory and Applications
0
4
of Satisfiability Testing –
3
9

S
C SAT 2015
N
L
18th International Conference
Austin, TX, USA, September 24–27, 2015
Proceedings
123
Lecture Notes in Computer Science 9340
Commenced Publication in 1973
Founding and Former Series Editors:
Gerhard Goos, Juris Hartmanis, and Jan van Leeuwen
Editorial Board
David Hutchison
Lancaster University, Lancaster, UK
Takeo Kanade
Carnegie Mellon University, Pittsburgh, PA, USA
Josef Kittler
University of Surrey, Guildford, UK
Jon M. Kleinberg
Cornell University, Ithaca, NY, USA
Friedemann Mattern
ETH Zurich, Zürich, Switzerland
John C. Mitchell
Stanford University, Stanford, CA, USA
Moni Naor
Weizmann Institute of Science, Rehovot, Israel
C. Pandu Rangan
Indian Institute of Technology, Madras, India
Bernhard Steffen
TU Dortmund University, Dortmund, Germany
Demetri Terzopoulos
University of California, Los Angeles, CA, USA
Doug Tygar
University of California, Berkeley, CA, USA
Gerhard Weikum
Max Planck Institute for Informatics, Saarbrücken, Germany
More information about this series at http://www.springer.com/series/7407
Marijn Heule Sean Weaver (Eds.)
(cid:129)
Theory and Applications
fi –
of Satis ability Testing
SAT 2015
18th International Conference
–
Austin, TX, USA, September 24 27, 2015
Proceedings
123
Editors
Marijn Heule SeanWeaver
University of Texas TrustedSystems Research Group
Austin, TX Fort Meade, MD
USA USA
ISSN 0302-9743 ISSN 1611-3349 (electronic)
Lecture Notesin Computer Science
ISBN 978-3-319-24317-7 ISBN978-3-319-24318-4 (eBook)
DOI 10.1007/978-3-319-24318-4
LibraryofCongressControlNumber:2015948854
LNCSSublibrary:SL1–TheoreticalComputerScienceandGeneralIssues
SpringerChamHeidelbergNewYorkDordrechtLondon
©SpringerInternationalPublishingSwitzerland2015
Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpartofthe
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodologynow
knownorhereafterdeveloped.
Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication
doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
Thepublisher,theauthorsandtheeditorsaresafetoassumethattheadviceandinformationinthisbookare
believedtobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsortheeditors
give a warranty, express or implied, with respect to the material contained herein or for any errors or
omissionsthatmayhavebeenmade.
Printedonacid-freepaper
SpringerInternationalPublishingAGSwitzerlandispartofSpringerScience+BusinessMedia
(www.springer.com)
Preface
This volume contains the papers presented at the 18th International Conference on
Theory and Applications of Satisfiability Testing (SAT 2015), held during September
24–27,2015inAustin,Texas,USA.SAT2015wascolocatedwithFormalMethodsin
Computer-AidedDesign(FMCAD2015)andwashostedbytheUniversityofTexasat
Austin.
The International Conference on Theory and Applications of Satisfiability Testing
(SAT) is the primary annual meeting for researchers focusing on the theory and
applications of the propositional satisfiability problem, broadly construed: Besides
plain propositional satisfiability, it includes Boolean optimization (including MaxSAT
and Pseudo-Boolean (PB), constraints), Quantified Boolean Formulas (QBF), Satisfi-
ability Modulo Theories (SMT), and Constraint Programming (CP) for problems with
clear connections to propositional reasoning. Many hard combinatorial problems can
be tackled using SAT-based techniques, including problems that arise in formal veri-
fication, artificial intelligence, operations research, biology, cryptology, data mining,
machine learning, mathematics, etc. Indeed, the theoretical and practical advances in
SAT research over the past 20 years have contributed to making SAT technology an
indispensable tool in various domains.
SAT 2015 welcomed scientific contributions addressing different aspects of SAT,
including(butnotrestrictedto)theoreticaladvances(includingexactalgorithms,proof
complexity, and other complexity issues), practical search algorithms, knowledge
compilation, implementation-level details of SAT solvers and SAT-based systems,
problemencodingsandreformulations,applications,aswellascasestudiesandreports
on insightful findings based on rigorous experimentation.
Atotalof70papersweresubmittedtoSAT2015,distributedinto44regularpapers
(up to 15 pages excluding references), 17 short papers (up to eight pages excluding
references), and nine tool papers (up to six pages excluding references). In contrast to
recent SAT conferences, no paper submission was found to be out of scope for the
conference. All 70 submissions were assigned for review to at least four Program
Committee members and their selected external reviewers. Continuing the procedure
initiated in SAT 2012, the review process included an author-response period, during
which the authors of submitted papers were given the opportunity to respond to the
initial reviews for their submissions. For reaching final decisions, a Program Com-
mittee discussion period followed the author-response period. This year, external
reviewerssupportingtheProgramCommitteewerealsoinvitedtoparticipatedirectlyin
the discussions for the papers they reviewed. In the end, the Program Committee
decidedtoaccept21regularpapers,twoshortpapers,andseventoolpapers.Twoshort
papers were downgraded to tool papers.
In addition to presentations on the accepted papers, the scientific program of SAT
2015 included three invited talks:
VI Preface
– Dimitris Achlioptas (University of California Santa Cruz, USA)
Random Formulas are Irrelevant, Right?
– Anna Slobodova (Centaur Technology, USA)
Pragmatic Approach to Formal Verification
– Aaron Tomb (Galois, Inc., USA)
Applying Satisfiability to the Analysis of Cryptography
SAT 2015 hosted various affiliated events, including two workshops
on September 23:
– Sixth International Workshop on Pragmatics of SAT (PoS 2015)
Organizers: Daniel Le Berre and Allen Van Gelder;
– Third International Workshop on Quantified Boolean Formulas (QBF 2015)
Organizers: Florian Lonsing and Martina Seidl;
and three competitions and system evaluations:
– SAT Race 2015
Organizers: Tomas Balyo, Carsten Sinz, and Markus Iser;
– Max-SAT Evaluation 2015
Organizers: Josep Argelich, Chu-Min Li, Felip Manya, and Jordi Planes;
– Pseudo-Boolean Evaluation 2015
Organizers: Norbert Manthey and Peter Steinke
We would like to thank everyone who contributed to making SAT 2015 a success.
FirstandforemostwewouldliketothankthemembersoftheProgramCommitteeand
theadditionalexternalreviewersfortheircareful andthoroughwork,withoutwhichit
would not have been possible for us to put together such an outstanding conference
program. We also wish to thank all the authors who submitted their work for our
consideration. We thank the SAT Association chair Armin Biere, vice chair John
Franco, and treasurer Hans Kleine Büning for their help and advice in organizational
matters. We wish to thank the workshop chair Albert Oliveras. The EasyChair con-
ference system provided invaluable assistance in coordinating the submission and
reviewprocess,aswellasintheassemblyoftheseproceedings.Wealsothankthelocal
organization team for their efforts with practical aspects of local organization.
Finally,wegratefullythanktheUniversityofTexasatAustin,theSATAssociation,
the Artificial Intelligence journal, CyberPoint, Galois, Inc., Intel, and Microsoft
Research for financial and organizational support for SAT 2015.
July 2015 Marijn Heule
Sean Weaver
Organization
Program Committee
Fahiem Bacchus University of Toronto, Canada
Olaf Beyersdorff University of Leeds, UK
Armin Biere Johannes Kepler University, Austria
Leonardo De Moura Microsoft Research, USA
Uwe Egly Vienna University of Technology, Austria
John Franco University of Cincinnati, USA
Enrico Giunchiglia DIST - University of Genova, Italy
Youssef Hamadi Microsoft Research, UK
Marijn Heule The University of Texas at Austin, USA
Holger Hoos University of British Columbia, Canada
Alexander Ivrii IBM, Israel
Jie-Hong Roland Jiang National Taiwan University, Taiwan
Matti Järvisalo University of Helsinki, Finland
Oliver Kullmann Swansea University, UK
Daniel Le Berre CNRS - Université d’Artois, France
Ines Lynce INESC-ID/IST, University of Lisbon, Portugal
Sharad Malik Princeton University, USA
Panagiotis Manolios Northeastern University, USA
Norbert Manthey TU Dresden, Germany
Joao Marques-Silva University College Dublin, Ireland
Alexander Nadel Intel, Israel
Nina Narodytska Samsung Research America, USA
Jakob Nordström KTH Royal Institute of Technology, Sweden
Albert Oliveras Technical University of Catalonia, Spain
Karem Sakallah University of Michigan, USA
Roberto Sebastiani DISI, University of Trento, Italy
Martina Seidl Johannes Kepler University Linz, Austria
Bart Selman Cornell University, USA
Laurent Simon Labri, Bordeaux Institute of Technology, France
Carsten Sinz Karlsruhe Institute of Technology, Germany
Stefan Szeider Vienna University of Technology, Austria
Sean Weaver Department of Defense, USA
Xishun Zhao Institute of Logic and Cognition, Sun Yat-Sen
University, China
VIII Organization
Additional Reviewers
Abío, Ignasi Lonsing, Florian
Aleksandrowicz, Gadi López-Ibáñez, Manuel
Arbel, Eli Manquinho, Vasco
Audemard, Gilles Marek, Victor
Balabanov, Valeriy Martins, Ruben
Balyo, Tomáš Mencía, Carlos
Bapst, Victor Miksa, Mladen
Bayless, Sam Morgado, Antonio
Ben-Haim, Yael Nevo, Ziv
Berkholz, Christoph Nieuwenhuis, Robert
Bingham, Jesse Oetsch, Johannes
Bova, Simone Oikarinen, Emilia
Cao, Weiwei Ordyniak, Sebastian
Chamarthi, Harsh Raju Previti, Alessandro
Creignou, Nadia Ray, Sayak
Diller, Martin de Rezende, Susanna F.
Elffers, Jan Ryvchin, Vadim
Fawcett, Chris Shen, Yuping
Fröhlich, Andreas Slivovsky, Friedrich
Ganian, Robert Steinke, Peter
de Haan, Ronald Strichman, Ofer
Ignatiev, Alexey Subramanyan, Pramod
Jain, Mitesh Trentin, Patrick
Janota, Mikolas Tu, Kuan-Hua
Kim, Eun Jung Van den Broeck, Guy
Kotthoff, Lars Vinyals, Marc
Krakovski, Roi Vizel, Yakir
Lagniez, Jean-Marie Wang, Hung-En
Lauria, Massimo Wetzler, Nathan
Lee, Nian-Ze Widl, Magdalena
Li, Chu-Min Yamada, Akihisa
Lonca, Emmanuel Yue, Weiya
Pragmatic Approach to Formal Verification
Anna Slobodova
CentaurTechnology, Taipet, Taiwan
[email protected]
After more than two decades of hard work by researchers in academia and industry,
formal methods have been accepted as a viable part of the hardware design and vali-
dation process. We now have a better understanding of what is a cost-effective use of
formal methods, and companies even set aside some resources for the further devel-
opment offormal tools. Spreading formal methods into industrial scale software ver-
ification broadened the user population and increased motivation for development of
such tools.
Centaur Technology is one of the companies that adopted formal verification
(FV)asapartoftheirproductionflow.OurcompanydesignsIntelcompatible x86-64
microprocessors. It does it with a relatively small team. To assure the quality of the
design, a lot of effort is spent in the process of validation. Since new additions to
x86-64 architecture widened the data on which instructions are performed, classic
simulationprovidesevenlesscoveragewithrespecttoallpossibleinputstothesystem
thanafewyearsago.TheFVteamatCentaurTechnologywascreatedasareactionto
thistrendaswellasthefactthatthecapacityofformaltoolshasreachedalevelwhere
they can be successful even on industrial scale designs. There are also more publicly
available off-shelf formal point tools (SAT, SMT, Model-checkers, etc.) that can be
incorporated into more complex validation framework. A pilot project [1] that dis-
coveredacornercasebuginfloating-pointarithmeticwasconvincingenoughtojustify
investing in a small FV team.
Our verification framework is built on top of the ACL2 theorem prover [2]. There
aremanydecisionproceduresbuiltinthelogicofACL2andprovedcorrectwithinthis
logic. For example, packages exist, defined inside the logic, for Binary Decision
Diagrams (BDD) [3] and for And-Inverter Graph (AIG) manipulation [4]. It has a
symbolic simulator called GL [5] that can automate the proof of theorems over finite
domains. GL can be combined with word-level symbolic simulation of a hardware
model to relate that model to its specification.
While we strive for rigorous analysis, our verification approach is very pragmatic.
Weconnectedsome“trusted”toolstoACL2,forinstancevariousSatisfiabilitysolvers
(e.g.,Glucose,Penelope,Lingeling,Riss3G),ortheABCmodel/equivalencechecking
tool [6]. The results from these tools are tagged as “trusted” (unverified) by ACL2.
However, for tools that provide a proof trace, in some cases, we can verify the cor-
rectness of those results within ACL2. We prefer this approach to blindly trusting the
tools. However, when we exhaust approaches that have verifiable results, we are
willing to use unverified methods as well.
Our team has worked on a variety of design and verification problems [7, 8]
including microcode verification [9] and transistor-level validation. Our main focus