Is it a tree, a DAG, or a cyclic graph? A shape analysis for heap-directed pointers in C

Ghiya, Rakesh; Hendren, Laurie

doi:10.1145/237721.237724

Cited by 220 publications

(150 citation statements)

References 27 publications

Supporting

Mentioning

150

Contrasting

Order By: Relevance

“…Splash2 benchmarks call the Pthreads li- 6 , and mutual exclusion is enforced by pthread mutex lock(<lock var>) and pthread mutex unlock(<lock var>) functions with explicit lock variables. For the purpose of our performance study, we manually transformed each lock/unlock region into a critical section.…”

Section: Performance Study On Sun-firementioning

confidence: 99%

“…The unnecessary serialization will then increase lock contention among threads during the execution time. Some more sophisticated analysis techniques such as shape analysis [6], or dynamic conflict resolving techniques such as transactional memory and synchronization state buffer (SSB) [7] are needed to exploit further concurrency among critical sections in these benchmarks.…”

Section: Performance Study On Sun-firementioning

confidence: 99%

See 1 more Smart Citation

Minimum Lock Assignment: A Method for Exploiting Concurrency among Critical Sections

Zhang

Sreedhar

Zhu

et al. 2008

Languages and Compilers for Parallel Computing

View full text Add to dashboard Cite

Abstract. In this paper we propose a lock assignment technique to simplify the mutual exclusion enforcement in multithreaded programs. Programmers are allowed to annotate the regions of code that are expected to be mutually exclusive as critical sections, without using explicit locks. The compiler then automatically infers an assignment of the minimum number of locks to critical sections by solving the Minimum Lock Assignment (MLA) problem so as to enforce mutual exclusion without any loss of concurrency. We show that the MLA problem is NP-hard. We have proposed a heuristic to solve the MLA problem, and tested the optimality of the heuristic with the Integer Linear Programming (ILP) solver. We have also tested the efficiency of the heuristic using scientific applications, from which we obtain up to 30% performance gain with respect to the programs in which all critical sections are controlled by a single lock.

show abstract

Section: Performance Study On Sun-firementioning

confidence: 99%

Section: Performance Study On Sun-firementioning

confidence: 99%

Minimum Lock Assignment: A Method for Exploiting Concurrency among Critical Sections

Zhang

Sreedhar

Zhu

et al. 2008

Languages and Compilers for Parallel Computing

View full text Add to dashboard Cite

show abstract

“…This refinement allows their approach to handle list-reversal type of operations and doubly-linked lists. Ghyia and Hendren [5] describe an approach that sacrifices precision for time and space complexity. They describe an interprocedural dataflow analysis in which for every pair of pointer variables handles the analysis keeps track of connectivity, direction and acyclicity.…”

Section: Shape Analysismentioning

confidence: 99%

Increasing the Accuracy of Shape and Safety Analysis of Pointer-Based Codes

Diniz

2004

Languages and Compilers for Parallel Computing

View full text Add to dashboard Cite

Abstract. Analyses and transformations of programs that manipulate pointer-based data structures rely on understanding the topological relationships between the nodes ı.e., the overall shape of the data structures. Current static shape analyses either assume correctness of the code or trade-off accuracy for analysis performance, leading in most cases to shape information that is of little use for practical purposes. This paper introduces four novel analysis techniques, namely structural fields, scan loops, assumed/verified shape properties and context tracing. Analysis of structural fields allows compilers to uncover node configurations that play key roles in the data structure. Analysis of scan loops allows compilers to establish accurate relationship between variables that traverse the data structure. Assumed/verified property analysis derives sufficient shape properties that guarantee termination of loops. These properties must be verified during shape analysis for consistency. Context tracing allows the compiler to isolate data structure nodes by evaluation of conditional statements and hence preserve shape accuracy. We believe that future static shape and safety analysis algorithms will have to include some if not all of these techniques to attain a high level accuracy. In this paper we illustrate the application of the proposed techniques to codes that build (correctly as well as incorrectly) sophisticated data structures that are beyond the reach of current approaches.

show abstract

“…Shape analyses [10,20,19,2] can handle structural constraints using abstract heap representations, predicate abstraction etc. However, shape analyses typically do not consider rich properties of data values in structures and mostly abstract away from the data.…”

Section: Static Analysesmentioning

confidence: 99%

“…Existing dynamic and static analyses [19,4,8,20,2,10] that check non-trivial properties of such programs impose a substantial burden on the users, e.g., by requiring the users to provide invariants, such as loop or representation invariants, or to provide complete executable implementations as well as specifications.…”

Section: Introductionmentioning

confidence: 99%

Generating Representation Invariants of Structurally Complex Data

Malik

Pervaiz

Khurshid

Tools and Algorithms for the Construction and Analysis of Systems

View full text Add to dashboard Cite

Abstract. Generating likely invariants using dynamic analyses is becoming an increasingly effective technique in software checking methodologies. This paper presents Deryaft, a novel algorithm for generating likely representation invariants of structurally complex data. Given a small set of concrete structures, Deryaft analyzes their key characteristics to formulate local and global properties that the structures exhibit. For effective formulation of structural invariants, Deryaft focuses on graph properties, including reachability, and views the program heap as an edge-labeled graph.Deryaft outputs a Java predicate that represents the invariants; the predicate takes an input structure and returns true if and only if it satisfies the invariants. The invariants generated by Deryaft directly enable automation of various existing frameworks, such as the Korat test generation framework and the Juzi data structure repair framework, which otherwise require the user to provide the invariants. Experimental results with the Deryaft prototype show that it feasibly generates invariants for a range of subject structures, including libraries as well as a stand-alone application.

show abstract

Is it a tree, a DAG, or a cyclic graph? A shape analysis for heap-directed pointers in C

Cited by 220 publications

References 27 publications

Minimum Lock Assignment: A Method for Exploiting Concurrency among Critical Sections

Minimum Lock Assignment: A Method for Exploiting Concurrency among Critical Sections

Increasing the Accuracy of Shape and Safety Analysis of Pointer-Based Codes

Generating Representation Invariants of Structurally Complex Data

Contact Info

Product

Resources

About