Detecting equality of variables in programs

Alpern, Bowen; Wegman, Mark N.; Zadeck, F. Kenneth

doi:10.1145/73560.73561

Cited by 312 publications

(219 citation statements)

References 14 publications

Supporting

Mentioning

219

Contrasting

Order By: Relevance

“…The variable ε of the TDS will be set to {Write x(v 1 ), Write x(v 2 )}, for some values v 1 , v 2 , but the variable ε of the FSP will be set to Write x(v 2 ). To handle this problem, one can always translate both programs into SSA form [Alpern et al, 1988] such that in the program texts there is only one assignment to each variable. That is, the translation of the TDS into SSA form results in x1 = x0 + 1 ; x2 = x1 + 1 ; and the translation of the FSP into SSA form results in x1' = x0' + 2.…”

Section: Verimag Research Report N O Tr-2008-14mentioning

confidence: 99%

Certification of Smart-Card Applications in Common Criteria

Narasamdya¹,

Périn

2009

Fundamental Approaches to Software Engineering

View full text Add to dashboard Cite

This report describes a certification method of smart-card applications in the framework of Common Criteria. In this framework, a smart-card application is represented consecutively by a model of its specification, a functional specification describing an input-output relationship, a low-level design, and implementation code. The certification process consists of the following tasks: (1) prove that the model, the functional specification, the low-level design, and the code satisfy security properties in the smart-card application's specification, and (2) prove that there is a representation correspondence between each two consecutive representations. For each task, a certificate or a collection of certificates are needed to certify the accomplishment of the task. We describe in this report the application of a theory of program properties to the certification process. The theory provides foundations for describing and proving properties of a single program and properties relating two programs. The theory provides a notion of verification condition as a notion of certificate. The theory is applicable to the certification process because all representations of a smart-card application are essentially programs and the representation correspondences are properties relating two programs.

show abstract

Section: Verimag Research Report N O Tr-2008-14mentioning

confidence: 99%

Certification of Smart-Card Applications in Common Criteria

Narasamdya¹,

Périn

2009

Fundamental Approaches to Software Engineering

View full text Add to dashboard Cite

show abstract

“…Kildall [13] discovers these equivalences by performing an abstract interpretation [3] over the lattice of Herbrand equivalences in exponential time. There are several polynomialtime, but less precise, algorithms that are complete for basic blocks, but are imprecise in the presence of joins and loops in a program [1,20,6]. A polynomial time intraprocedural algorithm was given by Gulwani and Necula [8,9] and Müller-Olm, Rüthing, and Seidl [14].…”

Section: History Of Global Value Numberingmentioning

confidence: 99%

“…Since full reasoning about the multiplication operator is in general undecidable, we can use some sound abstraction of the multiplication operator. One option is to model the multiplication operator as a binary uninterpreted function 1 . Such a model is sufficient to prove the validity of the assertion a = b.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Assertion Checking Unified

Gulwani

Tiwari

2007

Lecture Notes in Computer Science

View full text Add to dashboard Cite

This paper establishes an interesting connection between assertion checking in programs and unification in the theory underlying the program expressions. Using this connection, we describe how unification algorithms from theorem proving can be used to perform backward analysis over programs for assertion checking. Interestingly enough, this connection also helps prove hardness results for assertion checking for classes of program abstractions. In particular, we show (a) Assertion checking is PTIME for programs with nondeterministic conditionals that use expressions from a strict unitary theory. (b) Assertion checking is coNP-hard for programs with nondeterministic conditionals that use expressions from a bitary theory. (c) Assertion checking is decidable for programs with disequality guards that use expressions from a convex finitary theory. (d) Summary computation for interprocedural analysis can be performed using backward analysis, enabled with unification, on generic assertions. This helps generalize result (a) to interprocedural analysis. These results generalize several recently published results using a uniform framework. They also provide several new results, and partially solve the long standing open problem of interprocedural global value numbering. In essence, they provide new techniques for backward analysis of programs based on novel integration of theorem proving technology in program analysis.

show abstract

“…Reordering transformations change the order in which statements are executed, either within or across loop iterations. 4 They are safe if all program dependences in the original program are preserved. Reordering transformations are used to expose or enhance loop-level parallelism.…”

Section: Transformationsmentioning

confidence: 99%

Analysis and transformation in an interactive parallel programming tool

Kennedy

McKinley

Tseng

1993

Concurrency: Pract. Exper.

View full text Add to dashboard Cite

The ParaScope Editor is a new kind of interactive parallel programming tool for developing scienti c Fortran programs. It assists the knowledgeable user by displaying the results of sophisticated program analyses and by providing editing and a set of powerful interactive transformations. After an edit or parallelism-enhancing transformation, the ParaScope Editor incrementally updates both the analyses and source quickly. This paper describes the underlying implementation of the ParaScope Editor, paying particular attention to the analysis and representation of dependence information and its reconstruction after changes to the program.

show abstract

Detecting equality of variables in programs

Cited by 312 publications

References 14 publications

Certification of Smart-Card Applications in Common Criteria

Certification of Smart-Card Applications in Common Criteria

Assertion Checking Unified

Analysis and transformation in an interactive parallel programming tool

Contact Info

Product

Resources

About