Better Debugging via Output Tracing and Callstack-Sensitive Slicing

Horwitz, Susan; Liblit, Ben; Polishchuk, Marina

doi:10.1109/tse.2009.66

Cited by 29 publications

(23 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For test cases where core dumps were already produced, we used the generated core file. If a test case produced bad output without crashing, we used the output tracing tool of Horwitz et al [18] to identify the first character of incorrect output, and forced the application to abort at that point. We aggregated results by taking arithmetic means across all failing tests of each faulty build, then across all faulty builds of each version.…”

Section: B Analysis Effectivenessmentioning

confidence: 99%

See 1 more Smart Citation

Lightweight control-flow instrumentation and postmortem analysis in support of debugging

Ohmann

Liblit

2016

Autom Softw Eng

View full text Add to dashboard Cite

Abstract-Debugging is difficult and costly. As a human programmer looks for a bug, it would be helpful to see a complete trace of events leading to the point of failure. Unfortunately, full tracing is simply too slow to use in deployment, and may even be impractical during testing.We aid post-deployment debugging by giving programmers additional information about program activity shortly before failure. We use latent information in post-failure memory dumps, augmented by low-overhead, tunable run-time tracing. Our results with a realistically-tuned tracing scheme show low enough overhead (0-5%) to be used in production runs. We demonstrate several potential uses of this enhanced information, including a novel postmortem static slice restriction technique and a reduced view of potentially-executed code. Experimental evaluation shows our approach to be very effective, such as shrinking stack-sensitive interprocedural static slices by 49-78% in larger applications.

show abstract

Section: B Analysis Effectivenessmentioning

confidence: 99%

“…We compute interprocedural static slices backward from the crash point in the innermost stack frame; intraprocedural slices work backward from the crash point in each function in the crash stack. All interprocedural slices are callstack-sensitive [9,18,24].…”

Section: B Analysis Effectivenessmentioning

confidence: 99%

Lightweight control-flow instrumentation and postmortem analysis in support of debugging

Ohmann

Liblit

2016

Autom Softw Eng

View full text Add to dashboard Cite

show abstract

“…For the Siemens suite, gzip, space, and flex, the experiments use slices taken from (PDG-vertex, call-stack) pairs at which symptoms of runtime errors have been observed (i.e., unexpected output or crashes). Those configurations were obtained using the debugging tools described by Horwitz et al [2010]. For wc and go, slices were taken with respect to all of the calling contexts of printf in the unrolled SDG (by using an automaton that expressed this slicing criterion).…”

Section: Specialization-slicing Experimentsmentioning

confidence: 99%

Specialization Slicing

Aung

Horwitz

Joiner

et al. 2014

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

This paper defines a new variant of program slicing, called specialization slicing, and presents an algorithm for the specialization-slicing problem that creates an optimal output slice. An algorithm for specialization slicing is polyvariant: for a given procedure p, the algorithm may create multiple specialized copies of p. In creating specialized procedures, the algorithm must decide for which patterns of formal parameters a given procedure should be specialized, and which program elements should be included in each specialized procedure.We formalize the specialization-slicing problem as a partitioning problem on the elements of the possibly-infinite unrolled program. To manipulate possibly-infinite sets of program elements, the algorithm makes use of automata-theoretic techniques originally developed in the model-checking community. The algorithm returns a finite answer that is optimal. In particular, (i) each element replicated by the specialization-slicing algorithm provides information about specialized patterns of program behavior that are intrinsic to the program, and (ii) the answer is of minimal size (i.e., among all possible answers with property (i), there is no smaller one).The specialization-slicing algorithm provides a new way to create executable slices. Moreover, by combining specialization slicing with forward slicing, we obtain a method for removing unwanted features from a program. While it was previously known how to solve the feature-removal problem for single-procedure programs, it was not known how to solve it for programs with procedure calls.

show abstract

“…Essa técnica tende a produzir históricos de acompanhamento muito extensos, o que em geral é eficaz durante os testes, pois fornece informações sobre a execução do sistema. Entretanto, durante o uso, apresenta um impacto considerável de desempenho (Mirgorodskiy et al, 2006;Horwitz et al, 2010) e gera um volume de dados muito grande, sendo poucos destes de interesse para fins de avaliação, pois dados são gerados, sistematicamente, mesmo se nada tiverem a ver com as falhas observadas. Isso decorre do fato que, em tempo de uso, a instrumentação deve gerar dados a priori para potenciais falhas sempre desconhecidas.…”

Section: Lista De Figurasunclassified

Um Mecanismo Baseado Em Logs Com Meta-Informações Para a Verificação De Contratos Em Sistemas Distribuídos

ROCHA¹

View full text Add to dashboard Cite

Ao meu orientador Prof Arndt von Staa, pelo apoio, incentivo e profissionalismo presentes durante o desenvolvimento da pesquisa.Ao CNPq e à PUC-Rio, pelos auxílios concedidos, sem os quais este trabalho não poderia ter sido realizado. Aos professores que participaram da Comissão examinadora.A todos os professores e funcionários do Departamento pelos ensinamentos e pela ajuda.À AevoTech e seus funcionários pelo apoio e por tornar possível a aplicação deste trabalho ao sistema de software do Projeto de um Robô para Monitoramento Ambiental (RoMA).Ao meu grande amigo, Thiago Araújo, por me manter no rumo em momentos de tormenta, incentivar e motivar nos momentos mais difíceis.À minha mãe e minha avó, Alice e Nadir, pelo amor, dedicação e confiança sem os quais não seria possível vencer mais este desafio.Ao meu pai, irmão, irmã e avós, Alexandre, Vinicius, Daniela, Judith e Valdir, por terem acreditado na minha capacidade. Software contracts can be written as assertions that identify failures observed while using the software. Software contracts can be implemented through executable assertions. However, conventional assertions are not directly applicable in distributed systems, as they present difficulties to evaluate temporal expressions, as well as expressions involving properties of different processes.This work proposes a mechanism based on logs with meta-information to evaluate contracts in distributed systems. A grammar to write contracts enable temporal operations, e.g., allows specifying conditions between events at different timestamps, or even guaranteeing a sequence of events over a period of time. The flow of events is evaluated asynchronously in relation to the system execution, by comparison with contracts, previously written according to the grammar, representing the expectations on the behavior of the system.

show abstract

Better Debugging via Output Tracing and Callstack-Sensitive Slicing

Cited by 29 publications

References 31 publications

Lightweight control-flow instrumentation and postmortem analysis in support of debugging

Lightweight control-flow instrumentation and postmortem analysis in support of debugging

Specialization Slicing

Um Mecanismo Baseado Em Logs Com Meta-Informações Para a Verificação De Contratos Em Sistemas Distribuídos

Contact Info

Product

Resources

About