arXiv subjects

Mireille Ducasse

Publications and source records attributed to Mireille Ducasse.

5 recordsLinked to original sources

arXiv2008

Design and Implementation of a Tracer Driver: Easy and Efficient Dynamic Analyses of Constraint Logic Programs

Tracers provide users with useful information about program executions. In this article, we propose a ``tracer driver''. From a single tracer, it provides a powerful front-end enabling multiple dynamic analysis tools to be easily implemented, while limiting the overhead of the trace generation. The relevant execution events are specified by flexible event patterns and a large variety of trace data can be given either systematically or ``on demand''. The proposed tracer driver has been designed in the context of constraint logic programming; experiments have been made within GNU-Prolog. Execution views provided by existing tools have been easily emulated with a negligible overhead. Experimental measures show that the flexibility and power of the described architecture lead to good performance. The tracer driver overhead is inversely proportional to the average time between two traced events. Whereas the principles of the tracer driver are independent of the traced programming language, it is best suited for high-level languages, such as constraint logic programming, where each traced execution event encompasses numerous low-level execution steps. Furthermore, constraint logic programming is especially hard to debug. The current environments do not provide all the useful dynamic analysis tools. They can significantly benefit from our tracer driver which enables dynamic analyses to be integrated at a very low cost.

cs.SE↗

arXiv2005

A Tracer Driver for Versatile Dynamic Analyses of Constraint Logic Programs

Programs with constraints are hard to debug. In this paper, we describe a general architecture to help develop new debugging tools for constraint programming. The possible tools are fed by a single general-purpose tracer. A tracer-driver is used to adapt the actual content of the trace, according to the needs of the tool. This enables the tools and the tracer to communicate in a client-server scheme. Each tool describes its needs of execution data thanks to event patterns. The tracer driver scrutinizes the execution according to these event patterns and sends only the data that are relevant to the connected tools. Experimental measures show that this approach leads to good performance in the context of constraint logic programming, where a large variety of tools exists and the trace is potentially huge.

cs.SE↗

arXiv2005

Proving or Disproving likely Invariants with Constraint Reasoning

A program invariant is a property that holds for every execution of the program. Recent work suggest to infer likely-only invariants, via dynamic analysis. A likely invariant is a property that holds for some executions but is not guaranteed to hold for all executions. In this paper, we present work in progress addressing the challenging problem of automatically verifying that likely invariants are actual invariants. We propose a constraint-based reasoning approach that is able, unlike other approaches, to both prove or disprove likely invariants. In the latter case, our approach provides counter-examples. We illustrate the approach on a motivating example where automatically generated likely invariants are verified.

cs.SE↗

arXiv2003

Rigorous design of tracers: an experiment for constraint logic programming

In order to design and implement tracers, one must decide what exactly to trace and how to produce this trace. On the one hand, trace designs are too often guided by implementation concerns and are not as useful as they should be. On the other hand, an interesting trace which cannot be produced efficiently, is not very useful either. In this article we propose a methodology which helps to efficiently produce accurate traces. Firstly, design a formal specification of the trace model. Secondly, derive a prototype tracer from this specification. Thirdly, analyze the produced traces. Fourthly, implement an efficient tracer. Lastly, compare the traces of the two tracers. At each step, problems can be found. In that case one has to iterate the process. We have successfully applied the proposed methodology to the design and implementation of a real tracer for constraint logic programming which is able to efficiently generate information required to build interesting graphical views of executions.

cs.SE↗

arXiv2001

Prototyping CLP(FD) Tracers: a Trace Model and an Experimental Validation Environment

Developing and maintaining CLP programs requires visualization and explanation tools. However, existing tools are built in an ad hoc way. Therefore porting tools from one platform to another is very difficult. We have shown in previous work that, from a fine-grained execution trace, a number of interesting views about logic program executions could be generated by trace analysis. In this article, we propose a trace model for constraint solving by narrowing. This trace model is the first one proposed for CLP(FD) and does not pretend to be the ultimate one. We also propose an instrumented meta-interpreter in order to experiment with the model. Furthermore, we show that the proposed trace model contains the necessary information to build known and useful execution views. This work sets the basis for generic execution analysis of CLP(FD) programs.

cs.PL↗