M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proc. ACM symposium on Principles of Programming Languages, pages 1--14, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....however, because the problem of computing precise reference information is undecidable. Mock et al. 10] used dynamic points to information, collected by executing C programs, to evaluate the precision of the static pointsto information computed by several existing points to analysis algorithms [4, 13, 14]. They reported that, for the programs they studied, static points to information may contain a significant amount of spurious information. Although these studies give insight into the precision of static points to information for several existing algorithms, they are limited to evaluation of ....

M. Shapiro and S. Horwitz. Fast and accurate flowinsensitive points-to analysis. In ACM, editor, Conference record of POPL '97, the 24th ACM SIGPLANSIGACT Symposium on Principles of Programming Languages: papers presented at the symposium, pages 1--14, 1997.

.... are similar to those used in Steensgaard s pointer analysis for C [43] the inclusion constraints are similar to those used by Andersen s pointer analysis for C [3] Precision differences between these constraint formulations for C pointer analysis were explained by Shapiro and Horwitz [37]. Considering the pointer assignment statement p = q, the unification analysis will union the points to set of p with the points to set of q, effectively saying both pointer variables can point to the same set of objects after this assignment; this union is accomplished recursively, so that if p ....

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Conference Record of the Twenty-fourth Annual ACM SIGACT/SIGPLAN Symposium on Principles of Programming Languages, pages 1--14, January 1997.

....to e.g. the analysis can determine which are the locations that p may point to at line 1. In addition, pointer analysis determines which function addresses may be stored in a given function pointer. Because of the importance of such information, a variety of pointer analyses have been developed [11, 12, 9, 5, 1, 22, 20, 25, 18, 13, 7, 4, 6, 3]. These analyses provide different tradeoffs between cost and precision. For example, flow and context insensitive pointer analyses [1, 20, 25, 18, 4] ignore the flow of control between program points and do not distinguish between different calling contexts of procedures. As a result, such ....

....function pointer. Because of the importance of such information, a variety of pointer analyses have been developed [11, 12, 9, 5, 1, 22, 20, 25, 18, 13, 7, 4, 6, 3] These analyses provide different tradeoffs between cost and precision. For example, flow and context insensitive pointer analyses [1, 20, 25, 18, 4] ignore the flow of control between program points and do not distinguish between different calling contexts of procedures. As a result, such analyses are relatively inexpensive and imprecise. In contrast, analyses with some degree of flow or context sensitivity are typically more expensive and ....

[Article contains additional citation context not shown here]

M. Shapiro and S. Horwitz. Fast and accurate flowinsensitive points-to analysis. In Symposium on Principles of Programming Languages, pages 1--14, 1997.

....of the functions in the program, and presents a graphical view of which functions call which other functions [5, 6] Such a view is useful in trying to anticipate how a change to one function may affect the rest of the system. Recent work has focused on the efficiency [7, 8, 9] and precision [10] of such tools. In order to compute correct information, an analysis tool often needs to analyze the entire program. However, the source code for libraries is often not available to the analysis tool. Today s programs make extensive use of library functions. These functions provide mechanisms for ....

M. Shapiro and S. Horwitz. Fast and accurate flowinsensitive points-to analysis. In Proceedings of the 24th ACM Symposium on Principles of Programming Languages, pages 1--14, Paris, France, January 1997.

....source level programs. For instance, whole program optimizations may open up opportunities for moving invariant memory instructions out of loops. However, alias information is key to identifying such opportunities. While there is an extensive body of work on pointer alias analysis of various kinds [30, 27, 12, 8, 16], these are mostly high level analyses carried out in terms of source language constructs. Unfortunately, such analyses turn out to be of limited utility at the machine code level. In fact, the problem of memory disambiguation is one of the weak points of object code modification, because high ....

....number of systems have been described for optimization of executable code [28, 17, 9, 23] to the best of our knowledge, alias analysis carried out by these systems is limited to fairly simple local analysis. On the other hand, there is an extensive work on pointer alias analysis of various kinds [30, 27, 12, 8, 16]. However, in almost all cases, these are high level analyses carried out in terms of source language constructs that ignore features typically encountered in executable programs. As a result, such analyses are of limited utility at the machine code level. Amme et al. 2] present a general method ....

M. Shapiro and S. Horwitz. Fast and accurate flowinsensitive points-to analysis. In The 24th ACM SIGPLANSIGACT Symposium on Principles of Programming Languages, pages 1--14, Jan. 1997.

....refers to the memory location which is also referred to by another expression) resulting from pointer variables might generate implicit DD relations. In order to analyze such relations, pointer analysis should be needed. Many researchers have already proposed static pointer analysis methods[8, 17, 18]; however, it is difficult for static analysis to generate practical analysis results of the pointer alias problem. On the other hand, dynamic slicing is based on a single execution path, and it can give narrower slices than static slices. However, the dynamic slicing requires a large amount of ....

Shapiro, M. and Horwitz, S.: "Fast and accurate flowinsensitive point-to analysis", Proceedings of the 24th ACM SIGPLAN-SIGACT symposium on Principles of programming languages, pp.1-14, Paris, France, January (1997).

....pointer targets is computed before profiling by a static points to analysis. By default, Tumi uses the One Level Flow algorithm [Das00] OLF) for its points to analysis; other implemented choices are Steensgaard s [Ste96b] algorithm, or an extension thereof proposed by Shapiro and Horwitz [SH97b] These algorithms create one distinct compile time representative for each memory allocation site, i.e. data structures allocated at different allocation sites are distinguished, while data structures allocated at the same malloc site are not. Consequently, multiple run time addresses may ....

....could potentially result in annotations with a higher overall speedup. To enable scaling Calpa to larger applications, I implemented three different flow and contextinsensitive pointer analysis algorithms: Steensgaard s algorithm [Ste96b] an extension thereof proposed by Shapiro and Horwitz [SH97b] and a recent improvement to Steensgaard s algorithm (called One Level Flow (OLF) proposed by Manuvir Das [Das00] These algorithms compute for each variable and expression in the program the sets of objects 62 (variables, heap allocated data structures, or procedures) the variable may point ....

[Article contains additional citation context not shown here]

Marc Shapiro and Susan Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proceedings of the 24th ACM Symposium on Principles of Programming Languages, pages 1--14, Paris, France, January 1997.

....of instances of general purpose classes and excessive detail associated with failing to abstract internal data structures. 6.2 Pointer Analysis Pointer analysis has been an active area of research for well over 15 years. Approaches range from e#cient flowand context insensitive approaches [3, 25, 24, 21, 14, 10, 15] to potentially more precise but less e#cient flow and context sensitive approaches [22, 27, 13, 7, 18, 23] These approaches vary in whether they create a result for each program point (flow sensitive analyses) or one result for the entire program (flow insensitive analyses) They also vary in ....

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proceedings of the 24th Annual ACM Symposium on the Principles of Programming Languages, Paris, France, Jan. 1997.

....beneficial for large programs that cannot be easily understood without the use of tools. Therefore, existing slicers typically use pointer analyses that trade precision for better performance to enable slicing of complex programs. These analyses are typically not fully flow or context sensitive [9, 28, 30, 34]. The pointer imprecision problem is particularly severe for C programs, which use pointers extensively to simulate call by reference semantics, to emulate object oriented dispatch via function pointers, to avoid the expensive copying of large objects, to implement list, tree, or other complex ....

....increasing context sensitivity of the points to analysis, and in our context sensitive slice experiments we found in fact only small improvement of slice size due to context sensitivity. In recent years, much work has been done to improve the precision and efficiency of pointer analyses. Horwitz [28] and Das [9] provide algorithms to improve the precision of points to analyses, the latter achieving the precision of Andersen s [2] algorithm while running almost as fast as Steensgaard s algorithm [30] Rountev and Chandra [25] use a technique called variable substitution to replace a set of ....

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proceedings of the 24th ACM Symposium on Principles of Programming Languages, pages 1--14, Paris, France, Jan. 1997.

....Java is no exception. This means that, to determine whether an object must be serialized, we need to keep track of how all of its aliases are used. Fortunately, the aliasing problem is a well known and thoroughly studied one in program analysis, and a number of suitable techniques exist (e.g. [11, 3, 22, 19, 12, 13]) Moreover, alias analysis is orthogonal to the type based analysis we describe here, and the two can be combined straightforwardly as follows. First, we can exploit the results of alias analysis to annotate each node of the control flow graph with the alias set associated to each parameter ....

M. Shapiro and S. Horwitz. Fast and Accurate Flow-Insensitive Points-To Analysis. In Proc. of the Symp. on Principles of Programming Languages, pages 1--14, 1997.

....from above, pointer analysis can determine what are the locations that p may point to. In addition, pointer analysis determines which function addresses may be stored in a given function pointer. Because of the importance of such points to information, a variety of analyses have been developed [10, 9, 5, 1, 18, 17, 20, 15, 11, 7, 4, 6, 3, 14, 8]. These analyses provide different tradeoffs between cost and precision. For example, flow and context insensitive pointer analyses [1, 17, 20, 15, 4] ignore the flow of control between program points and do not distinguish between different calling contexts of procedures. As a result, such ....

....pointer. Because of the importance of such points to information, a variety of analyses have been developed [10, 9, 5, 1, 18, 17, 20, 15, 11, 7, 4, 6, 3, 14, 8] These analyses provide different tradeoffs between cost and precision. For example, flow and context insensitive pointer analyses [1, 17, 20, 15, 4] ignore the flow of control between program points and do not distinguish between different calling contexts of procedures. As a result, such analyses are relatively inexpensive and imprecise. In contrast, analyses with some degree of flow or context sensitivity are typically more expensive and ....

[Article contains additional citation context not shown here]

M. Shapiro and S. Horwitz. Fast and accurate flowinsensitive points-to analysis. In Symposium on Principles of Programming Languages, pages 1--14, 1997.

....Java. Although we demonstrate our technique on Andersen s analysis, parameterized object sensitivity can be trivially applied to enhance the precision of other flow and context insensitive analyses for Java (e.g. analyses that are based on flow and context insensitive points to analyses for C [25, 24, 8]) Modification side e#ect analysis (MOD) determines, for each statement, the set of objects that may be modified by that statement. Similarly, USE analysis computes the set of objects that may be read by a statement. This information plays an important role in optimizing compilers and software ....

....context o. Our object sensitive analysis is based on Andersen s analysis for Java from Section 2.1. However, the same approach can be trivially applied to other flow and context insensitive analyses for Java (e.g. analyses derived from flow and context insensitive points to analyses for C [25, 24, 8]) Section 3.1 defines the semantics of the object sensitive analysis. Section 3.2 discusses why object sensitivity is appropriate for flow insensitive analysis of object oriented programs, and compares this approach with other context sensitive analyses. Our object sensitive analysis is defined ....

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Symposium on Principles of Programming Languages, pages 1--14, 1997.

No context found.

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proc. ACM symposium on Principles of Programming Languages, pages 1--14, 1997.

No context found.

Shapiro, M., and Horwitz, S.: 1997, `Fast and Accurate Flow-insensitive Points-to Analysis'. In: Symposium on Principles of Programming Languages. pp. 1--14.

No context found.

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Symposium on Principles of Programming Languages, pages 1--14, 1997.

No context found.

Marc Shapiro and Susan Horwitz. Fast and Accurate Flow-Insensitive Points-To Analysis. In Conference Record of the 24th ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages, pages 1--14, January 1997. 132

No context found.

Marc Shapiro II and Susan Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proceedings of the ACM SIGPLAN/SIGACT Symposium on Principles of Programming Languages, 1997.

No context found.

Marc Shapiro II and Susan Horwitz. Fast and accurate flow-insensitive points-to analysis. In POPL, 1997.

No context found.

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Symposium on Principles of Programming Languages, pages 1--14, 1997.

No context found.

Marc Shapiro and Susan Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proceedings of POPL'97, pages 1--14, Paris, France, Jan 1997.

No context found.

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In ACM Symposium on Principles of Programming Languages, pages 1--14, 1997.

No context found.

M. Shapiro and S. Horwitz. Fast and accurate flowinsensitive points-to analysis. In The 24th ACM SIGPLANSIGACT Symposium on Principles of Programming Languages, pages 1--14, Jan. 1997.

No context found.

Marc Shapiro and Susan Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proceedings of POPL'97, pages 1--14, Paris, France, Jan 1997.

No context found.

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Proc. ACM symposium on Principles of Programming Languages, pages 1--14, 1997.

No context found.

M. Shapiro and S. Horwitz. Fast and accurate flow-insensitive points-to analysis. In Symposium on Principles of Programming Languages, pages 1--14, 1997.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC