M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In Program Flow Analysis: Theory and Applications, chapter 7, pages 189-234. Prentice-Hall, 1981.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and memory elements: Every individual v represents either a heap cell in which case heap S (v) 1, or an activation record, in which case stack S (v) 1. The unary predicate cs label indicates the call site at which a procedure is invoked. Its similarities with the call strings of [SP81] are discussed in Chapter 8. The unary predicate top is true for the current activation record. The binary relation n captures the n successor relation between list elements. The binary relation pr connects an activation record to the activation record of the caller. For a local ....

....and summarize activation records in essentially the same way that linked list elements are summarized. By representing the call site in each activation record the analysis algorithm is capable of encoding the calling context, too. This approach bears some similarity to the call string approach of [SP81], since it avoids propagating information to return sites that do not match the call site of the current activation record. In our case there is no need to put an arbitrary bound on the length of the call string, the bounded representation is achieved indirectly by the summarization of ....

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S.S. Muchnick and N.D. Jones, editors, Program Flow Analysis: Theory and Applications, chapter 7, pages 189-234. Prentice-Hall, Englewood Clis, NJ, 1981.

....for o stage nodes. Definition 10. nodeCheck(n; G; S) holds i both predicates localCheck(n; G) and acycCheckAll(n; G; S) hold. 6. INTERPROCEDURAL ROLE ANALYSIS Our interprocedural role analysis can be seen as similar in the spirit to the functional approach to interprocedural data ow analysis [42]. However, simply tagging a data ow fact G with the abstract values of the initial procedure context G0 is not appropriate for a complex abstraction such as role graphs. We instead approximate the transfer functions in the concrete semantics with procedure interfaces consisting of: 1) an initial ....

Micha Sharir and Amir Pnueli. Two approaches to interprocedural data ow analysis problems. In Program Flow Analysis: Theory and Applications. Prentice-Hall, Inc., 1981.

....Core Erlang programs. We use the same basic idea as for the abstract interpretation. We construct an abstract graph representation of the program, which is nite state. The construction guarantees, that its semantics is safe with respect to the SOS. Our approach is a kind of call string approach [16] on program level. The main idea of the abstraction is to replace the calls and the returns by jumps. For Example 3 a good abstraction is, that rst n times an a is sent and after that m times a b. A property like no a is sent after a b could then be proven automatically. The idea of the ....

Micha Sharir and Amir Pnueli. Two approaches to interprocedural data ow analysis. In S. S. Muchnick and N. D. Jones, editors, Program Flow Analysis: Theory and Applications, Prentice-Hall Software Series, pages 189-233. Prentice-Hall, Englewood Clis , NJ , USA, 1981.

....and memory elements: Every individual v represents either a heap cell in which case heap S (v) 1, or an activation record, in which case stack S (v) 1. The unary predicate cs label indicates the call site at which a procedure is invoked. Its similarities with the call strings of [16] are discussed in Sect. 6. The unary predicate top is true for the current activation record. The binary relation n captures the n successor relation between list elements. The binary relation pr connects an activation record to the activation record of the caller. For a local variable ....

....and summarize activation records in essentially the same way that linked list elements are summarized. By representing the call site in each activation record the analysis algorithm is capable of encoding the calling context, too. This approach bears some similarity to the call string approach of [16], since it avoids propagating information to return sites that do not match the call site of the current activation record. In our case there is no need to put an arbitrary bound on the length of the call string, the bounded representation is achieved indirectly by the summarization of ....

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S.S. Muchnick and N.D. Jones, editors, Program Flow Analysis: Theory and Applications, chapter 7, pages 189-234. Prentice-Hall, Englewood Clis, NJ, 1981.

....G) and the predicate acycCheck(n; G; S) hold. 50 Chapter 5 Interprocedural Role Analysis This chapter describes the interprocedural aspects of our role analysis. Interprocedural role analysis can be viewed as an instance of the functional approach to interprocedural data ow analysis [80]. For each program point p, the role analysis approximates program traces from procedure entry to point p. The solution in [80] proposes tagging the entire data ow fact G at point p with the data ow fact G 0 at procedure entry. In contrast, our analysis computes the correspondence between the ....

....interprocedural aspects of our role analysis. Interprocedural role analysis can be viewed as an instance of the functional approach to interprocedural data ow analysis [80] For each program point p, the role analysis approximates program traces from procedure entry to point p. The solution in [80] proposes tagging the entire data ow fact G at point p with the data ow fact G 0 at procedure entry. In contrast, our analysis computes the correspondence between the heaps at procedure entry and the heaps at point p at the granularity of sets of objects that constitute the role graphs. This ....

Micha Sharir and Amir Pnueli. Two approaches to interprocedural data ow analysis problems. In Program Flow Analysis: Theory and Applications. PrenticeHall, Inc., 1981.

....and [21] a single edge might represent a large single entry single exit (SESE) region; ow across that edge is constrained by the entire system representing the region. 4.3. 1 Context sensitive Interpretation An interprocedural analysis may be treated similarly (as in the second approach of [28]) each edge corresponding to a call site represents the solution of the called procedure s local data ow graph. If this solution can be summarized by a single transfer function using techniques of Section 4.2, then we can use it in the interpretation of each call to that procedure in other ....

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S. Muchnick and N. Jones, editors, Program Flow Analysis: Theory and Applications. Prentice Hall Inc, 1981.

....programs, however, tools that require dependence information also require information about interprocedural dependencies: dependencies that exist because of interactions among procedures. Many de nitions of, and methods for computing, interprocedural data dependencies have been presented (e.g. [11, 13, 15, 19, 22]) In contrast, our search of the research literature reveals only one attempt to de ne and compute interprocedural control dependencies [14] Unfortunately, that approach omits control dependencies that may indicate the presence of semantic dependencies, and it incorrectly identi es control ....

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. Prentice-Hall, 1981. 41

....The key property of 0 CFA is that each function is analyzed just once (or not at all, if the analysis is demand driven) The idea of polyvariance is to analyze functions more than once and thereby obtain better precision for each call site. Polyvariant analysis was pioneered by Sharir and Pnueli [58], and Jones and Muchnick [36] In the 1990s the study of polyvariant analysis has been intensive. Well known are the k CFA of Shivers [59] the poly k CFA of Jagannathan and Weeks [31] the polymorphic splitting of Jagannathan and Wright [32] and the cartesian product algorithm of Agesen [1, 2] ....

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In Steven Muchnick and Neil Jones, editors, Program Flow Analysis, Theory and Applications. 1981.

....just a few) In terms of paths through the program, one tries to avoid working with a complete ow graph where all call sites are linked to all function entries and where all function exits are linked to all return sites. Often this is accomplished by means of contours [25] a la call strings [23] or tokens [12] so as to improve the precision of the information obtained. One way to specify the analysis is to show how to generate a set of constraints [8, 9, 18, 19] whose least solution is then computed using graph based ideas. However, the majority of papers on Control Flow Analysis (e.g. ....

.... (and this set of paths is a regular language) whereas the interprocedural view is that only those paths will be valid where procedure entries and exits match in the manner of parentheses (and this set of paths is a proper context free language) Most papers on Data Flow Analysis (e.g. [23, 13]) do not consider rst class procedures and therefore have no need for a component akin to Control Flow Analysis a notable exception to this is [20] One approach deals with the interprocedural analysis by obtaining transfer functions for entire call statements [23, 13] and to some ....

[Article contains additional citation context not shown here]

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S. S. Muchnick and N. D. Jones, editors, Program Flow Analysis. Prentice Hall International, 1981.

....just a few) In terms of paths through the program, one tries to avoid working with a complete ow graph where all call sites are linked to all function entries and where all function exits are linked to all return sites. Often this is accomplished by means of contours [29] a la call strings [27] or tokens [14] so as to improve the precision of the information obtained. One way to specify the analysis is to show how to generate a set of constraints [10, 11, 22, 23] whose least solution is then computed using graph based ideas. However, the majority of papers on Control Flow Analysis ....

.... a regular language) whereas the interprocedural view is that only those paths will be valid where procedure entries and exits match in the manner of parentheses (and this set of paths is a proper context free language) Looking at the literature, the majority of papers on Data Flow Analysis (e.g. [27, 17]) do not consider rst class procedures and therefore have no need for a component akin to Control Flow Analysis an exception to this is [24] that studies an object oriented language entry P1 entry P2 P1 P2 call of P1 or P2 call of exit exit P2 Figure 1: Function call. with virtual method ....

[Article contains additional citation context not shown here]

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S. S. Muchnick and N. D. Jones, editors, Program Flow Analysis. Prentice Hall International, 1981.

No context found.

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In Program Flow Analysis: Theory and Applications, chapter 7, pages 189-234. Prentice-Hall, 1981.

No context found.

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S.S. Muchnick and N.D. Jones, editors, Program Flow Analysis: Theory and Applications, pages 189-233. Prentice-Hall, 1981.

No context found.

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S.S.Muchnick and N.D.Jones, editors, Program Flow Analysis: Theory and Applications, pages 189-233. Prentice-Hall, Englewood Clis, NJ, 1981.

No context found.

M. Sharir and A. Pnueli. Two approaches to interprocedural data- ow analyis. In S.S. Muchnick and N. D. Jones, editors, Program Flow Analysis: Theory and Applications, chapter 7, pages 189-234. Prentice-Hall, 1981.

No context found.

M. Sharir and A. Pnueli. Two approaches to interprocedural data ow analysis. In S.S. Muchnick and N.D. Jones, editors, Program Flow Analysis: Theory and Applications, pages 189-233. Prentice-Hall, 1981.

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