C. Fecht and H. Seidl. Propagating differences: An efficient new fixpoint algorithm for distributive constraint systems. Nordic J. of Comput., (5):304--329, 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....analyzers. Being developed for static analyses of traditional call graphs PAG will have to be extended to cope with the problems of dynamic dispatch. Control flow analysis for OO languages is done using equation systems dynamically. JOSES will solve these systems by adopting a fast fixpoint solver [4] developed by Fecht and Seidl. 4.2 Memory consumption requires garbage collection. As real time garbage collection is a run time critical issue for Java code, methods to eliminate it as much as possible are of very high importance for embedded systems. These techniques are called static garbage ....

C. Fecht and H. Seidl, Propagating Differences: An Efficient New Fixpoint Algorithm for Distributive Constraint Systems, Proceedings of the European Symposium on Programming, Lisbon, 1998

....requirements of Duesterwald, Gupta, and Soffa s framework for demand driven interprocedural analysis [7] Thus, a demand algorithm for the problem follows directly from their framework. Alternatively, the constraints may be solved using Fecht Siedl s fast algorithm to compute solutions on demand [8, 29]. The flow constraints wherein the computation of summary edges is combined with the computation of slice, 40 4 , are new. The worklist algorithm resulting from these constraints is a contribution of this paper. Binkley s SDG based interprocedural constant propagation algorithm [3] and ....

H. Seidl and C. Fecht. Propagating differences: An efficient new fixpoint algorithm for distributive constraint systems. Nordic Journal of Computing, 5:304--329, 1998. (Also appeared in the Proceedings of 7th European Symposium on Programming, ESOP'1998, Lecture Notes in Computer Science, Vol. 1381, Springer, 1998, pp. 90-104).

No context found.

C. Fecht and H. Seidl. Propagating Differences: An Efficient New Fixpoint Algorithm for Distributive Constraint Systems. Nordic Journal of Computing (NJC), 5(4):304--329, 1998.

....a basis for the span of the resulting (k 1) matrices. By Gaussian elimination, this can be done in time O(k ) Altogether, we obtain an upper complexity bound of O(p k k ) O(p k 10 ) A better runtime can be obtained if we use a semi naive fixpoint iteration strategy [16, 1, 4]. The idea here is that when the value of a fixpoint variable changes, we do not propagate the complete new value to all uses of the variable in right hand sides of constraints but just the increment, i.e. in our case the new matrices extending the current basis (instead of the complete new ....

C. Fecht and H. Seidl. Propagating Differences: An Efficient New Fixpoint Algorithm for Distributive Constraint Systems. Nordic Journal of Computing (NJC), 5(4):304-- 329, 1998.

....In the second stage, a system of (in )equations is selected which (more or less precisely) characterizes the property to be analyzed. In the last stage, a standard solver is applied to compute a (least) solution of the system of (in )equations. Such solvers have systematically been studied in [6,20,8,22,23]. Therefore, we concentrate here on the second stage, that is, the design decisions for an analysis once the abstract operational semantics of a program is fixed. The main contributions of our paper are: ffl By giving a unified framework applicable to imperative as well as logic languages we hope ....

....of formulation (3) is that not all variables y f;d of a transformer f necessarily contribute to the questions the analysis is going to answer. Therefore, the least solution of the possibly huge system of inequations need to be only partially computed. This idea is exploited by local solvers [6,22,23] (see Section 9) Instead of using the somewhat clumsy representation (3) which is good for implementation but shifts everything one level down into the indices, we prefer to stick to the more readable representation (2) and keep in mind that in fact each transformer application represents a ....

[Article contains additional citation context not shown here]

Christian Fecht and Helmut Seidl. Propagating Differences: An Efficient New Fixpoint Algorithm for Distributive Constraint Systems. In European Symposium on Programming (ESOP), pages 90--104. LNCS 1381, Springer Verlag, 1998. Long version in Northern Journal of Computing 5, 304-329,1998.

....for all r F [q] and r T [q] necessary to compute r F [q ] for the start state q ] involves exactly the reachable states of DC . LH92] propose local solvers for computing such a solution. Because the only operation required for solving the system is set union, differential local solvers as in [FS98] can also be applied. The interested reader may consult that paper for efficient algorithms. Example 5: Consider automaton DC from Ex. 4. Running a local solver yields: r F [1] f1; 3g, r F [3] f3g, r F [2] f2g, and r T [1] r T [2] r T [3] f5; 4g. i.e. states 1; 5 from Fig. ....

C. Fecht and H. Seidl. Propagating Differences: An Efficient New Fixpoint Algorithm for Distributive Constraint Systems. In ESOP '98, LNCS 1381, pages 90--104. Springer, 1998.

....computing the reachable states of a PFA Computing a partial least solution for all r F [q] and r T [q] required for computing r F [q I ] involves exactly the reachable states of DC . The only operation required for solving the system is set union. Therefore, differential local solvers as in [FS98] can be applied. The interested reader may consult that paper for more efficient algorithms than ours. Example 7: Consider again automaton DC from example 6. Running the algorithm yields: r F [q I ] fq I ; q 0 g, r F [q 0 ] fq 0 g, r F [q ] fq g, and r T [q I ] r T [q ] r T [q 0 ....

C. Fecht and H. Seidl. Propagating Differences: An Efficient New Fixpoint Algorithm for Distributive Constraint Systems. In ESOP '98, LNCS 1381, pages 90--104. Springer Verlag, 1998.

No context found.

C. Fecht and H. Seidl. Propagating differences: An efficient new fixpoint algorithm for distributive constraint systems. Nordic J. of Comput., (5):304--329, 1998.

No context found.

C. Fecht and H. Seidl. Propagating differences: An efficient new fixpoint algorithm for distributive constraint systems. In Proc. European Symposium on Programming, pages 90-- 104, 1998.

No context found.

H. Seidl and C. Fecht. Propagating differences: An efficient new fixpoint algorithm for distributive constraint systems. Nordic Journal of Computing, 5:304--329, 1998.

No context found.

C. Fecht and H. Seidl. Propagating differences: An efficient new fixpoint algorithm for distributive constraint systems. In Proc. European Symposium on Programming, pages 90-- 104, 1998.

No context found.

C. Fecht and H. Seidl. Propagating differences: An efficient new fixpoint algorithm for distributive constraint systems. In Proc. European Symposium on Programming, pages 90-- 104, 1998.

No context found.

C. Fecht and H. Seidl, Propagating Differences: An Efficient New Fixpoint Algorithm for Distributive Constraint Systems, Proceedings of the European Symposium on Programming, Lisbon, 1998

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