Issue No. 08 - August (1994 vol. 5)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/71.298205
<p>As the number of available multiprocessors increases, so does the importance of providing software support for these systems, including parallel compilers. Data flow analysis, an important component of software tools, may be computed many times during the compilation of a program, especially when compiling for a multiprocessor. Althoughconverting a sequential data flow algorithm to a parallel algorithm can present someopportunities for computing data flow in parallel, more parallelism can be exposed by thedevelopment of new parallel data flow algorithms. We present a technique that computesrapid data flow problems in parallel and thus is applicable for commonly used classicaldata flow problems, including reaching definitions, reachable uses, available expressions,and very busy expressions. Unlike previous techniques, our technique exploits theinherent parallelism in the data flow computation that occurs across independent paths,within linear paths, and in paths through loops of a control flow graph. The technique first changes cyclic structures in a control flow graph to acyclic structures and then builds a combining directed acyclic graph (DAG) that represents the paths through the control flow graph needed to compute data flow. Data flow is then computed using two passesover the DAG by computing the data flow for the nodes on each level of the DAG inparallel. We also present experimental results comparing the performance of our algorithm with a sequential algorithm and a parallelized sequential algorithm.</p>
Index Termsdirected graphs; parallel programming; program compilers; parallel algorithms; parallel processing; DAG; parallel data flow analysis; multiprocessors; parallel compilers; software tools; sequential data flow algorithm; parallel algorithm; parallel data flow algorithms; rapid data flow problems; control flow graph; cyclic structures; combining directed acyclic graph; performance; sequential algorithm; parallelized sequential algorithm
M. Soffa, R. Gupta and R. Kramer, "The Combining DAG: A Technique for Parallel Data Flow Analysis," in IEEE Transactions on Parallel & Distributed Systems, vol. 5, no. , pp. 805-813, 1994.