Automatic Extraction of Coarse-Grained Data-Flow Threads from Imperative Programs

Feng Li (1) , Antoniu Pop (1) , Albert Cohen (1)

1 Parkas - Parallélisme de Kahn Synchrone

This article presents a general algorithm for transforming sequential imperative programs into parallel data-flow programs. The algorithm operates on a program dependence graph in static-single-assignment form, extracting task, pipeline, and data parallelism from arbitrary control flow, and coarsening its granularity using a generalized form of typed fusion. A prototype based on GNU Compiler Collection (GCC) is applied to the automatic parallelization of recursive C programs.

Mots clés: data flow computing; data flow graphs; parallel programming; program compilers; program interpreters; GNU compiler collection; arbitrary control flow; automatic coarse-grained data-flow thread extraction; automatic recursive C program parallelization; data parallelism; imperative programs; parallel data-flow programs; program dependence graph; sequential imperative program transformation; task extraction; Instruction sets; Parallel processing; Pipeline processing; Radiation detectors; Sequential analysis; Synchronization; SSA form; automatic parallelization; data-flow model; loop fusion; sequential imperative programs

Domaines: Informatique et langage [cs.CL]

Feng Li, Antoniu Pop, Albert Cohen. Automatic Extraction of Coarse-Grained Data-Flow Threads from Imperative Programs. IEEE Micro, 2012, 32 (4), pp.19-31. ⟨10.1109/MM.2012.49⟩. ⟨hal-00906099⟩