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

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

• Type: Journal
• Year: 2012
• Where: MICRO
• Authors: Feng Li, Antoniu Pop, Albert Cohen
• Keywords: Data Parallelism, Hardware, Imperative Programs, MICRO 2012, Program Dependence Graph