{"title":"Optimal loop parallelization for maximizing iteration-level parallelism","authors":"Duo Liu, Z. Shao, M. Wang, M. Guo, Jingling Xue","doi":"10.1145/1629395.1629407","DOIUrl":null,"url":null,"abstract":"This paper solves the open problem of extracting the maximal number of iterations from a loop that can be executed in parallel on chip multiprocessors. Our algorithm solves it optimally by migrating the weights of parallelism-inhibiting dependences on dependence cycles in two phases. First, we model dependence migration with retiming and formulate this classic loop parallelization into a graph optimization problem, i.e., one of finding retiming values for its nodes so that the minimum non-zero edge weight in the graph is maximized. We present our algorithm in three stages with each being built incrementally on the preceding one. Second, the optimal code for a loop is generated from the retimed graph of the loop found in the first phase. We demonstrate the effectiveness of our optimal algorithm by comparing with a number of representative non-optimal algorithms using a set of benchmarks frequently used in prior work.","PeriodicalId":136293,"journal":{"name":"International Conference on Compilers, Architecture, and Synthesis for Embedded Systems","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2009-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"18","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Compilers, Architecture, and Synthesis for Embedded Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/1629395.1629407","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 18
Abstract
This paper solves the open problem of extracting the maximal number of iterations from a loop that can be executed in parallel on chip multiprocessors. Our algorithm solves it optimally by migrating the weights of parallelism-inhibiting dependences on dependence cycles in two phases. First, we model dependence migration with retiming and formulate this classic loop parallelization into a graph optimization problem, i.e., one of finding retiming values for its nodes so that the minimum non-zero edge weight in the graph is maximized. We present our algorithm in three stages with each being built incrementally on the preceding one. Second, the optimal code for a loop is generated from the retimed graph of the loop found in the first phase. We demonstrate the effectiveness of our optimal algorithm by comparing with a number of representative non-optimal algorithms using a set of benchmarks frequently used in prior work.