ITR: An Optimizing Compiler for Languages with Programmable Memory Models

点击翻译按钮获取中文摘要

The memory model for a programming language specifiesthe relationship between the order in which data accessesappear in a program and the order seen during execution bythe different program components With the advent of parallelprogramming environments like Posix threads, Java, and OpenMP, multi-threaded explicitly parallel programs have become much morefrequent. This increases the need for memory models thatare easy to understand and efficient so that correct pro-grams can be developed and still give good performance.Unfortunately, the usability of memory models, their impacton performance, and the compiler technology needed to per-form optimizations of parallel programs are poorly under-stood. The result is that current memory models tend tofavor performance over usability by restricting the programsthat can be written with them, or by being difficult tounderstand. The objective of this project is to study com-piler techniques to optimize explicitly parallel programs byusing optimizations and analysis algorithms structured tohandle a broad class of consistency models. The techniquesstudied will be implemented in a compiler that will serveas a testbed for prototyping and studying programminglanguage memory models and for studying the optimizationand analysis of explicitly parallel programs.