- 作者: 楊朝棟; 曾憲雄; 陳昌盛
- 作者服務機構: 交通大學資訊科學所
- 中文摘要: 平行編譯器能從循序程式的迴圈中偵測出隱藏的平行性,並自動重組循序程式成為可平行處理的子工作。目前為止,已有許多研究都是針對在共享記憶體式多處理機系統上,如何找到一個有效率且精確的平行編譯器來使系統達到高速度。Parafrase-2是由伊利諾大學所發展的著名平行編譯器,主要目標在探測平行編譯器的程式轉換行為。在本文中,首先將解析Parafrase-2,並介紹資料結構與純量分析、資料相依分析、平行度分析、以及其他相關分析的邏輯觀點。為了發展一個雛型平行編譯器與評估此編譯器的效能,故將Parafrase-2移植到擁有Mach作業系統的Acer Altos10000系統上,同時也提出一些有關於可改進Parafrase-2的論點。最後,並提出結論以及未來的研究發展方向。
- 英文摘要: Parallel compilers analyze sequential programs, in particular their loops, to detect hidden parallelismfor automatic reconstructing of sequential programs into parallel subtasks. Up to now, many researcheshave concentrated on how to find efficient and precise parallel compilers for shared-memory multiprocessorsystems to achieve high speedup. Parafrase-2, a famous parallel compiler developed at Illinois University,aims at exploring program transformation for parallel compilers. In this paper, Parafrase-2 is first traced;the data structures and the logical views of scalar analysis, data dependence analysis, and parallelizationanalysis passes along with their relevant relations between passes are then presented. To develop a prototypeparallel compiler and evaluate its performance, Parafrase-2 was ported on an Acer Altos 10000 systemunder Mach OS. Furthermore, some improvable issues are also presented. Finally, concluding remarksand future work are also given.
- 中文關鍵字: shared-memory; multiprocessor system; multiprocessor operating system; parallel compiler; scalar analysis; data dependence analysis
- 英文關鍵字: --