經驗搜尋的儀表板之設計與實做

Design and Implementation of a Dashboard for an Empirical Searcher

黃禮俊(Li-Jiun Huang)

指導教授 ： 廖世偉

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.01622

摘要

現今編譯器擁有許多編譯選項，也提供了預先設定好的最佳化選項供使用者使用。這些選項如果挑選適當，對於程式的效能會有助益，然而增進效能的選項卻可能增加目的碼大小。由於選項數量眾多，如何挑選適當的選項，有一定專門性，對於使用者而言是個不小的負擔。即便是使用編譯器預先設定好的選項，也不一定會符合使用者的需求。本文描述我們實作的一個儀表板工具，在三種市面上可見的設備上，幫助使用者測試編譯選項，並在使用者輸入所期望的限制之下，顯示出適合的編譯選項。

關鍵字

編譯器 ； 編譯選項 ； 基因演算法 ； 最佳化 ； 目的碼尺寸

並列摘要

Nowadays, a compiler has many options. It also provides predefined options for certain optimization, such as performance and code size. It is good for performance if we select suitable options when compiling. However, good performance may hurt code size. How to pick up suitable options is not easy for users to meet their requirements, for example, maximize performance or minimize code size. In this thesis, we implement a dashboard to provide a way for examining the results of specific compiler options. Users can find suitable compiler options which conform to their constraint. We demonstrate out tool on three devices on the market. We show that our tool can help user to find a set of compiler options best for their requirements.

並列關鍵字

compiler ； compile options ； genetic algorithm ； optimization ； code size

參考文獻

[1] "GCC, the GNU Compiler Collection," http://gcc.gnu.org/.

[4] Z. Pan, and R. Eigenmann, “Rating Compiler Optimizations for Automatic Performance Tuning,” in Proceedings of the 2004 ACM/IEEE conference on Supercomputing, 2004.

[5] Z. Pan, and R. Eigenmann, “Fast, automatic, procedure-level performance tuning,” in Proceedings of the 15th international conference on Parallel architectures and compilation techniques, 2006.

[6] Z. Pan, and R. Eigenmann, “Fast and Effective Orchestration of Compiler Optimizations for Automatic Performance Tuning,” in Proceedings of the International Symposium on Code Generation and Optimization, 2006.

[7] J. Cavazos, G. Fursin, F. Agakov et al., “Rapidly Selecting Good Compiler Optimizations using Performance Counters,” in Proceedings of the International Symposium on Code Generation and Optimization, 2007.