DocumentCode :
1799911
Title :
Specializing Compiler Optimizations through Programmable Composition for Dense Matrix Computations
Author :
Qing Yi ; Qian Wang ; Huimin Cui
Author_Institution :
Dept. of Comput. Sci., Univ. of Colorado, Colorado Springs, CO, USA
fYear :
2014
fDate :
13-17 Dec. 2014
Firstpage :
596
Lastpage :
608
Abstract :
General purpose compilers aim to extract the best average performance for all possible user applications. Due to the lack of specializations for different types of computations, compiler attained performance often lags behind those of the manually optimized libraries. In this paper, we demonstrate a new approach, programmable composition, to enable the specialization of compiler optimizations without compromising their generality. Our approach uses a single pass of source-level analysis to recognize a common pattern among dense matrix computations. It then tags the recognized patterns to trigger a sequence of general-purpose compiler optimizations specially composed for them. We show that by allowing different optimizations to adequately communicate with each other through a set of coordination handles and dynamic tags inserted inside the optimized code, we can specialize the composition of general-purpose compiler optimizations to attain a level of performance comparable to those of manually written assembly code by experts, thereby allowing selected computations in applications to benefit from similar levels of optimizations as those manually applied by experts.
Keywords :
mathematics computing; matrix algebra; optimising compilers; pattern recognition; coordination handles; dense matrix computation; dynamic tags; general purpose compilers; general-purpose compiler optimization; pattern recognition; programmable composition; source-level analysis; Arrays; Kernel; Libraries; Optimization; Pattern recognition; Runtime; Safety; Automatic programming; Computer science; Computers and information processing; Programming;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Microarchitecture (MICRO), 2014 47th Annual IEEE/ACM International Symposium on
Conference_Location :
Cambridge
ISSN :
1072-4451
Type :
conf
DOI :
10.1109/MICRO.2014.14
Filename :
7011420
Link To Document :
بازگشت