期刊
COMMUNICATIONS OF THE ACM
卷 59, 期 2, 页码 114-122出版社
ASSOC COMPUTING MACHINERY
DOI: 10.1145/2863701
关键词
-
类别
资金
- NSF [CCF-0915766]
- Army High Performance Computing Research Center
- Division of Computing and Communication Foundations
- Direct For Computer & Info Scie & Enginr [1409813] Funding Source: National Science Foundation
The optimization of short sequences of loop-free, fixed-point assembly code sequences is an important problem in high-performance computing. However, the competing constraints of transformation correctness and performance improvement often force even special purpose compilers to produce sub-optimal code. We show that by encoding these constraints as terms in a cost function, and using a Markov Chain Monte Carlo sampler to rapidly explore the space of all possible code sequences, we are able to generate aggressively optimized versions of a given target code sequence. Beginning from binaries compiled by 11vm -O0, we are able to produce provably correct code sequences that either match or outperform the code produced by gcc -O3, icc -O3, and in some cases expert handwritten assembly.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据