Abstract | ||
---|---|---|
Modern processors' multimedia extensions (MME) provide SIMD ISAs to boost the performance of typical operations in multimedia applications. However, automatic vectorization support for them is not very mature. The key difficulty is how to vectorize those SIMD-ISA-supported idioms in source code in an efficient and general way. In this paper, we introduce a powerful and ex-tendable recognition engine to solve this problem, which only needs a small amount of rules to recognize many such idioms and generate efficient SIMD in-structions. We integrated this engine into the classic vectorization framework and obtained very good performance speedup for some real-life applications. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1007/978-3-540-31985-6_5 | CC |
Keywords | Field | DocType |
ex-tendable recognition engine,multimedia application,efficient simd in-structions,simd-isa-supported idiom,multimedia extension,simd instruction,automatic vectorization support,good performance speedup,classic vectorization framework,key difficulty,simd isas,source code | Programming language,Source code,Computer science,Parallel computing,Vectorization (mathematics),SIMD,Compiler,Image tracing,Boosting (machine learning),Multimedia,Coordination language,Speedup | Conference |
Volume | ISSN | ISBN |
3443 | 0302-9743 | 3-540-25411-0 |
Citations | PageRank | References |
5 | 0.46 | 17 |
Authors | ||
7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Weihua Jiang | 1 | 5 | 0.46 |
Chao Mei | 2 | 182 | 11.88 |
Bo Huang | 3 | 339 | 17.03 |
Jianhui Li | 4 | 146 | 31.34 |
Jiahua Zhu | 5 | 6 | 1.16 |
Binyu Zang | 6 | 984 | 62.75 |
Chuan-qi Zhu | 7 | 240 | 39.42 |