Username   Password       Forgot your password?  Forgot your username? 

A New Improved Algorithm for SLP

Volume 13, Number 7, November 2017 - Paper 11  - pp. 1087-1093
DOI: 10.23940/ijpe.17.07.p11.10871093

Zhan-Jie Guoa, Hui Liub,c,*

aZhengzhou Technical College, Zhengzhou,450000, China
bPLA Information Engineering University, Zhengzhou,450000, China
cHenan Normal University, Xinxiang 453007, China 

(Submitted on July 25, 2017; Revised on August 30, 2017; Accepted on September 15, 2017)


superword level parallel (SLP) algorithm cannot effectively handle the large-scale applications which covered few parallel codes, and the codes which can be vectorized may be adverse to the vectorization. A new improved algorithm for SLP is proposed. First of all, attempt to transform the non-isomorphic statements, which can’t be vectorized to isomorphic statements as far as possible. Namely, locate the opportunities of vectorization which SLP has lost, and then build the Max Common Subgraph (MCS) through adding redundant nodes, process some optimization such as redundant deleting to get the supplement diagram of SLP, it can greatly increase the parallelism of program. At last, using the method of cutting, eliminate the codes harmful to the vectorization, and execute them in serial. This vectorizes the revenue codes, improving the efficiency of programs as far as possible. Experimental results show that, compared with the SLP algorithm, its performance in average is better than it 9.1%.


References: 15

      1. J. Fritts, F. Steiling, and J. Tucek, “MediaBench II video: Expediting the Next Generation of Video Systems Research,” Microprocessors & Microsystems, 2005, vol. 33, no. 4, pp. 301-318
      2. W. Gao, R. C. Zhao, L Han, “Research on SIMD Auto-vectorization Compiling Optimization,” Journal of Software, 2015, 26(6):1265−1284 (in Chinese)
      3. T. Hiroaki, Y. Akeuchi, K. Sakanushi, et al. “Pack Instruction Generation for Media Processors Using Multi-valued Decision Diagram,” in Proceedings of the 4th International Conference on Hardware/Software Codesign and System Synthesis. ACM, 2006:154-159
      4. S. Larsen, S. Amarasinghe, “Exploiting Superword level Parallelism with Multimedia Instruction Sets,” Acm Sigplan Notices, 2000, 35(5), 145-156
      5. NAS parallel benchmark suite, Avaiklable at Resources/Software/npb.html, Last accessed on June 16, 2014
      6. M. Prieto, L. Pinuel, F. Catthoor, et al. “Improving Superword Level Parallelism Support in Modern Compilers,” IEEE/ACM/IFIP International Conference on Hardware/software Codesign and System Synthesis, CODES+ISSS 2005, Jersey City, Nj, Usa, September. 2005:303-308
      7. L. N. Pouchet, “PolyBench: The polyhedral benchmark suite,” Available at˜pouchet/software/
      8. V. Porpodas and T. Jones, “Throttling Automatic Vectorization: When Less is More,” International Conference on Parallel Architecture & Compilation. IEEE, 2015:432-444
      9. V. Porpodas, A. Magni, T. M. JONES, “PSLP: Padded SLP Automatic Vectorization,” Code Generation and Optimization (CGO), 2015 IEEE/ACM International Symposium on. IEEE, 2015:190-201
      10. “Spec cpu2006,”  Available at, Last accessed on  August 24, 2015
      11. W. M. Joseph, “High Performance Compilers for Parallel Computing,” 1996
      12. W. Y. Suo, R. C. Zhao, Y. Yao, “Superword Level Parallelism Instruction Analysis and Redundancy Optimization Algorithm on DSP,” Journal of Computer Applications, 2012, 32(12):3303-3307
      13. Z. Gtoumavitis, B. WANG, “Towards a Holistic Approach to Auto-parallelization: Integrating Profile-driven Parallelism Detection and Machine-learning Based Mapping,” Acm Sigplan Notices, 2009, 44(6):177-187
      14. S. Wei, R. C. Zhao, Y. Yao, “Loop-Nest Auto-Vectorization Based on SLP,” Journal of Software, 2012, 23(07):1717-1728
      15. J. L. Xu, R. C. Zhao, L. Han, “Vector Exploring Path Optimization Algorithm of Superworld Level Parallelism with Subsection Constraints,” Journal of Computer Applications, 2015, 35(04):950-955


          Click here to download the paper.

          Please note : You will need Adobe Acrobat viewer to view the full articles.Get Free Adobe Reader

          This site uses encryption for transmitting your passwords.