Structure optimization method based on automatic vectorization

  • Yu-ping LiEmail author
  • Zhan-jie Guo
  • Hui Liu
Special Issue


Structure is used more extensively used in program such as scientific computing. But the non-continuity and the non-aligment of vectorization structure array have a dramatic influence on the efficiency of program’s vectorizaton. To reduce the access to these addresses during the SIMD vectorization, a structure peeling model is proposed based on the structure which combines domain access affinity and domain data type. At the same time, to meet t the requirement of memory access continuity and alignment in the vectorization of structured array, an address conversion method is proposed which structure arrays are mapped one by one map to two-dimensional arrays, further reducing the failure rate of cache. By using the test suites of gcc_vec, spec2000 and spec2006, the experimental results on the compiler of automatic vector show that the performance of optimized method can be improved by more than 8%.


Structure peeling Address mapping SIMD vectorization Access affinity 



  1. 1.
    Chen J-F, Wang Y-D (2012) Automatic generation of attack vectors for stored-XSS. J Grad Univ Chin Acad Sci 29(6):815–820Google Scholar
  2. 2.
    Yan J, Chen W, Zheng W (2011) Efficient memory management for structured data-layout optimization. J Tsinghua Univ 51(1):68–72Google Scholar
  3. 3.
    Li Y-X, Shi H, Chen L (2009) Vectorization-oriented local data regrouping. J Chin Comput Syst 30(8):1528–1534Google Scholar
  4. 4.
    Hagog M, Tice C (2005) Cache aware data layout reorgan optimization in GCC. In: Proceeding of the GCC Developers’Summit, pp 69–92Google Scholar
  5. 5.
    Tanaka H, Ota Y, Matsumoto N, et al (2011) A new compilation technique for SIMD code generation across basic block boundaries. In: Proceedings of the 2010 Asia and South Pacific design automation conference. IEEE Press, pp 101–106Google Scholar
  6. 6.
    Zhang Y-Y, Zhao R-C, Han L (2012) Vectorization benefit evaluation method based on polyhedron representation. Comput Eng 38(7):266–268Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  1. 1.School of Information TechnologyShangqiu Normal UniversityShangqiuChina
  2. 2.Department of Electrical and Electronic EngineeringZhengzhou Technical CollegeZhengzhouChina
  3. 3.Collge of Information Systems EngineeringInformation Engineering UniversityZhengzhouChina

Personalised recommendations