Advertisement

Continuous Analysis Based on Profile Information for Vectorization Optimization

  • Yuan Yao
  • Rong-cai Zhao
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 145)

Abstract

Aiming at the non-continuous reference problem inside the loop, a method which can implement the continuous analysis based on feedback-directed compiling optimization technique is proposed. It can make up for the traditional vector-based static analysis identified inadequate for the program appears contains a lot of pointers, array of structures and other complex data structures, and to provide more accurate identification and optimization of the vector.

Keywords

Vectorization Optimization Structure Array Continuous Analysis Reference Type Profile Information 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Overview of the open64 Compiler Infrastructure, http://open64.sourceforge.net
  2. 2.
    Nuzman, D., Rosen, I., Zaks, A.: Auto-Vectorization of Interleaved Data for SIMD. In: PLDI 2006, Ottawa, Canada, pp. 132–143 (2006)Google Scholar
  3. 3.
    WHIRL Symbol Table Specification, http://open64.sourceforge.net/symtab_Pro64_SGI.pdf
  4. 4.
    SPEC2000 overview, http://www.spec.org/cpu2000/
  5. 5.
    Slingerland, N., Smith, A.J.: Design and characterization of the Berkeley multimedia workload. Multimedia Systems 8(4), 315–327 (2002)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Berlin Heidelberg 2012

Authors and Affiliations

  1. 1.ZhengZhou Information Science and Technology InstituteZhengzhouChina

Personalised recommendations