• DocumentCode
    2270980
  • Title

    Bug analysis and corresponding error models in real designs

  • Author

    Lv, Tao ; Xu, Tong ; Zhao, Yang ; Li, Huawei ; Li, Xiaowei

  • Author_Institution
    Chinese Acad. of Sci., Beijing
  • fYear
    2007
  • fDate
    7-9 Nov. 2007
  • Firstpage
    59
  • Lastpage
    64
  • Abstract
    This paper presents the item-missing error model. It stems from the analysis of real bugs that are collected in two market-oriented projects: (1) the AMBA interface of a general-purpose microprocessor IP core; (2) a wireless sensor network oriented embedded processor. The bugs are analyzed via code structure comparison, and it is found that item-missing errors merit attention. The test generation method for item-missing error model is proposed. Structural information obtained from this error model is helpful to reach a greater probability of bug detection than that in random-generation verification with only functional constraints. Finally, the proposed test method is applied in verification of our designs, and experimental results demonstrate the effectiveness of this method.
  • Keywords
    error analysis; industrial property; microprocessor chips; probability; program debugging; program verification; wireless sensor networks; AMBA interface; bug analysis; bug detection probability; embedded processor; item-missing error model; microprocessor IP core; random-generation verification; wireless sensor network; Automatic test pattern generation; Circuit faults; Circuit testing; Computer architecture; Computer bugs; Computer errors; Error analysis; Genetic mutations; Laboratories; Microprocessors; error models; item-missing bugs; simulation-based validation; test generation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    High Level Design Validation and Test Workshop, 2007. HLVDT 2007. IEEE International
  • Conference_Location
    Irvine, CA
  • ISSN
    1552-6674
  • Print_ISBN
    978-1-4244-1480-2
  • Type

    conf

  • DOI
    10.1109/HLDVT.2007.4392788
  • Filename
    4392788