Search Patents
  • Publication number: 20030018459
    Abstract: An improved method for debugging of analog and mixed signal behavioral models during simulation using Newton-Raphson iteration replay. The method according to the invention has substantially modified the prior art solution by limiting the interactive debugging steps in a replay of the last iteration of the accepted timepoints. Using this method, the user only interacts with the simulation during the iteration replay, and only for the accepted solution points. If the user is single stepping through this simulation, the simulator enters interactive mode at each statement during the replay. Similarly, if not single stepping, but a breakpoint has been triggered, the simulator enters the interactive mode at the appropriate statement to honor the breakpoint. While the iteration replay is performed, the system of equations does not need to be solved again. Instead, the solution vector is reinstated from the known solution of the last iteration.
    Type: Application
    Filed: June 20, 2001
    Publication date: January 23, 2003
    Inventors: Donald J. O' Riordan, Richard Trihy
  • Patent number: 7085700
    Abstract: An improved method for debugging of analog and mixed signal behavioral models during simulation using Newton-Raphson iteration replay. The method according to the invention has substantially modified the prior art solution by limiting the interactive debugging steps in a replay of the last iteration of the accepted timepoints. Using this method, the user only interacts with the simulation during the iteration replay, and only for the accepted solution points. If the user is single stepping through this simulation, the simulator enters interactive mode at each statement during the replay. Similarly, if not single stepping, but a breakpoint has been triggered, the simulator enters the interactive mode at the appropriate statement to honor the breakpoint. While the iteration replay is performed, the system of equations does not need to be solved again. Instead, the solution vector is reinstated from the known solution of the last iteration.
    Type: Grant
    Filed: June 20, 2001
    Date of Patent: August 1, 2006
    Assignee: Cadence Design Systems, Inc.
    Inventors: Donald J. O'Riordan, Richard Trihy