According to the Simulink Tool Limitations for R2011a breakpoints cannot be set inside a referenced model. I tend to use the debugger quite extensively to follow the signal flow throughout a model, not just at the boundaries of a system, when I'm debugging it so this seems like a very severe limitation to me. I've tried to think of a work-around to this limitation like creating the model references from libraries. Then embedding the model reference inside another configurable library so that during development I can configure with the library and get access to the signals in the library. Then when needed switch the configuration to the version with the model reference. While I think this would work, this seems like a crude hack that I can't quite justify and I'm sure I couldn't convince other project team members that it's justified.
So I have two questions:
1. Does anyone know of a better work-around to get the benefits of both model referencing when needed and still be able to create a single source for a model that can be debugged with the Simulink debugger.
2. If Mathworks pushes for model referencing as the solution to large model issues, will the Simulink debugger then eventually be considered only useable for small standalone models?
Best Answer