Execution Indirection or Dynamic Calls?
When would the Execution Indirection be used as opposed to explicitly adding a relationship?
Using execution indirection for programs that use parameters to determine which entry point to call, which data file to access, etc. For example, there maybe a program called "LINK" that takes one parameter. LINK just loads and calls the program name that's passed in as a parameter.
Suppose program PROGA calls LINK like this:
If LINK is set up using Execution indirection, Revolve knows to automatically set up a system_invokes_entry relation between PROGA and PROGB. No manual addition of the relation between PROGA and PROGB is then needed.