Clarify lifetime of argument functions in fmi2Instantiate() about fmi-standard HOT 3 CLOSED

TorstenS: If you do not take a copy of the function pointers, you can run into trouble.
Andreas: "The struct contains pointers to functions provided by the environment to be used by the
FMU. It is not allowed to change these functions between fmi2Instantiate(..) and
fmi2Terminate(..) calls."
TorstenS: My question was: can one keep a pointer to the struct?
Andreas: Maybe it is not specified (I found it for strings)
TorstenS: Should we define pointers are only valid for the call?
Andreas: Anything that points into memory of the importer ... (not logging).
Andreas, TorstenS: We have to think this through more carefully. And perhaps mandate what is best practice.

from fmi-standard.

The current text makes little sense.

C does not allow a function to change - whereas function pointers are often changed, so a sensible interpretation is that when it says "It is not allowed to change these functions between fmi2Instantiate(..) and fmi2Terminate(..) calls." the intent was "It is not allowed to change these functions pointers between fmi2Instantiate(..) and fmi2Terminate(..) calls." which means that the pointed to struct cannot be changed by the caller - in that case it doesn't matter if the FMI makes a copy of the function pointers or not; as the values cannot change.

If we only wanted that the struct should be valid during the call we could have sent in a struct instead of a pointer to struct (the overhead of sending a struct by value during instantiation is minimal).

from fmi-standard.

Fixed with #1931

from fmi-standard.