> A function that is called once will always be inlined.
That is objectively false. Even though I agree with the spirit. For starters, unoptimized builds won't have that. Second, exceptions tend to inhibit inlining. Compare https://godbolt.org/z/c8Jayf with https://godbolt.org/z/Uoo2YA. Third, it's easy enough to push one of the heuristics used for inlining high enough in the wrong direction (e.g. function size). I recommend the clang optimizer view on godbolt.
Also note that it can be hard for a compiler to prove that a function is not used outside the current translation unit (although anonymous namespaces help with that). Number of calls is only one of the heuristics for inlining though anyway.
