I'm not sure I understand what happened in
SI-5005. It's clearly a bug, I opened it with a clear description of the bug, but it ended up being closed because it had become "confused". Okay, I will unconfuse things. Here is the same report again. One should not have to choose between @specialize working and @inline working.
To answer the question of what it would take to be considered fixed - it's fixed with respect to this ticket when the bytecode of methods m1 and m2 both is constant true.