Oh yeah - absolutely. This isn't even just a PowerPC thing - it happens more generally, because usually if you're forwarding both arguments and return values for a function that has the same calling conventions, you don't really have to do much adaptation, everything was already set up for you. It hit me a ton on x86 when I first started my journey into trying to reverse engineer. It still occasionally hits me, especially when I'm relying too hard on decompiler output and not paying enough attention.

There are plenty of examples where two very semantically different source codes can have the same output, which is really counter-intuitive when combined with the difficulty that often comes with trying to find a single source code that does match.

This is just another reason why debug information is such a godsend; having symbol names for C++ code will usually give you most of the function signature, and type information will give you the rest. That greatly enriches the disassembly, and the automated decompilation output, and no doubt constrains the number of possible source codes that could match both the output and the debug information, somewhat alleviating this issue.