encoding - Emacs c-mode can't recognize utf-8?

I need to read one C++ head file which has some Chinese and was encoded using utf-8.

Emacs should recognize this encoding, but it turns out:

Then, I changed it to text-mode, it works:

I also tested for python-mode, lisp-mode, etc, all works except c-mode, c++-mode, java-mode, seems there are something wrong with cc-mode, or the cc-vars?

Please help me if you know how to fix this weird problem.