On the contrary, R. Bryant has shown that any closed oriented real analytic riemannian manifold is the real locus of an antiholomorphic, isometric involution of a Calabi-Yau 3-fold (see http://arxiv.org/abs/math/9912246).