https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84525
--- Comment #6 from Zhao Chun <buaa.zhaoc at gmail dot com> --- (In reply to Jakub Jelinek from comment #5) > I wrote it above. memcpy or packed struct. And there is no reason to think > about memcpy as something inefficient, GCC will turn those single element > memcpy calls into efficient unaligned loads or stores. OK, Thanks!