https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81803
--- Comment #17 from Eric Botcazou <ebotcazou at gcc dot gnu.org> --- > If it ends up as a simple register move though then that shouldn't be > harmful either as it would be a WORD_REGISTER_OPERATION. I have no objection > to removing the restriction though; it is easier to reason about without the > special case. Yes, this would be my recommendation. I just successfully tested this variant of the fix on SPARC64/Linux.
