http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51628
--- Comment #8 from rguenther at suse dot de <rguenther at suse dot de> 2011-12-20 11:23:48 UTC --- On Tue, 20 Dec 2011, ebotcazou at gcc dot gnu.org wrote: > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51628 > > --- Comment #7 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2011-12-20 > 11:18:24 UTC --- > > Huh, it's not. It's the same as a packed struct or enum type. > > No, it isn't, the mode is integral instead of BLKmode. In Ada we do support > misaligned integers, but we simply wrap them in a BLKmode record. > > > Why can't you strict-align people simply fix this case? > > Because integral modes are naturally aligned. The only reasonable way to > support the aforementioned abomination is to use the Ada approach. You mean that handling the TYPE_ALIGN != MODE_ALIGN case when expanding a MEM_REF (thus, INDIRECT_REF on old branches) won't work? Why not? You'd simply have to emit the same RTL as when expanding that wrapped struct case. Richard.
