https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111231
--- Comment #40 from Andrew Pinski <pinskia at gcc dot gnu.org> --- (In reply to Andrew Pinski from comment #39) > (In reply to Richard Earnshaw from comment #34) > > To be honest, I'm more concerned that we aren't eliminating a lot of these > > copies during the gimple optimization phase. The memcpy is really a type > > punning step (that's strictly ISO C compliant, rather than using the GCC > > union extension), so ideally we'd recognize that and eliminate as many of > > the copies as possible (perhaps using some form of view_convert or whatever > > gimple is appropriate for changing the view without changing the contents). > > > > But that's for another day... > > I will be looking into improving the situation here in the next couple of > weeks. I noticed a few patterns for this code which I will file tomorrow and > link back in the see also. Note Gcc 16 does remove at least a few copies but > not the memcpy ones. Filed PR 121417 and 121418 for the copies I saw which should be able to be fixed.
