https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89889
Andrew Pinski <pinskia at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Keywords| |missed-optimization Component|c++ |middle-end --- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> --- This is because LLVM promotes the alloca to an array that is "statically" allocated on the stack. I would say this is a bad micro-benchmark really.