https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127610
--- Comment #3 from Drea Pinski <pinskia at gcc dot gnu.org> --- (In reply to Filip Kastl from comment #1) > I've seen no exec time speedups of perlbench in the relevant timeframe, btw. Right because the full cs-elim handles the same thing. Just that is after inlining (and partioning). The idea is to simplify the cfg and ir before inlinung for better costing of the ir.
