[Bug rtl-optimization/127814] [14/15/16/17 Regression] wrong code at -O3 with -fno-vect-cost-model on x86_64-linux-gnu

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127814

--- Comment #4 from Drea Pinski <pinskia at gcc dot gnu.org> ---
In the original there is no signed integer overflow but in my versions there
is.  Back to reducing it differently.