[Bug target/87599] Broadcasting scalar to vector uses stack unnecessarily on x86

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87599

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Try with -mtune=intel. So AMD cores are faster with the move between gpr and
sse register sets via memory rather than direct