[Bug target/116738] Constant folding of _mm_min_ss and _mm_max_ss is wrong

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116738

Alexander Monakov <amonakov at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |amonakov at gcc dot gnu.org

--- Comment #1 from Alexander Monakov <amonakov at gcc dot gnu.org> ---
Yeah, expanding intrinsics to RTL smin/smax is wrong, because smin/smax are
undefined for NaNs and signed zeroes. For full-vector intrinsic _mm_min_ps we
correctly use UNSPEC_IEEE_MIN.