[Bug target/80874] gcc does not emit cmov for minmax

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80874

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |x86_64-linux-gnu
          Component|tree-optimization           |target
           Keywords|                            |missed-optimization

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
aarch64 produces the cmov:
minmax1(int, int):
        cmp     w0, w1
        csel    w2, w0, w1, gt
        csel    w0, w1, w0, gt
        orr     x0, x0, x2, lsl 32
        ret
minmax2(int, int):
        cmp     w0, w1
        csel    w2, w1, w0, le
        csel    w0, w0, w1, le
        orr     x0, x0, x2, lsl 32
        ret

Note minmax3 code is not correct itself returns a std::pair<int&,int&>