[Bug target/101846] Improve __builtin_shufflevector emitted code

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101846

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |enhancement

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
With just -mavx512f we produce a bunch of instructions (looking like we went to
scalar mode) while LLVM is able to produce:
foo(short __vector(16)):                           # @foo(short __vector(16))
        .cfi_startproc
# %bb.0:
        vpmovzxwd       ymm1, xmm0              # ymm1 =
xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
        vextracti128    xmm0, ymm0, 1
        vpmovzxwd       ymm0, xmm0              # ymm0 =
xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
        vinserti64x4    zmm0, zmm1, ymm0, 1
        ret


bar(short __vector(32)):                           # @bar(short __vector(32))
        .cfi_startproc
# %bb.0:
        vpmovdw ymm0, zmm0
        ret


For -march=skylake512 we do produce now:
foo(short __vector(16)):
        vpmovzxwd       zmm0, ymm0
        ret
bar(short __vector(32)):
        vpmovdw ymm0, zmm0
        ret

So still confirmed for the -mavx512f case.