[Bug tree-optimization/111793] OpenMP SIMD inbranch clones for AVX512 are highly sub-optimal

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111793

--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
non-openmp testcase for the AVX512 mask inefficiency:

unsigned foo (unsigned *a, unsigned short mask)
{
  unsigned sum = 0;
  for (int i = 0; i < 16; ++i)
    if ((mask >> i) & 1)
      sum += a[i];
  return sum;
}

I think the AVX2 one is settled as being as good as it can get.