[Bug tree-optimization/119351] [15 Regression] Wrong code in GROMACS for AArch64 generic SVE VLS target since r15-6807-g68326d5d1a593d

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119351

--- Comment #15 from Tamar Christina <tnfchris at gcc dot gnu.org> ---
The following example reproduces the CFG but not the bad codegen:
https://godbolt.org/z/Thzo7hz8P

This generates the actual code I expected:

  _55 = {_2, _2, _2, _2};
  _56 = {_11, _11, _11, _11};
  _57 = {ci_y_16, ci_y_16, ci_y_16, ci_y_16};
  _60 = _57 + { 0, 1, 2, 3 };
  max_mask_71 = .WHILE_ULT (0, niters.4_38, { 0, 0, 0, 0 });

So I have look at what scenario we caused the other bad reduction.