[Bug tree-optimization/92596] [10 Regression] ICE in exact_div, at poly-int.h:2162 since r278406

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92596

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
So I think the issue is that we have

      /* Calculate the unrolling factor based on the smallest type.  */
      poly_uint64 unrolling_factor
        = calculate_unrolling_factor (max_nunits, group_size);

with max_nunits == 1 and group_size == 3 because we end up with
V1DImode as vectype.  But then later we end up with V2SImode vectors
for

_2 = c.0_1 == 0;

but the vectorization factor is 1.