[Bug tree-optimization/127975] [16/17 Regression] vect: ICE in vect_get_loop_mask on omp declare simd simdlen(2) call with -mavx512vl and vect-partial-vector-usage=2

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127975

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
So we end up with

(gdb) p debug_generic_expr (vectype)
vector(2) float
(gdb) p factor
$6 = 8
(gdb) p rgm->type
$3 = <vector_type 0x7ffff66ec0a8>
(gdb) p debug_generic_expr ($3)
vector(16) <signed-boolean:1>
(gdb) p debug_generic_expr (mask_type)
vector(2) <signed-boolean:32>

that is a sofar unexpected setup for how we record loop masks and we
do not verify that constraint.