[Bug target/114801] [14/15 Regression] arm: ICE in find_cached_value, at rtx-vector-builder.cc:100 with MVE intrinsics

["jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114801

--- Comment #9 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
(In reply to Christophe Lyon from comment #7)
> This fails because down the call chain from lowpart_subreg, we reach
> gcc_unreachable in rtx_vector_builder::find_cached_value because m_mode ==
> V4BImode (so is MODE_VECTOR_BOOL), but is not a valid expected boolean
> constant vector.

Ah, so the builtin prototype says the argument is unsigned short (or unsigned
char?)
but mode (as the insn predicate mode) is V*BImode in that case?
Then it would still be passed as SImode, so one would need
if (GET_MODE (x) == SImode || GET_MODE (x) == VOIDmode)
  {
    x = lowpart_subreg ((mode == HImode || mode == V16BImode) ? HImode :
QImode, x, SImode);
    if (mode != HImode)
      x = lowpart_subreg (mode, x, mode == V16BImode ? HImode : QImode);
  }
Does that second lowpart_subreg create CONST_VECTOR you are looking for?