[Bug target/95969] Use of __builtin_aarch64_im_lane_boundsi in AArch64 arm_neon.h interferes with gimple optimisation

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95969

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
There are two bugs with this change.
First is in the change itself; I forgot to check to make sure totalsize and
elementsize are non-zero. I have that fixed.
The second issue is some gimple level passes are not ready to change the
function to a GIMPLE_NOP yet.  I will fix those tomorrow.

Right now I see evpr and cunroll so far.