[Bug tree-optimization/123799] [16 Regression] ICE: verify_gimple failed: invalid types in conversion to floating-point with -O3 -mavx10.1 since r16-6976

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123799

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Of course you can convert _BitInt(64) to float or double, and you can bitcast
(VCE) double to _BitInt(64) and back.
What you can't create is VECTOR_TYPE with BITINT_TYPE elements, similarly how
VECTOR_TYPE with POINTER_TYPE elements is not supported.