Vector Normalization

The normalize instruction scales a floating-point vector to unit length.

Opcode	Arguments	Operands (stack)	Description
normalize	type	values	Replaces one floating-point array with an equal-length unit vector in the selected type; type must be hlf, flt, or dbl, and a zero vector produces an equal-length zero vector.