Function is_subnormal_f32 

pub const fn is_subnormal_f32(v: f32) -> bool

Is this a subnormal — a non-zero value with a zero exponent field?

Load-bearing on the VR4300 in a way it is not on most FPUs: this processor cannot compute with subnormals at all. A subnormal operand, or a subnormal result with FCSR.FS clear, raises the unmaskable unimplemented operation cause rather than producing a number. See arith_unimplemented_s.