Trait UlpFloat 

pub trait UlpFloat: Copy {
    const MAX_ULP: u64;
    const SIGN_MASK: u64;

    // Required methods
    fn ulp_bits(self) -> u64;
    fn ulp_is_nan(self) -> bool;

    // Provided methods
    fn ulp_magnitude_bits(self) -> u64 { ... }
    fn ulp_is_sign_negative(self) -> bool { ... }
}

Floating point types for which an integer ULP distance is defined.

Required Associated Constants

const MAX_ULP: u64

Largest distance representable for this type. Also the distance reported between a NaN and a non-NaN value.

const SIGN_MASK: u64

Bit pattern of the sign bit, widened to u64.

Required Methods

fn ulp_bits(self) -> u64

Raw bit pattern, widened to u64.

Within a single sign, the bit patterns of finite floats are monotonic in magnitude, which is what makes the subtraction below meaningful.

fn ulp_is_nan(self) -> bool

Provided Methods

fn ulp_magnitude_bits(self) -> u64

Bit pattern with the sign bit cleared, i.e. the bits of |self|.

fn ulp_is_sign_negative(self) -> bool

Dyn Compatibility

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementations on Foreign Types

impl UlpFloat for f32

const MAX_ULP: u64

const SIGN_MASK: u64

fn ulp_bits(self) -> u64

fn ulp_is_nan(self) -> bool

impl UlpFloat for f64

const MAX_ULP: u64

const SIGN_MASK: u64

fn ulp_bits(self) -> u64

fn ulp_is_nan(self) -> bool

Implementors

impl UlpFloat for f16

const MAX_ULP: u64

const SIGN_MASK: u64