```1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
```
``````/// Fused multiply-add. Computes `(self * a) + b` with only one rounding
/// error, yielding a more accurate result than an unfused multiply-add.
///
/// Using `mul_add` can be more performant than an unfused multiply-add if
/// the target architecture has a dedicated `fma` CPU instruction.
///
/// Note that `A` and `B` are `Self` by default, but this is not mandatory.
///
/// # Example
///
/// ```
/// use std::f32;
///
/// let m = 10.0_f32;
/// let x = 4.0_f32;
/// let b = 60.0_f32;
///
/// // 100.0
/// let abs_difference = (m.mul_add(x, b) - (m*x + b)).abs();
///
/// assert!(abs_difference <= 100.0 * f32::EPSILON);
/// ```
pub trait MulAdd<A = Self, B = Self> {
/// The resulting type after applying the fused multiply-add.
type Output;

/// Performs the fused multiply-add operation `(self * a) + b`
fn mul_add(self, a: A, b: B) -> Self::Output;
}

/// The fused multiply-add assignment operation `*self = (*self * a) + b`
pub trait MulAddAssign<A = Self, B = Self> {
/// Performs the fused multiply-add assignment operation `*self = (*self * a) + b`
fn mul_add_assign(&mut self, a: A, b: B);
}

#[cfg(any(feature = "std", feature = "libm"))]
impl MulAdd<f32, f32> for f32 {
type Output = Self;

#[inline]
fn mul_add(self, a: Self, b: Self) -> Self::Output {
}
}

#[cfg(any(feature = "std", feature = "libm"))]
impl MulAdd<f64, f64> for f64 {
type Output = Self;

#[inline]
fn mul_add(self, a: Self, b: Self) -> Self::Output {
}
}

(\$trait_name:ident for \$(\$t:ty)*) => {\$(
impl \$trait_name for \$t {
type Output = Self;

#[inline]
fn mul_add(self, a: Self, b: Self) -> Self::Output {
(self * a) + b
}
}
)*}
}

#[cfg(any(feature = "std", feature = "libm"))]
impl MulAddAssign<f32, f32> for f32 {
#[inline]
fn mul_add_assign(&mut self, a: Self, b: Self) {
*self = <Self as crate::Float>::mul_add(*self, a, b)
}
}

#[cfg(any(feature = "std", feature = "libm"))]
impl MulAddAssign<f64, f64> for f64 {
#[inline]
fn mul_add_assign(&mut self, a: Self, b: Self) {
*self = <Self as crate::Float>::mul_add(*self, a, b)
}
}

(\$trait_name:ident for \$(\$t:ty)*) => {\$(
impl \$trait_name for \$t {
#[inline]
fn mul_add_assign(&mut self, a: Self, b: Self) {
*self = (*self * a) + b
}
}
)*}
}

#[cfg(test)]
mod tests {
use super::*;

#[test]
(\$(\$t:ident)+) => {
\$(
{
let m: \$t = 2;
let x: \$t = 3;
let b: \$t = 4;

}
)+
};
}

test_mul_add!(usize u8 u16 u32 u64 isize i8 i16 i32 i64);
}

#[test]
#[cfg(feature = "std")]
(\$(\$t:ident)+) => {
\$(
{
use core::\$t;

let m: \$t = 12.0;
let x: \$t = 3.4;
let b: \$t = 5.6;