cpu 0.2.0

cpu
Documentation 
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

// cache-line padding to reduce false sharing
//
// false sharing occurs when different threads write to distinct variables on same cache line
// cache coherence operates at cache-line granularity causing unnecessary invalidation traffic
//
// CachePadded<T> aligns value to 64-byte boundary and ensures size is multiple of 64
// so adjacent instances dont share cache lines

use core::fmt;
use core::ops::{Deref, DerefMut};

pub const CACHE_LINE_SIZE: usize = 64;

// pads and aligns value to 64-byte boundary
// reduces false sharing by ensuring adjacent CachePadded<T> values don't share cache lines
//
// guarantees:
// - alignment 64
// - value field at offset 0
// - size rounded up to multiple of 64
#[repr(C, align(64))]
pub struct CachePadded<T> {
    value: T,
}

impl<T> CachePadded<T> {
    #[inline]
    pub const fn new(value: T) -> Self {
        Self { value }
    }

    #[inline]
    pub fn into_inner(self) -> T {
        self.value
    }

    #[inline]
    pub fn get_mut(&mut self) -> &mut T {
        &mut self.value
    }
}

impl<T> Deref for CachePadded<T> {
    type Target = T;

    #[inline]
    fn deref(&self) -> &T {
        &self.value
    }
}

impl<T> DerefMut for CachePadded<T> {
    #[inline]
    fn deref_mut(&mut self) -> &mut T {
        &mut self.value
    }
}

impl<T: Default> Default for CachePadded<T> {
    fn default() -> Self {
        Self::new(T::default())
    }
}

impl<T: Clone> Clone for CachePadded<T> {
    fn clone(&self) -> Self {
        Self::new(self.value.clone())
    }
}

impl<T: Copy> Copy for CachePadded<T> {}

impl<T: fmt::Debug> fmt::Debug for CachePadded<T> {
    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
        f.debug_struct("CachePadded")
            .field("value", &self.value)
            .finish()
    }
}

impl<T: PartialEq> PartialEq for CachePadded<T> {
    fn eq(&self, other: &Self) -> bool {
        self.value == other.value
    }
}

impl<T: Eq> Eq for CachePadded<T> {}

impl<T> From<T> for CachePadded<T> {
    fn from(value: T) -> Self {
        Self::new(value)
    }
}

#[cfg(test)]
mod tests {
    use super::{CACHE_LINE_SIZE, CachePadded};
    use core::mem::{align_of, size_of};

    #[test]
    fn aligns_to_cache_line() {
        assert_eq!(align_of::<CachePadded<u8>>(), CACHE_LINE_SIZE);
    }

    #[test]
    fn size_is_cache_line_multiple() {
        assert_eq!(size_of::<CachePadded<u8>>() % CACHE_LINE_SIZE, 0);
        assert_eq!(size_of::<CachePadded<[u8; 65]>>() % CACHE_LINE_SIZE, 0);
    }

    #[test]
    fn derefs_to_inner_value() {
        let padded = CachePadded::new(11u32);
        assert_eq!(*padded, 11);
    }
}