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
150
151
152
153
154
155
156
157

use rand_core::{RngCore, Error, SeedableRng, le};

#[cfg(feature = "serde1")]
use serde::{Deserialize, Serialize};

// Deliberately poor constants for testing:
// 2562598503 - Lag-2 or 3 Truly awful spectra
// 2487410280 - Lag-2 or 3 Very bad spectra

const MULTIPLIER: u32 = 3487286589; //Suitable for lag-2,3,4 acceptably good spectra

/// A PCG random number generator (MWC X A 128/32 variant).
///
/// Permuted Congruential Generator with 128-bit state, internal multiply
/// with carry Generator, and 32-bit output via a xor and an add.
#[derive(Clone, PartialEq, Eq)]
#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))]
pub struct Mwc128XXA32 {
    pub(crate) x1: u32,
    pub(crate) x2: u32,
    pub(crate) x3: u32,
    pub(crate) c: u32,
}

impl Mwc128XXA32 {
    /// Construct an instance given two keys.
    pub fn new(k1: u32, k2: u32) -> Self {
        // X3 is 0xcafef00d 0xd15ea5e5 (default state from PCG paper because it cannot be 0.
        // C must be initialized to a value > 1 and < MULTIPLIER
        Mwc128XXA32::from_state_incr(k1, k2, 0xcafef00d, 0xd15ea5e5)
    }

    #[inline]
    fn from_state_incr(x1: u32, x2: u32, x3: u32, c: u32) -> Self {
        let mut pcg = Mwc128XXA32 { x1, x2, x3, c };
        //Advance 6 steps to fully mix the keys.
        pcg.gen6();
        pcg
    }

    pub fn next(&mut self) -> u32 {
        self.step()
    }

    #[inline]
    fn step(&mut self) -> u32 {
        // prepare the MCG for the next round
        let (low, hi) = multiply(self.x3);
        let result = permute(self.x1, self.x2, self.x3, self.c, low, hi);
        let (x1, b) = low.overflowing_add(self.c);
        self.x3 = self.x2;
        self.x2 = self.x1;
        self.x1 = x1;
        self.c = hi.wrapping_add(b as u32);
        result
    }

    #[inline]
    fn gen6(&mut self) -> [u32; 6] {
        //This is faster than calling `next_u32` 6 times because it avoids the intermediate assignments to the member variables.
        //For some reason the compiler doesn't figure this out automatically.
        let mut result = [0; 6];
        let (low, hi) = multiply(self.x3);
        result[0] = permute(self.x1, self.x2, self.x3, self.c, low, hi);
        let (r1, b) = low.overflowing_add(self.c);
        let c = hi.wrapping_add(b as u32);
        let (low, hi) = multiply(self.x2);
        result[1] = permute(r1, self.x1, self.x2, c, low, hi);
        let (r2, b) = low.overflowing_add(c);
        let c = hi.wrapping_add(b as u32);
        let (low, hi) = multiply(self.x1);
        result[2] = permute(r2, r1, self.x1, c, low, hi);
        let (r3, b) = low.overflowing_add(c);
        let c = hi.wrapping_add(b as u32);

        let (low, hi) = multiply(r1);
        result[3] = permute(r3, r2, r1, c, low, hi);
        let (r1, b) = low.overflowing_add(c);
        let c = hi.wrapping_add(b as u32);
        let (low, hi) = multiply(r2);
        result[4] = permute(r1, r3, r2, c, low, hi);
        let (r2, b) = low.overflowing_add(c);
        let c = hi.wrapping_add(b as u32);
        let (low, hi) = multiply(r3);
        result[5] = permute(r2, r1, r3, c, low, hi);
        let (r3, b) = low.overflowing_add(c);
        let c = hi.wrapping_add(b as u32);

        self.c = c;
        self.x1 = r3;
        self.x2 = r2;
        self.x3 = r1;
        return result;
    }
}


/// We use a single 121-bit seed to initialise the state and select a stream.
/// Of the 128 `seed` bits 7 are ignored.
impl SeedableRng for Mwc128XXA32 {
    type Seed = [u8; 16];

    fn from_seed(seed: Self::Seed) -> Self {
        let mut seed_u32 = [0u32; 4];
        le::read_u32_into(&seed, &mut seed_u32);
        // c must be < MULTIPLE and not all 1s or 0s
        let c = (seed_u32[0] & 0x3fff_fff8) | 5;
        // X3 must be non-zero and not all 1s, hence we discard 2 bits
        let x3 = (seed_u32[3] << 2) | 1;
        Mwc128XXA32::from_state_incr(seed_u32[1], seed_u32[2], x3, c)
    }
}

impl RngCore for Mwc128XXA32 {
    #[inline]
    fn next_u32(&mut self) -> u32 {
        self.step() as u32
    }

    #[inline]
    fn next_u64(&mut self) -> u64 {
        let result = self.step() as u64;
        return (result << 32) | (self.step() as u64);
    }

    #[inline]
    fn fill_bytes(&mut self, dest: &mut [u8]) {
        let mut dest_chunks = dest.chunks_exact_mut(6 * 4);
        for mut dest_chunk in &mut dest_chunks {
            for &num in self.gen6().iter() {
                let (l, r) = dest_chunk.split_at_mut(4);
                l.copy_from_slice(&num.to_le_bytes());
                dest_chunk = r;
            }
        }
        for dest_chunk in dest_chunks.into_remainder().chunks_mut(4) {
            dest_chunk.copy_from_slice(&self.step().to_le_bytes()[..dest_chunk.len()]);
        }
    }

    #[inline(always)]
    fn try_fill_bytes(&mut self, dest: &mut [u8]) -> Result<(), Error> {
        self.fill_bytes(dest);
        Ok(())
    }
}

#[inline(always)]
fn multiply(val: u32) -> (u32, u32) {
    let t = (val as u64).wrapping_mul(MULTIPLIER as u64);
    return (t as u32, (t >> 32) as u32);
}

#[inline(always)]
fn permute(x1: u32, x2: u32, x3: u32, _c: u32, _low: u32, hi: u32) -> u32 {
    (x3 ^ x2).wrapping_add(x1 ^ hi)
}