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
|
/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this file,
* You can obtain one at http://mozilla.org/MPL/2.0/. */
#include <math.h>
#include "mozilla/Assertions.h"
#include "mozilla/PodOperations.h"
#include "mozilla/XorShift128PlusRNG.h"
using mozilla::non_crypto::XorShift128PlusRNG;
static void TestDumbSequence() {
XorShift128PlusRNG rng(1, 4);
// Calculated by hand following the algorithm given in the paper. The upper
// bits are mostly zero because we started with a poor seed; once it has run
// for a while, we'll get an even mix of ones and zeros in all 64 bits.
MOZ_RELEASE_ASSERT(rng.next() == 0x800049);
MOZ_RELEASE_ASSERT(rng.next() == 0x3000186);
MOZ_RELEASE_ASSERT(rng.next() == 0x400003001145);
// Using ldexp here lets us write out the mantissa in hex, so we can compare
// them with the results generated by hand.
MOZ_RELEASE_ASSERT(rng.nextDouble() ==
ldexp(static_cast<double>(0x1400003105049), -53));
MOZ_RELEASE_ASSERT(rng.nextDouble() ==
ldexp(static_cast<double>(0x2000802e49146), -53));
MOZ_RELEASE_ASSERT(rng.nextDouble() ==
ldexp(static_cast<double>(0x248300468544d), -53));
}
static size_t Population(uint64_t n) {
size_t pop = 0;
while (n > 0) {
n &= n - 1; // Clear the rightmost 1-bit in n.
pop++;
}
return pop;
}
static void TestPopulation() {
XorShift128PlusRNG rng(698079309544035222ULL, 6012389156611637584ULL);
// Give it some time to warm up; it should tend towards more
// even distributions of zeros and ones.
for (size_t i = 0; i < 40; i++) rng.next();
for (size_t i = 0; i < 40; i++) {
size_t pop = Population(rng.next());
MOZ_RELEASE_ASSERT(24 <= pop && pop <= 40);
}
}
static void TestSetState() {
static const uint64_t seed[2] = {1795644156779822404ULL,
14162896116325912595ULL};
XorShift128PlusRNG rng(seed[0], seed[1]);
const size_t n = 10;
uint64_t log[n];
for (size_t i = 0; i < n; i++) log[i] = rng.next();
rng.setState(seed[0], seed[1]);
for (size_t i = 0; i < n; i++) MOZ_RELEASE_ASSERT(log[i] == rng.next());
}
static void TestDoubleDistribution() {
XorShift128PlusRNG rng(0xa207aaede6859736, 0xaca6ca5060804791);
const size_t n = 100;
size_t bins[n];
mozilla::PodArrayZero(bins);
// This entire file runs in 0.006s on my laptop. Generating
// more numbers lets us put tighter bounds on the bins.
for (size_t i = 0; i < 100000; i++) {
double d = rng.nextDouble();
MOZ_RELEASE_ASSERT(0.0 <= d && d < 1.0);
bins[(int)(d * n)]++;
}
for (size_t i = 0; i < n; i++) {
MOZ_RELEASE_ASSERT(900 <= bins[i] && bins[i] <= 1100);
}
}
int main() {
TestDumbSequence();
TestPopulation();
TestSetState();
TestDoubleDistribution();
return 0;
}
|
