Coverage Report

Created: 2026-07-08 14:14

next uncovered line (L), next uncovered region (R), next uncovered branch (B)
/tmp/bitcoin/src/crypto/muhash.cpp
Line
Count
Source
1
// Copyright (c) 2017-present The Bitcoin Core developers
2
// Distributed under the MIT software license, see the accompanying
3
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
4
5
#include <crypto/muhash.h>
6
7
#include <crypto/chacha20.h>
8
#include <crypto/common.h>
9
#include <hash.h>
10
#include <span.h>
11
#include <uint256.h>
12
#include <util/check.h>
13
14
#include <bit>
15
#include <cstring>
16
#include <limits>
17
18
namespace {
19
20
using limb_t = Num3072::limb_t;
21
using signed_limb_t = Num3072::signed_limb_t;
22
using double_limb_t = Num3072::double_limb_t;
23
using signed_double_limb_t = Num3072::signed_double_limb_t;
24
constexpr int LIMB_SIZE = Num3072::LIMB_SIZE;
25
constexpr int SIGNED_LIMB_SIZE = Num3072::SIGNED_LIMB_SIZE;
26
constexpr int LIMBS = Num3072::LIMBS;
27
constexpr int SIGNED_LIMBS = Num3072::SIGNED_LIMBS;
28
constexpr int FINAL_LIMB_POSITION = 3072 / SIGNED_LIMB_SIZE;
29
constexpr int FINAL_LIMB_MODULUS_BITS = 3072 % SIGNED_LIMB_SIZE;
30
constexpr limb_t MAX_LIMB = (limb_t)(-1);
31
constexpr limb_t MAX_SIGNED_LIMB = (((limb_t)1) << SIGNED_LIMB_SIZE) - 1;
32
/** 2^3072 - 1103717, the largest 3072-bit safe prime number, is used as the modulus. */
33
constexpr limb_t MAX_PRIME_DIFF = 1103717;
34
/** The modular inverse of (2**3072 - MAX_PRIME_DIFF) mod (MAX_SIGNED_LIMB + 1). */
35
constexpr limb_t MODULUS_INVERSE = limb_t(0x70a1421da087d93);
36
37
38
/** Extract the lowest limb of [c0,c1,c2] into n, and left shift the number by 1 limb. */
39
inline void extract3(limb_t& c0, limb_t& c1, limb_t& c2, limb_t& n)
40
499k
{
41
499k
    n = c0;
42
499k
    c0 = c1;
43
499k
    c1 = c2;
44
499k
    c2 = 0;
45
499k
}
46
47
/** [c0,c1] = a * b */
48
inline void mul(limb_t& c0, limb_t& c1, const limb_t& a, const limb_t& b)
49
489k
{
50
489k
    double_limb_t t = (double_limb_t)a * b;
51
489k
    c1 = t >> LIMB_SIZE;
52
489k
    c0 = t;
53
489k
}
54
55
/* [c0,c1,c2] += n * [d0,d1,d2]. c2 is 0 initially */
56
inline void mulnadd3(limb_t& c0, limb_t& c1, limb_t& c2, limb_t& d0, limb_t& d1, limb_t& d2, const limb_t& n)
57
489k
{
58
489k
    double_limb_t t = (double_limb_t)d0 * n + c0;
59
489k
    c0 = t;
60
489k
    t >>= LIMB_SIZE;
61
489k
    t += (double_limb_t)d1 * n + c1;
62
489k
    c1 = t;
63
489k
    t >>= LIMB_SIZE;
64
489k
    c2 = t + d2 * n;
65
489k
}
66
67
/* [c0,c1] *= n */
68
inline void muln2(limb_t& c0, limb_t& c1, const limb_t& n)
69
10.4k
{
70
10.4k
    double_limb_t t = (double_limb_t)c0 * n;
71
10.4k
    c0 = t;
72
10.4k
    t >>= LIMB_SIZE;
73
10.4k
    t += (double_limb_t)c1 * n;
74
10.4k
    c1 = t;
75
10.4k
}
76
77
/** [c0,c1,c2] += a * b */
78
inline void muladd3(limb_t& c0, limb_t& c1, limb_t& c2, const limb_t& a, const limb_t& b)
79
23.5M
{
80
23.5M
    double_limb_t t = (double_limb_t)a * b;
81
23.5M
    limb_t th = t >> LIMB_SIZE;
82
23.5M
    limb_t tl = t;
83
84
23.5M
    c0 += tl;
85
23.5M
    th += (c0 < tl) ? 1 : 0;
86
23.5M
    c1 += th;
87
23.5M
    c2 += (c1 < th) ? 1 : 0;
88
23.5M
}
89
90
/**
91
 * Add limb a to [c0,c1]: [c0,c1] += a. Then extract the lowest
92
 * limb of [c0,c1] into n, and left shift the number by 1 limb.
93
 * */
94
inline void addnextract2(limb_t& c0, limb_t& c1, const limb_t& a, limb_t& n)
95
500k
{
96
500k
    limb_t c2 = 0;
97
98
    // add
99
500k
    c0 += a;
100
500k
    if (c0 < a) {
101
3.51k
        c1 += 1;
102
103
        // Handle case when c1 has overflown
104
3.51k
        if (c1 == 0) c2 = 1;
105
3.51k
    }
106
107
    // extract
108
500k
    n = c0;
109
500k
    c0 = c1;
110
500k
    c1 = c2;
111
500k
}
112
113
} // namespace
114
115
/** Indicates whether d is larger than the modulus. */
116
bool Num3072::IsOverflow() const
117
22.7k
{
118
22.7k
    if (this->limbs[0] <= std::numeric_limits<limb_t>::max() - MAX_PRIME_DIFF) return false;
119
528
    for (int i = 1; i < LIMBS; ++i) {
120
517
        if (this->limbs[i] != std::numeric_limits<limb_t>::max()) return false;
121
517
    }
122
11
    return true;
123
11
}
124
125
void Num3072::FullReduce()
126
11
{
127
11
    limb_t c0 = MAX_PRIME_DIFF;
128
11
    limb_t c1 = 0;
129
539
    for (int i = 0; i < LIMBS; ++i) {
130
528
        addnextract2(c0, c1, this->limbs[i], this->limbs[i]);
131
528
    }
132
11
}
133
134
namespace {
135
/** A type representing a number in signed limb representation. */
136
struct Num3072Signed
137
{
138
    /** The represented value is sum(limbs[i]*2^(SIGNED_LIMB_SIZE*i), i=0..SIGNED_LIMBS-1).
139
     *  Note that limbs may be negative, or exceed 2^SIGNED_LIMB_SIZE-1. */
140
    signed_limb_t limbs[SIGNED_LIMBS];
141
142
    /** Construct a Num3072Signed with value 0. */
143
    Num3072Signed()
144
16.4k
    {
145
16.4k
        memset(limbs, 0, sizeof(limbs));
146
16.4k
    }
147
148
    /** Convert a Num3072 to a Num3072Signed. Output will be normalized and in
149
     *  range 0..2^3072-1. */
150
    void FromNum3072(const Num3072& in)
151
4.11k
    {
152
4.11k
        double_limb_t c = 0;
153
4.11k
        int b = 0, outpos = 0;
154
201k
        for (int i = 0; i < LIMBS; ++i) {
155
197k
            c += double_limb_t{in.limbs[i]} << b;
156
197k
            b += LIMB_SIZE;
157
399k
            while (b >= SIGNED_LIMB_SIZE) {
158
201k
                limbs[outpos++] = limb_t(c) & MAX_SIGNED_LIMB;
159
201k
                c >>= SIGNED_LIMB_SIZE;
160
201k
                b -= SIGNED_LIMB_SIZE;
161
201k
            }
162
197k
        }
163
4.11k
        Assume(outpos == SIGNED_LIMBS - 1);
164
4.11k
        limbs[SIGNED_LIMBS - 1] = c;
165
4.11k
        c >>= SIGNED_LIMB_SIZE;
166
4.11k
        Assume(c == 0);
167
4.11k
    }
168
169
    /** Convert a Num3072Signed to a Num3072. Input must be in range 0..modulus-1. */
170
    void ToNum3072(Num3072& out) const
171
4.11k
    {
172
4.11k
        double_limb_t c = 0;
173
4.11k
        int b = 0, outpos = 0;
174
209k
        for (int i = 0; i < SIGNED_LIMBS; ++i) {
175
205k
            c += double_limb_t(limbs[i]) << b;
176
205k
            b += SIGNED_LIMB_SIZE;
177
205k
            if (b >= LIMB_SIZE) {
178
197k
                out.limbs[outpos++] = c;
179
197k
                c >>= LIMB_SIZE;
180
197k
                b -= LIMB_SIZE;
181
197k
            }
182
205k
        }
183
4.11k
        Assume(outpos == LIMBS);
184
4.11k
        Assume(c == 0);
185
4.11k
    }
186
187
    /** Take a Num3072Signed in range 1-2*2^3072..2^3072-1, and:
188
     *  - optionally negate it (if negate is true)
189
     *  - reduce it modulo the modulus (2^3072 - MAX_PRIME_DIFF)
190
     *  - produce output with all limbs in range 0..2^SIGNED_LIMB_SIZE-1
191
     */
192
    void Normalize(bool negate)
193
4.11k
    {
194
        // Add modulus if this was negative. This brings the range of *this to 1-2^3072..2^3072-1.
195
4.11k
        signed_limb_t cond_add = limbs[SIGNED_LIMBS-1] >> (LIMB_SIZE-1); // -1 if this is negative; 0 otherwise
196
4.11k
        limbs[0] += signed_limb_t(-MAX_PRIME_DIFF) & cond_add;
197
4.11k
        limbs[FINAL_LIMB_POSITION] += (signed_limb_t(1) << FINAL_LIMB_MODULUS_BITS) & cond_add;
198
        // Next negate all limbs if negate was set. This does not change the range of *this.
199
4.11k
        signed_limb_t cond_negate = -signed_limb_t(negate); // -1 if this negate is true; 0 otherwise
200
209k
        for (int i = 0; i < SIGNED_LIMBS; ++i) {
201
205k
            limbs[i] = (limbs[i] ^ cond_negate) - cond_negate;
202
205k
        }
203
        // Perform carry (make all limbs except the top one be in range 0..2^SIGNED_LIMB_SIZE-1).
204
205k
        for (int i = 0; i < SIGNED_LIMBS - 1; ++i) {
205
201k
            limbs[i + 1] += limbs[i] >> SIGNED_LIMB_SIZE;
206
201k
            limbs[i] &= MAX_SIGNED_LIMB;
207
201k
        }
208
        // Again add modulus if *this was negative. This brings the range of *this to 0..2^3072-1.
209
4.11k
        cond_add = limbs[SIGNED_LIMBS-1] >> (LIMB_SIZE-1); // -1 if this is negative; 0 otherwise
210
4.11k
        limbs[0] += signed_limb_t(-MAX_PRIME_DIFF) & cond_add;
211
4.11k
        limbs[FINAL_LIMB_POSITION] += (signed_limb_t(1) << FINAL_LIMB_MODULUS_BITS) & cond_add;
212
        // Perform another carry. Now all limbs are in range 0..2^SIGNED_LIMB_SIZE-1.
213
205k
        for (int i = 0; i < SIGNED_LIMBS - 1; ++i) {
214
201k
            limbs[i + 1] += limbs[i] >> SIGNED_LIMB_SIZE;
215
201k
            limbs[i] &= MAX_SIGNED_LIMB;
216
201k
        }
217
4.11k
    }
218
};
219
220
/** 2x2 transformation matrix with signed_limb_t elements. */
221
struct SignedMatrix
222
{
223
    signed_limb_t u, v, q, r;
224
};
225
226
/** Compute the transformation matrix for SIGNED_LIMB_SIZE divsteps.
227
 *
228
 * eta: initial eta value
229
 * f:   bottom SIGNED_LIMB_SIZE bits of initial f value
230
 * g:   bottom SIGNED_LIMB_SIZE bits of initial g value
231
 * out: resulting transformation matrix, scaled by 2^SIGNED_LIMB_SIZE
232
 * return: eta value after SIGNED_LIMB_SIZE divsteps
233
 */
234
inline limb_t ComputeDivstepMatrix(signed_limb_t eta, limb_t f, limb_t g, SignedMatrix& out)
235
412k
{
236
    /** inv256[i] = -1/(2*i+1) (mod 256) */
237
412k
    static const uint8_t NEGINV256[128] = {
238
412k
        0xFF, 0x55, 0x33, 0x49, 0xC7, 0x5D, 0x3B, 0x11, 0x0F, 0xE5, 0xC3, 0x59,
239
412k
        0xD7, 0xED, 0xCB, 0x21, 0x1F, 0x75, 0x53, 0x69, 0xE7, 0x7D, 0x5B, 0x31,
240
412k
        0x2F, 0x05, 0xE3, 0x79, 0xF7, 0x0D, 0xEB, 0x41, 0x3F, 0x95, 0x73, 0x89,
241
412k
        0x07, 0x9D, 0x7B, 0x51, 0x4F, 0x25, 0x03, 0x99, 0x17, 0x2D, 0x0B, 0x61,
242
412k
        0x5F, 0xB5, 0x93, 0xA9, 0x27, 0xBD, 0x9B, 0x71, 0x6F, 0x45, 0x23, 0xB9,
243
412k
        0x37, 0x4D, 0x2B, 0x81, 0x7F, 0xD5, 0xB3, 0xC9, 0x47, 0xDD, 0xBB, 0x91,
244
412k
        0x8F, 0x65, 0x43, 0xD9, 0x57, 0x6D, 0x4B, 0xA1, 0x9F, 0xF5, 0xD3, 0xE9,
245
412k
        0x67, 0xFD, 0xDB, 0xB1, 0xAF, 0x85, 0x63, 0xF9, 0x77, 0x8D, 0x6B, 0xC1,
246
412k
        0xBF, 0x15, 0xF3, 0x09, 0x87, 0x1D, 0xFB, 0xD1, 0xCF, 0xA5, 0x83, 0x19,
247
412k
        0x97, 0xAD, 0x8B, 0xE1, 0xDF, 0x35, 0x13, 0x29, 0xA7, 0x3D, 0x1B, 0xF1,
248
412k
        0xEF, 0xC5, 0xA3, 0x39, 0xB7, 0xCD, 0xAB, 0x01
249
412k
    };
250
    // Coefficients of returned SignedMatrix; starts off as identity matrix. */
251
412k
    limb_t u = 1, v = 0, q = 0, r = 1;
252
    // The number of divsteps still left.
253
412k
    int i = SIGNED_LIMB_SIZE;
254
2.64M
    while (true) {
255
        /* Use a sentinel bit to count zeros only up to i. */
256
2.64M
        int zeros = std::countr_zero(g | (MAX_LIMB << i));
257
        /* Perform zeros divsteps at once; they all just divide g by two. */
258
2.64M
        g >>= zeros;
259
2.64M
        u <<= zeros;
260
2.64M
        v <<= zeros;
261
2.64M
        eta -= zeros;
262
2.64M
        i -= zeros;
263
         /* We're done once we've performed SIGNED_LIMB_SIZE divsteps. */
264
2.64M
        if (i == 0) break;
265
        /* If eta is negative, negate it and replace f,g with g,-f. */
266
2.23M
        if (eta < 0) {
267
770k
            limb_t tmp;
268
770k
            eta = -eta;
269
770k
            tmp = f; f = g; g = -tmp;
270
770k
            tmp = u; u = q; q = -tmp;
271
770k
            tmp = v; v = r; r = -tmp;
272
770k
        }
273
        /* eta is now >= 0. In what follows we're going to cancel out the bottom bits of g. No more
274
         * than i can be cancelled out (as we'd be done before that point), and no more than eta+1
275
         * can be done as its sign will flip once that happens. */
276
2.23M
        int limit = ((int)eta + 1) > i ? i : ((int)eta + 1);
277
        /* m is a mask for the bottom min(limit, 8) bits (our table only supports 8 bits). */
278
2.23M
        limb_t m = (MAX_LIMB >> (LIMB_SIZE - limit)) & 255U;
279
        /* Find what multiple of f must be added to g to cancel its bottom min(limit, 8) bits. */
280
2.23M
        limb_t w = (g * NEGINV256[(f >> 1) & 127]) & m;
281
        /* Do so. */
282
2.23M
        g += f * w;
283
2.23M
        q += u * w;
284
2.23M
        r += v * w;
285
2.23M
    }
286
412k
    out.u = (signed_limb_t)u;
287
412k
    out.v = (signed_limb_t)v;
288
412k
    out.q = (signed_limb_t)q;
289
412k
    out.r = (signed_limb_t)r;
290
412k
    return eta;
291
412k
}
292
293
/** Apply matrix t/2^SIGNED_LIMB_SIZE to vector [d,e], modulo modulus.
294
 *
295
 * On input and output, d and e are in range 1-2*modulus..modulus-1.
296
 */
297
inline void UpdateDE(Num3072Signed& d, Num3072Signed& e, const SignedMatrix& t)
298
412k
{
299
412k
    const signed_limb_t u = t.u, v=t.v, q=t.q, r=t.r;
300
301
    /* [md,me] start as zero; plus [u,q] if d is negative; plus [v,r] if e is negative. */
302
412k
    signed_limb_t sd = d.limbs[SIGNED_LIMBS - 1] >> (LIMB_SIZE - 1);
303
412k
    signed_limb_t se = e.limbs[SIGNED_LIMBS - 1] >> (LIMB_SIZE - 1);
304
412k
    signed_limb_t md = (u & sd) + (v & se);
305
412k
    signed_limb_t me = (q & sd) + (r & se);
306
    /* Begin computing t*[d,e]. */
307
412k
    signed_limb_t di = d.limbs[0], ei = e.limbs[0];
308
412k
    signed_double_limb_t cd = (signed_double_limb_t)u * di + (signed_double_limb_t)v * ei;
309
412k
    signed_double_limb_t ce = (signed_double_limb_t)q * di + (signed_double_limb_t)r * ei;
310
    /* Correct md,me so that t*[d,e]+modulus*[md,me] has SIGNED_LIMB_SIZE zero bottom bits. */
311
412k
    md -= (MODULUS_INVERSE * limb_t(cd) + md) & MAX_SIGNED_LIMB;
312
412k
    me -= (MODULUS_INVERSE * limb_t(ce) + me) & MAX_SIGNED_LIMB;
313
    /* Update the beginning of computation for t*[d,e]+modulus*[md,me] now md,me are known. */
314
412k
    cd -= (signed_double_limb_t)1103717 * md;
315
412k
    ce -= (signed_double_limb_t)1103717 * me;
316
    /* Verify that the low SIGNED_LIMB_SIZE bits of the computation are indeed zero, and then throw them away. */
317
412k
    Assume((cd & MAX_SIGNED_LIMB) == 0);
318
412k
    Assume((ce & MAX_SIGNED_LIMB) == 0);
319
412k
    cd >>= SIGNED_LIMB_SIZE;
320
412k
    ce >>= SIGNED_LIMB_SIZE;
321
    /* Now iteratively compute limb i=1..SIGNED_LIMBS-2 of t*[d,e]+modulus*[md,me], and store them in output
322
     * limb i-1 (shifting down by SIGNED_LIMB_SIZE bits). The corresponding limbs in modulus are all zero,
323
     * so modulus/md/me are not actually involved here. */
324
20.2M
    for (int i = 1; i < SIGNED_LIMBS - 1; ++i) {
325
19.8M
        di = d.limbs[i];
326
19.8M
        ei = e.limbs[i];
327
19.8M
        cd += (signed_double_limb_t)u * di + (signed_double_limb_t)v * ei;
328
19.8M
        ce += (signed_double_limb_t)q * di + (signed_double_limb_t)r * ei;
329
19.8M
        d.limbs[i - 1] = (signed_limb_t)cd & MAX_SIGNED_LIMB; cd >>= SIGNED_LIMB_SIZE;
330
19.8M
        e.limbs[i - 1] = (signed_limb_t)ce & MAX_SIGNED_LIMB; ce >>= SIGNED_LIMB_SIZE;
331
19.8M
    }
332
    /* Compute limb SIGNED_LIMBS-1 of t*[d,e]+modulus*[md,me], and store it in output limb SIGNED_LIMBS-2. */
333
412k
    di = d.limbs[SIGNED_LIMBS - 1];
334
412k
    ei = e.limbs[SIGNED_LIMBS - 1];
335
412k
    cd += (signed_double_limb_t)u * di + (signed_double_limb_t)v * ei;
336
412k
    ce += (signed_double_limb_t)q * di + (signed_double_limb_t)r * ei;
337
412k
    cd += (signed_double_limb_t)md << FINAL_LIMB_MODULUS_BITS;
338
412k
    ce += (signed_double_limb_t)me << FINAL_LIMB_MODULUS_BITS;
339
412k
    d.limbs[SIGNED_LIMBS - 2] = (signed_limb_t)cd & MAX_SIGNED_LIMB; cd >>= SIGNED_LIMB_SIZE;
340
412k
    e.limbs[SIGNED_LIMBS - 2] = (signed_limb_t)ce & MAX_SIGNED_LIMB; ce >>= SIGNED_LIMB_SIZE;
341
    /* What remains goes into output limb SINGED_LIMBS-1 */
342
412k
    d.limbs[SIGNED_LIMBS - 1] = (signed_limb_t)cd;
343
412k
    e.limbs[SIGNED_LIMBS - 1] = (signed_limb_t)ce;
344
412k
}
345
346
/** Apply matrix t/2^SIGNED_LIMB_SIZE to vector (f,g).
347
 *
348
 * The matrix t must be chosen such that t*(f,g) results in multiples of 2^SIGNED_LIMB_SIZE.
349
 * This is the case for matrices computed by ComputeDivstepMatrix().
350
 */
351
inline void UpdateFG(Num3072Signed& f, Num3072Signed& g, const SignedMatrix& t, int len)
352
412k
{
353
412k
    const signed_limb_t u = t.u, v=t.v, q=t.q, r=t.r;
354
355
412k
    signed_limb_t fi, gi;
356
412k
    signed_double_limb_t cf, cg;
357
    /* Start computing t*[f,g]. */
358
412k
    fi = f.limbs[0];
359
412k
    gi = g.limbs[0];
360
412k
    cf = (signed_double_limb_t)u * fi + (signed_double_limb_t)v * gi;
361
412k
    cg = (signed_double_limb_t)q * fi + (signed_double_limb_t)r * gi;
362
    /* Verify that the bottom SIGNED_LIMB_BITS bits of the result are zero, and then throw them away. */
363
412k
    Assume((cf & MAX_SIGNED_LIMB) == 0);
364
412k
    Assume((cg & MAX_SIGNED_LIMB) == 0);
365
412k
    cf >>= SIGNED_LIMB_SIZE;
366
412k
    cg >>= SIGNED_LIMB_SIZE;
367
    /* Now iteratively compute limb i=1..SIGNED_LIMBS-1 of t*[f,g], and store them in output limb i-1 (shifting
368
     * down by SIGNED_LIMB_BITS bits). */
369
14.8M
    for (int i = 1; i < len; ++i) {
370
14.4M
        fi = f.limbs[i];
371
14.4M
        gi = g.limbs[i];
372
14.4M
        cf += (signed_double_limb_t)u * fi + (signed_double_limb_t)v * gi;
373
14.4M
        cg += (signed_double_limb_t)q * fi + (signed_double_limb_t)r * gi;
374
14.4M
        f.limbs[i - 1] = (signed_limb_t)cf & MAX_SIGNED_LIMB; cf >>= SIGNED_LIMB_SIZE;
375
14.4M
        g.limbs[i - 1] = (signed_limb_t)cg & MAX_SIGNED_LIMB; cg >>= SIGNED_LIMB_SIZE;
376
14.4M
    }
377
    /* What remains is limb SIGNED_LIMBS of t*[f,g]; store it as output limb SIGNED_LIMBS-1. */
378
412k
    f.limbs[len - 1] = (signed_limb_t)cf;
379
412k
    g.limbs[len - 1] = (signed_limb_t)cg;
380
381
412k
}
382
} // namespace
383
384
Num3072 Num3072::GetInverse() const
385
4.11k
{
386
    // Compute a modular inverse based on a variant of the safegcd algorithm:
387
    // - Paper: https://gcd.cr.yp.to/papers.html
388
    // - Inspired by this code in libsecp256k1:
389
    //   https://github.com/bitcoin-core/secp256k1/blob/master/src/modinv32_impl.h
390
    // - Explanation of the algorithm:
391
    //   https://github.com/bitcoin-core/secp256k1/blob/master/doc/safegcd_implementation.md
392
393
    // Local variables d, e, f, g:
394
    // - f and g are the variables whose gcd we compute (despite knowing the answer is 1):
395
    //   - f is always odd, and initialized as modulus
396
    //   - g is initialized as *this (called x in what follows)
397
    // - d and e are the numbers for which at every step it is the case that:
398
    //   - f = d * x mod modulus; d is initialized as 0
399
    //   - g = e * x mod modulus; e is initialized as 1
400
4.11k
    Num3072Signed d, e, f, g;
401
4.11k
    e.limbs[0] = 1;
402
    // F is initialized as modulus, which in signed limb representation can be expressed
403
    // simply as 2^3072 + -MAX_PRIME_DIFF.
404
4.11k
    f.limbs[0] = -MAX_PRIME_DIFF;
405
4.11k
    f.limbs[FINAL_LIMB_POSITION] = ((limb_t)1) << FINAL_LIMB_MODULUS_BITS;
406
4.11k
    g.FromNum3072(*this);
407
4.11k
    int len = SIGNED_LIMBS; //!< The number of significant limbs in f and g
408
4.11k
    signed_limb_t eta = -1; //!< State to track knowledge about ratio of f and g
409
    // Perform divsteps on [f,g] until g=0 is reached, keeping (d,e) synchronized with them.
410
412k
    while (true) {
411
        // Compute transformation matrix t that represents the next SIGNED_LIMB_SIZE divsteps
412
        // to apply. This can be computed from just the bottom limb of f and g, and eta.
413
412k
        SignedMatrix t;
414
412k
        eta = ComputeDivstepMatrix(eta, f.limbs[0], g.limbs[0], t);
415
        // Apply that transformation matrix to the full [f,g] vector.
416
412k
        UpdateFG(f, g, t, len);
417
        // Apply that transformation matrix to the full [d,e] vector (mod modulus).
418
412k
        UpdateDE(d, e, t);
419
420
        // Check if g is zero.
421
412k
        if (g.limbs[0] == 0) {
422
181k
            signed_limb_t cond = 0;
423
8.86M
            for (int j = 1; j < len; ++j) {
424
8.68M
                cond |= g.limbs[j];
425
8.68M
            }
426
            // If so, we're done.
427
181k
            if (cond == 0) break;
428
181k
        }
429
430
        // Check if the top limbs of both f and g are both 0 or -1.
431
408k
        signed_limb_t fn = f.limbs[len - 1], gn = g.limbs[len - 1];
432
408k
        signed_limb_t cond = ((signed_limb_t)len - 2) >> (LIMB_SIZE - 1);
433
408k
        cond |= fn ^ (fn >> (LIMB_SIZE - 1));
434
408k
        cond |= gn ^ (gn >> (LIMB_SIZE - 1));
435
408k
        if (cond == 0) {
436
            // If so, drop the top limb, shrinking the size of f and g, by
437
            // propagating the sign to the previous limb.
438
201k
            f.limbs[len - 2] |= (limb_t)f.limbs[len - 1] << SIGNED_LIMB_SIZE;
439
201k
            g.limbs[len - 2] |= (limb_t)g.limbs[len - 1] << SIGNED_LIMB_SIZE;
440
201k
            --len;
441
201k
        }
442
408k
    }
443
    // At some point, [f,g] will have been rewritten into [f',0], such that gcd(f,g) = gcd(f',0).
444
    // This is proven in the paper. As f started out being modulus, a prime number, we know that
445
    // gcd is 1, and thus f' is 1 or -1.
446
4.11k
    Assume((f.limbs[0] & MAX_SIGNED_LIMB) == 1 || (f.limbs[0] & MAX_SIGNED_LIMB) == MAX_SIGNED_LIMB);
447
    // As we've maintained the invariant that f = d * x mod modulus, we get d/f mod modulus is the
448
    // modular inverse of x we're looking for. As f is 1 or -1, it is also true that d/f = d*f.
449
    // Normalize d to prepare it for output, while negating it if f is negative.
450
4.11k
    d.Normalize(f.limbs[len - 1] >> (LIMB_SIZE  - 1));
451
4.11k
    Num3072 ret;
452
4.11k
    d.ToNum3072(ret);
453
4.11k
    return ret;
454
4.11k
}
455
456
void Num3072::Multiply(const Num3072& a)
457
10.4k
{
458
10.4k
    limb_t c0 = 0, c1 = 0, c2 = 0;
459
10.4k
    Num3072 tmp;
460
461
    /* Compute limbs 0..N-2 of this*a into tmp, including one reduction. */
462
499k
    for (int j = 0; j < LIMBS - 1; ++j) {
463
489k
        limb_t d0 = 0, d1 = 0, d2 = 0;
464
489k
        mul(d0, d1, this->limbs[1 + j], a.limbs[LIMBS + j - (1 + j)]);
465
11.7M
        for (int i = 2 + j; i < LIMBS; ++i) muladd3(d0, d1, d2, this->limbs[i], a.limbs[LIMBS + j - i]);
466
489k
        mulnadd3(c0, c1, c2, d0, d1, d2, MAX_PRIME_DIFF);
467
12.2M
        for (int i = 0; i < j + 1; ++i) muladd3(c0, c1, c2, this->limbs[i], a.limbs[j - i]);
468
489k
        extract3(c0, c1, c2, tmp.limbs[j]);
469
489k
    }
470
471
    /* Compute limb N-1 of a*b into tmp. */
472
10.4k
    assert(c2 == 0);
473
510k
    for (int i = 0; i < LIMBS; ++i) muladd3(c0, c1, c2, this->limbs[i], a.limbs[LIMBS - 1 - i]);
474
10.4k
    extract3(c0, c1, c2, tmp.limbs[LIMBS - 1]);
475
476
    /* Perform a second reduction. */
477
10.4k
    muln2(c0, c1, MAX_PRIME_DIFF);
478
510k
    for (int j = 0; j < LIMBS; ++j) {
479
499k
        addnextract2(c0, c1, tmp.limbs[j], this->limbs[j]);
480
499k
    }
481
482
10.4k
    assert(c1 == 0);
483
10.4k
    assert(c0 == 0 || c0 == 1);
484
485
    /* Perform up to two more reductions if the internal state has already
486
     * overflown the MAX of Num3072 or if it is larger than the modulus or
487
     * if both are the case.
488
     * */
489
10.4k
    if (this->IsOverflow()) this->FullReduce();
490
10.4k
    if (c0) this->FullReduce();
491
10.4k
}
492
493
void Num3072::SetToOne()
494
23.3k
{
495
23.3k
    this->limbs[0] = 1;
496
1.12M
    for (int i = 1; i < LIMBS; ++i) this->limbs[i] = 0;
497
23.3k
}
498
499
void Num3072::Divide(const Num3072& a)
500
4.11k
{
501
4.11k
    if (this->IsOverflow()) this->FullReduce();
502
503
4.11k
    Num3072 inv{};
504
4.11k
    if (a.IsOverflow()) {
505
0
        Num3072 b = a;
506
0
        b.FullReduce();
507
0
        inv = b.GetInverse();
508
4.11k
    } else {
509
4.11k
        inv = a.GetInverse();
510
4.11k
    }
511
512
4.11k
    this->Multiply(inv);
513
4.11k
    if (this->IsOverflow()) this->FullReduce();
514
4.11k
}
515
516
6.07k
Num3072::Num3072(const unsigned char (&data)[BYTE_SIZE]) {
517
297k
    for (int i = 0; i < LIMBS; ++i) {
518
291k
        if (sizeof(limb_t) == 4) {
519
0
            this->limbs[i] = ReadLE32(data + 4 * i);
520
291k
        } else if (sizeof(limb_t) == 8) {
521
291k
            this->limbs[i] = ReadLE64(data + 8 * i);
522
291k
        }
523
291k
    }
524
6.07k
}
525
526
4.11k
void Num3072::ToBytes(unsigned char (&out)[BYTE_SIZE]) {
527
201k
    for (int i = 0; i < LIMBS; ++i) {
528
197k
        if (sizeof(limb_t) == 4) {
529
0
            WriteLE32(out + i * 4, this->limbs[i]);
530
197k
        } else if (sizeof(limb_t) == 8) {
531
197k
            WriteLE64(out + i * 8, this->limbs[i]);
532
197k
        }
533
197k
    }
534
4.11k
}
535
536
6.07k
Num3072 MuHash3072::ToNum3072(std::span<const unsigned char> in) {
537
6.07k
    unsigned char tmp[Num3072::BYTE_SIZE];
538
539
6.07k
    uint256 hashed_in{(HashWriter{} << in).GetSHA256()};
540
6.07k
    static_assert(sizeof(tmp) % ChaCha20Aligned::BLOCKLEN == 0);
541
6.07k
    ChaCha20Aligned{MakeByteSpan(hashed_in)}.Keystream(MakeWritableByteSpan(tmp));
542
6.07k
    Num3072 out{tmp};
543
544
6.07k
    return out;
545
6.07k
}
546
547
MuHash3072::MuHash3072(std::span<const unsigned char> in) noexcept
548
186
{
549
186
    m_numerator = ToNum3072(in);
550
186
}
551
552
void MuHash3072::Finalize(uint256& out) noexcept
553
4.11k
{
554
4.11k
    m_numerator.Divide(m_denominator);
555
4.11k
    m_denominator.SetToOne();  // Needed to keep the MuHash object valid
556
557
4.11k
    unsigned char data[Num3072::BYTE_SIZE];
558
4.11k
    m_numerator.ToBytes(data);
559
560
4.11k
    out = (HashWriter{} << data).GetSHA256();
561
4.11k
}
562
563
MuHash3072& MuHash3072::operator*=(const MuHash3072& mul) noexcept
564
124
{
565
124
    m_numerator.Multiply(mul.m_numerator);
566
124
    m_denominator.Multiply(mul.m_denominator);
567
124
    return *this;
568
124
}
569
570
MuHash3072& MuHash3072::operator/=(const MuHash3072& div) noexcept
571
79
{
572
79
    m_numerator.Multiply(div.m_denominator);
573
79
    m_denominator.Multiply(div.m_numerator);
574
79
    return *this;
575
79
}
576
577
5.40k
MuHash3072& MuHash3072::Insert(std::span<const unsigned char> in) noexcept {
578
5.40k
    m_numerator.Multiply(ToNum3072(in));
579
5.40k
    return *this;
580
5.40k
}
581
582
484
MuHash3072& MuHash3072::Remove(std::span<const unsigned char> in) noexcept {
583
484
    m_denominator.Multiply(ToNum3072(in));
584
484
    return *this;
585
484
}