Coverage Report

Created: 2026-07-23 20:35

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
528k
{
41
528k
    n = c0;
42
528k
    c0 = c1;
43
528k
    c1 = c2;
44
528k
    c2 = 0;
45
528k
}
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
517k
{
50
517k
    double_limb_t t = (double_limb_t)a * b;
51
517k
    c1 = t >> LIMB_SIZE;
52
517k
    c0 = t;
53
517k
}
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
517k
{
58
517k
    double_limb_t t = (double_limb_t)d0 * n + c0;
59
517k
    c0 = t;
60
517k
    t >>= LIMB_SIZE;
61
517k
    t += (double_limb_t)d1 * n + c1;
62
517k
    c1 = t;
63
517k
    t >>= LIMB_SIZE;
64
517k
    c2 = t + d2 * n;
65
517k
}
66
67
/* [c0,c1] *= n */
68
inline void muln2(limb_t& c0, limb_t& c1, const limb_t& n)
69
11.0k
{
70
11.0k
    double_limb_t t = (double_limb_t)c0 * n;
71
11.0k
    c0 = t;
72
11.0k
    t >>= LIMB_SIZE;
73
11.0k
    t += (double_limb_t)c1 * n;
74
11.0k
    c1 = t;
75
11.0k
}
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
24.8M
{
80
24.8M
    double_limb_t t = (double_limb_t)a * b;
81
24.8M
    limb_t th = t >> LIMB_SIZE;
82
24.8M
    limb_t tl = t;
83
84
24.8M
    c0 += tl;
85
24.8M
    th += (c0 < tl) ? 1 : 0;
86
24.8M
    c1 += th;
87
24.8M
    c2 += (c1 < th) ? 1 : 0;
88
24.8M
}
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
528k
{
96
528k
    limb_t c2 = 0;
97
98
    // add
99
528k
    c0 += a;
100
528k
    if (c0 < a) {
101
3.66k
        c1 += 1;
102
103
        // Handle case when c1 has overflown
104
3.66k
        if (c1 == 0) c2 = 1;
105
3.66k
    }
106
107
    // extract
108
528k
    n = c0;
109
528k
    c0 = c1;
110
528k
    c1 = c2;
111
528k
}
112
113
} // namespace
114
115
/** Indicates whether d is larger than the modulus. */
116
bool Num3072::IsOverflow() const
117
24.2k
{
118
24.2k
    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
17.6k
    {
145
17.6k
        memset(limbs, 0, sizeof(limbs));
146
17.6k
    }
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
4.41k
    {
152
4.41k
        double_limb_t c = 0;
153
4.41k
        int b = 0, outpos = 0;
154
216k
        for (int i = 0; i < LIMBS; ++i) {
155
211k
            c += double_limb_t{in.limbs[i]} << b;
156
211k
            b += LIMB_SIZE;
157
428k
            while (b >= SIGNED_LIMB_SIZE) {
158
216k
                limbs[outpos++] = limb_t(c) & MAX_SIGNED_LIMB;
159
216k
                c >>= SIGNED_LIMB_SIZE;
160
216k
                b -= SIGNED_LIMB_SIZE;
161
216k
            }
162
211k
        }
163
4.41k
        Assume(outpos == SIGNED_LIMBS - 1);
164
4.41k
        limbs[SIGNED_LIMBS - 1] = c;
165
4.41k
        c >>= SIGNED_LIMB_SIZE;
166
4.41k
        Assume(c == 0);
167
4.41k
    }
168
169
    /** Convert a Num3072Signed to a Num3072. Input must be in range 0..modulus-1. */
170
    void ToNum3072(Num3072& out) const
171
4.41k
    {
172
4.41k
        double_limb_t c = 0;
173
4.41k
        int b = 0, outpos = 0;
174
225k
        for (int i = 0; i < SIGNED_LIMBS; ++i) {
175
220k
            c += double_limb_t(limbs[i]) << b;
176
220k
            b += SIGNED_LIMB_SIZE;
177
220k
            if (b >= LIMB_SIZE) {
178
211k
                out.limbs[outpos++] = c;
179
211k
                c >>= LIMB_SIZE;
180
211k
                b -= LIMB_SIZE;
181
211k
            }
182
220k
        }
183
4.41k
        Assume(outpos == LIMBS);
184
4.41k
        Assume(c == 0);
185
4.41k
    }
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
4.41k
    {
194
        // Add modulus if this was negative. This brings the range of *this to 1-2^3072..2^3072-1.
195
4.41k
        signed_limb_t cond_add = limbs[SIGNED_LIMBS-1] >> (LIMB_SIZE-1); // -1 if this is negative; 0 otherwise
196
4.41k
        limbs[0] += signed_limb_t(-MAX_PRIME_DIFF) & cond_add;
197
4.41k
        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
4.41k
        signed_limb_t cond_negate = -signed_limb_t(negate); // -1 if this negate is true; 0 otherwise
200
225k
        for (int i = 0; i < SIGNED_LIMBS; ++i) {
201
220k
            limbs[i] = (limbs[i] ^ cond_negate) - cond_negate;
202
220k
        }
203
        // Perform carry (make all limbs except the top one be in range 0..2^SIGNED_LIMB_SIZE-1).
204
220k
        for (int i = 0; i < SIGNED_LIMBS - 1; ++i) {
205
216k
            limbs[i + 1] += limbs[i] >> SIGNED_LIMB_SIZE;
206
216k
            limbs[i] &= MAX_SIGNED_LIMB;
207
216k
        }
208
        // Again add modulus if *this was negative. This brings the range of *this to 0..2^3072-1.
209
4.41k
        cond_add = limbs[SIGNED_LIMBS-1] >> (LIMB_SIZE-1); // -1 if this is negative; 0 otherwise
210
4.41k
        limbs[0] += signed_limb_t(-MAX_PRIME_DIFF) & cond_add;
211
4.41k
        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
220k
        for (int i = 0; i < SIGNED_LIMBS - 1; ++i) {
214
216k
            limbs[i + 1] += limbs[i] >> SIGNED_LIMB_SIZE;
215
216k
            limbs[i] &= MAX_SIGNED_LIMB;
216
216k
        }
217
4.41k
    }
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
442k
{
236
    /** inv256[i] = -1/(2*i+1) (mod 256) */
237
442k
    static const uint8_t NEGINV256[128] = {
238
442k
        0xFF, 0x55, 0x33, 0x49, 0xC7, 0x5D, 0x3B, 0x11, 0x0F, 0xE5, 0xC3, 0x59,
239
442k
        0xD7, 0xED, 0xCB, 0x21, 0x1F, 0x75, 0x53, 0x69, 0xE7, 0x7D, 0x5B, 0x31,
240
442k
        0x2F, 0x05, 0xE3, 0x79, 0xF7, 0x0D, 0xEB, 0x41, 0x3F, 0x95, 0x73, 0x89,
241
442k
        0x07, 0x9D, 0x7B, 0x51, 0x4F, 0x25, 0x03, 0x99, 0x17, 0x2D, 0x0B, 0x61,
242
442k
        0x5F, 0xB5, 0x93, 0xA9, 0x27, 0xBD, 0x9B, 0x71, 0x6F, 0x45, 0x23, 0xB9,
243
442k
        0x37, 0x4D, 0x2B, 0x81, 0x7F, 0xD5, 0xB3, 0xC9, 0x47, 0xDD, 0xBB, 0x91,
244
442k
        0x8F, 0x65, 0x43, 0xD9, 0x57, 0x6D, 0x4B, 0xA1, 0x9F, 0xF5, 0xD3, 0xE9,
245
442k
        0x67, 0xFD, 0xDB, 0xB1, 0xAF, 0x85, 0x63, 0xF9, 0x77, 0x8D, 0x6B, 0xC1,
246
442k
        0xBF, 0x15, 0xF3, 0x09, 0x87, 0x1D, 0xFB, 0xD1, 0xCF, 0xA5, 0x83, 0x19,
247
442k
        0x97, 0xAD, 0x8B, 0xE1, 0xDF, 0x35, 0x13, 0x29, 0xA7, 0x3D, 0x1B, 0xF1,
248
442k
        0xEF, 0xC5, 0xA3, 0x39, 0xB7, 0xCD, 0xAB, 0x01
249
442k
    };
250
    // Coefficients of returned SignedMatrix; starts off as identity matrix. */
251
442k
    limb_t u = 1, v = 0, q = 0, r = 1;
252
    // The number of divsteps still left.
253
442k
    int i = SIGNED_LIMB_SIZE;
254
2.80M
    while (true) {
255
        /* Use a sentinel bit to count zeros only up to i. */
256
2.80M
        int zeros = std::countr_zero(g | (MAX_LIMB << i));
257
        /* Perform zeros divsteps at once; they all just divide g by two. */
258
2.80M
        g >>= zeros;
259
2.80M
        u <<= zeros;
260
2.80M
        v <<= zeros;
261
2.80M
        eta -= zeros;
262
2.80M
        i -= zeros;
263
         /* We're done once we've performed SIGNED_LIMB_SIZE divsteps. */
264
2.80M
        if (i == 0) break;
265
        /* If eta is negative, negate it and replace f,g with g,-f. */
266
2.35M
        if (eta < 0) {
267
779k
            limb_t tmp;
268
779k
            eta = -eta;
269
779k
            tmp = f; f = g; g = -tmp;
270
779k
            tmp = u; u = q; q = -tmp;
271
779k
            tmp = v; v = r; r = -tmp;
272
779k
        }
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
2.35M
        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
2.35M
        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
2.35M
        limb_t w = (g * NEGINV256[(f >> 1) & 127]) & m;
281
        /* Do so. */
282
2.35M
        g += f * w;
283
2.35M
        q += u * w;
284
2.35M
        r += v * w;
285
2.35M
    }
286
442k
    out.u = (signed_limb_t)u;
287
442k
    out.v = (signed_limb_t)v;
288
442k
    out.q = (signed_limb_t)q;
289
442k
    out.r = (signed_limb_t)r;
290
442k
    return eta;
291
442k
}
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
442k
{
299
442k
    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
442k
    signed_limb_t sd = d.limbs[SIGNED_LIMBS - 1] >> (LIMB_SIZE - 1);
303
442k
    signed_limb_t se = e.limbs[SIGNED_LIMBS - 1] >> (LIMB_SIZE - 1);
304
442k
    signed_limb_t md = (u & sd) + (v & se);
305
442k
    signed_limb_t me = (q & sd) + (r & se);
306
    /* Begin computing t*[d,e]. */
307
442k
    signed_limb_t di = d.limbs[0], ei = e.limbs[0];
308
442k
    signed_double_limb_t cd = (signed_double_limb_t)u * di + (signed_double_limb_t)v * ei;
309
442k
    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
442k
    md -= (MODULUS_INVERSE * limb_t(cd) + md) & MAX_SIGNED_LIMB;
312
442k
    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
442k
    cd -= (signed_double_limb_t)1103717 * md;
315
442k
    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
442k
    Assume((cd & MAX_SIGNED_LIMB) == 0);
318
442k
    Assume((ce & MAX_SIGNED_LIMB) == 0);
319
442k
    cd >>= SIGNED_LIMB_SIZE;
320
442k
    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
21.7M
    for (int i = 1; i < SIGNED_LIMBS - 1; ++i) {
325
21.2M
        di = d.limbs[i];
326
21.2M
        ei = e.limbs[i];
327
21.2M
        cd += (signed_double_limb_t)u * di + (signed_double_limb_t)v * ei;
328
21.2M
        ce += (signed_double_limb_t)q * di + (signed_double_limb_t)r * ei;
329
21.2M
        d.limbs[i - 1] = (signed_limb_t)cd & MAX_SIGNED_LIMB; cd >>= SIGNED_LIMB_SIZE;
330
21.2M
        e.limbs[i - 1] = (signed_limb_t)ce & MAX_SIGNED_LIMB; ce >>= SIGNED_LIMB_SIZE;
331
21.2M
    }
332
    /* Compute limb SIGNED_LIMBS-1 of t*[d,e]+modulus*[md,me], and store it in output limb SIGNED_LIMBS-2. */
333
442k
    di = d.limbs[SIGNED_LIMBS - 1];
334
442k
    ei = e.limbs[SIGNED_LIMBS - 1];
335
442k
    cd += (signed_double_limb_t)u * di + (signed_double_limb_t)v * ei;
336
442k
    ce += (signed_double_limb_t)q * di + (signed_double_limb_t)r * ei;
337
442k
    cd += (signed_double_limb_t)md << FINAL_LIMB_MODULUS_BITS;
338
442k
    ce += (signed_double_limb_t)me << FINAL_LIMB_MODULUS_BITS;
339
442k
    d.limbs[SIGNED_LIMBS - 2] = (signed_limb_t)cd & MAX_SIGNED_LIMB; cd >>= SIGNED_LIMB_SIZE;
340
442k
    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
442k
    d.limbs[SIGNED_LIMBS - 1] = (signed_limb_t)cd;
343
442k
    e.limbs[SIGNED_LIMBS - 1] = (signed_limb_t)ce;
344
442k
}
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
442k
{
353
442k
    const signed_limb_t u = t.u, v=t.v, q=t.q, r=t.r;
354
355
442k
    signed_limb_t fi, gi;
356
442k
    signed_double_limb_t cf, cg;
357
    /* Start computing t*[f,g]. */
358
442k
    fi = f.limbs[0];
359
442k
    gi = g.limbs[0];
360
442k
    cf = (signed_double_limb_t)u * fi + (signed_double_limb_t)v * gi;
361
442k
    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
442k
    Assume((cf & MAX_SIGNED_LIMB) == 0);
364
442k
    Assume((cg & MAX_SIGNED_LIMB) == 0);
365
442k
    cf >>= SIGNED_LIMB_SIZE;
366
442k
    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
15.9M
    for (int i = 1; i < len; ++i) {
370
15.5M
        fi = f.limbs[i];
371
15.5M
        gi = g.limbs[i];
372
15.5M
        cf += (signed_double_limb_t)u * fi + (signed_double_limb_t)v * gi;
373
15.5M
        cg += (signed_double_limb_t)q * fi + (signed_double_limb_t)r * gi;
374
15.5M
        f.limbs[i - 1] = (signed_limb_t)cf & MAX_SIGNED_LIMB; cf >>= SIGNED_LIMB_SIZE;
375
15.5M
        g.limbs[i - 1] = (signed_limb_t)cg & MAX_SIGNED_LIMB; cg >>= SIGNED_LIMB_SIZE;
376
15.5M
    }
377
    /* What remains is limb SIGNED_LIMBS of t*[f,g]; store it as output limb SIGNED_LIMBS-1. */
378
442k
    f.limbs[len - 1] = (signed_limb_t)cf;
379
442k
    g.limbs[len - 1] = (signed_limb_t)cg;
380
381
442k
}
382
} // namespace
383
384
Num3072 Num3072::GetInverse() const
385
4.41k
{
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
4.41k
    Num3072Signed d, e, f, g;
401
4.41k
    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
4.41k
    f.limbs[0] = -MAX_PRIME_DIFF;
405
4.41k
    f.limbs[FINAL_LIMB_POSITION] = ((limb_t)1) << FINAL_LIMB_MODULUS_BITS;
406
4.41k
    g.FromNum3072(*this);
407
4.41k
    int len = SIGNED_LIMBS; //!< The number of significant limbs in f and g
408
4.41k
    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
442k
    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
442k
        SignedMatrix t;
414
442k
        eta = ComputeDivstepMatrix(eta, f.limbs[0], g.limbs[0], t);
415
        // Apply that transformation matrix to the full [f,g] vector.
416
442k
        UpdateFG(f, g, t, len);
417
        // Apply that transformation matrix to the full [d,e] vector (mod modulus).
418
442k
        UpdateDE(d, e, t);
419
420
        // Check if g is zero.
421
442k
        if (g.limbs[0] == 0) {
422
196k
            signed_limb_t cond = 0;
423
9.58M
            for (int j = 1; j < len; ++j) {
424
9.39M
                cond |= g.limbs[j];
425
9.39M
            }
426
            // If so, we're done.
427
196k
            if (cond == 0) break;
428
196k
        }
429
430
        // Check if the top limbs of both f and g are both 0 or -1.
431
438k
        signed_limb_t fn = f.limbs[len - 1], gn = g.limbs[len - 1];
432
438k
        signed_limb_t cond = ((signed_limb_t)len - 2) >> (LIMB_SIZE - 1);
433
438k
        cond |= fn ^ (fn >> (LIMB_SIZE - 1));
434
438k
        cond |= gn ^ (gn >> (LIMB_SIZE - 1));
435
438k
        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
216k
            f.limbs[len - 2] |= (limb_t)f.limbs[len - 1] << SIGNED_LIMB_SIZE;
439
216k
            g.limbs[len - 2] |= (limb_t)g.limbs[len - 1] << SIGNED_LIMB_SIZE;
440
216k
            --len;
441
216k
        }
442
438k
    }
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
4.41k
    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
4.41k
    d.Normalize(f.limbs[len - 1] >> (LIMB_SIZE  - 1));
451
4.41k
    Num3072 ret;
452
4.41k
    d.ToNum3072(ret);
453
4.41k
    return ret;
454
4.41k
}
455
456
void Num3072::Multiply(const Num3072& a)
457
11.0k
{
458
11.0k
    limb_t c0 = 0, c1 = 0, c2 = 0;
459
11.0k
    Num3072 tmp;
460
461
    /* Compute limbs 0..N-2 of this*a into tmp, including one reduction. */
462
528k
    for (int j = 0; j < LIMBS - 1; ++j) {
463
517k
        limb_t d0 = 0, d1 = 0, d2 = 0;
464
517k
        mul(d0, d1, this->limbs[1 + j], a.limbs[LIMBS + j - (1 + j)]);
465
12.4M
        for (int i = 2 + j; i < LIMBS; ++i) muladd3(d0, d1, d2, this->limbs[i], a.limbs[LIMBS + j - i]);
466
517k
        mulnadd3(c0, c1, c2, d0, d1, d2, MAX_PRIME_DIFF);
467
12.9M
        for (int i = 0; i < j + 1; ++i) muladd3(c0, c1, c2, this->limbs[i], a.limbs[j - i]);
468
517k
        extract3(c0, c1, c2, tmp.limbs[j]);
469
517k
    }
470
471
    /* Compute limb N-1 of a*b into tmp. */
472
11.0k
    assert(c2 == 0);
473
539k
    for (int i = 0; i < LIMBS; ++i) muladd3(c0, c1, c2, this->limbs[i], a.limbs[LIMBS - 1 - i]);
474
11.0k
    extract3(c0, c1, c2, tmp.limbs[LIMBS - 1]);
475
476
    /* Perform a second reduction. */
477
11.0k
    muln2(c0, c1, MAX_PRIME_DIFF);
478
539k
    for (int j = 0; j < LIMBS; ++j) {
479
528k
        addnextract2(c0, c1, tmp.limbs[j], this->limbs[j]);
480
528k
    }
481
482
11.0k
    assert(c1 == 0);
483
11.0k
    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
11.0k
    if (this->IsOverflow()) this->FullReduce();
490
11.0k
    if (c0) this->FullReduce();
491
11.0k
}
492
493
void Num3072::SetToOne()
494
24.8k
{
495
24.8k
    this->limbs[0] = 1;
496
1.19M
    for (int i = 1; i < LIMBS; ++i) this->limbs[i] = 0;
497
24.8k
}
498
499
void Num3072::Divide(const Num3072& a)
500
4.41k
{
501
4.41k
    if (this->IsOverflow()) this->FullReduce();
502
503
4.41k
    Num3072 inv{};
504
4.41k
    if (a.IsOverflow()) {
505
0
        Num3072 b = a;
506
0
        b.FullReduce();
507
0
        inv = b.GetInverse();
508
4.41k
    } else {
509
4.41k
        inv = a.GetInverse();
510
4.41k
    }
511
512
4.41k
    this->Multiply(inv);
513
4.41k
    if (this->IsOverflow()) this->FullReduce();
514
4.41k
}
515
516
6.37k
Num3072::Num3072(const unsigned char (&data)[BYTE_SIZE]) {
517
312k
    for (int i = 0; i < LIMBS; ++i) {
518
305k
        if (sizeof(limb_t) == 4) {
519
0
            this->limbs[i] = ReadLE32(data + 4 * i);
520
305k
        } else if (sizeof(limb_t) == 8) {
521
305k
            this->limbs[i] = ReadLE64(data + 8 * i);
522
305k
        }
523
305k
    }
524
6.37k
}
525
526
4.41k
void Num3072::ToBytes(unsigned char (&out)[BYTE_SIZE]) {
527
216k
    for (int i = 0; i < LIMBS; ++i) {
528
211k
        if (sizeof(limb_t) == 4) {
529
0
            WriteLE32(out + i * 4, this->limbs[i]);
530
211k
        } else if (sizeof(limb_t) == 8) {
531
211k
            WriteLE64(out + i * 8, this->limbs[i]);
532
211k
        }
533
211k
    }
534
4.41k
}
535
536
6.37k
Num3072 MuHash3072::ToNum3072(std::span<const unsigned char> in) {
537
6.37k
    unsigned char tmp[Num3072::BYTE_SIZE];
538
539
6.37k
    uint256 hashed_in{(HashWriter{} << in).GetSHA256()};
540
6.37k
    static_assert(sizeof(tmp) % ChaCha20Aligned::BLOCKLEN == 0);
541
6.37k
    ChaCha20Aligned{MakeByteSpan(hashed_in)}.Keystream(MakeWritableByteSpan(tmp));
542
6.37k
    Num3072 out{tmp};
543
544
6.37k
    return out;
545
6.37k
}
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
4.41k
{
554
4.41k
    m_numerator.Divide(m_denominator);
555
4.41k
    m_denominator.SetToOne();  // Needed to keep the MuHash object valid
556
557
4.41k
    unsigned char data[Num3072::BYTE_SIZE];
558
4.41k
    m_numerator.ToBytes(data);
559
560
4.41k
    out = (HashWriter{} << data).GetSHA256();
561
4.41k
}
562
563
MuHash3072& MuHash3072::operator*=(const MuHash3072& mul) noexcept
564
108
{
565
108
    m_numerator.Multiply(mul.m_numerator);
566
108
    m_denominator.Multiply(mul.m_denominator);
567
108
    return *this;
568
108
}
569
570
MuHash3072& MuHash3072::operator/=(const MuHash3072& div) noexcept
571
95
{
572
95
    m_numerator.Multiply(div.m_denominator);
573
95
    m_denominator.Multiply(div.m_numerator);
574
95
    return *this;
575
95
}
576
577
5.70k
MuHash3072& MuHash3072::Insert(std::span<const unsigned char> in) noexcept {
578
5.70k
    m_numerator.Multiply(ToNum3072(in));
579
5.70k
    return *this;
580
5.70k
}
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
}