Coverage Report

Created: 2025-11-16 06:46

next uncovered line (L), next uncovered region (R), next uncovered branch (B)
/src/gmp/mpn/hgcd2.c
Line
Count
Source
1
/* hgcd2.c
2
3
   THE FUNCTIONS IN THIS FILE ARE INTERNAL WITH MUTABLE INTERFACES.  IT IS ONLY
4
   SAFE TO REACH THEM THROUGH DOCUMENTED INTERFACES.  IN FACT, IT IS ALMOST
5
   GUARANTEED THAT THEY'LL CHANGE OR DISAPPEAR IN A FUTURE GNU MP RELEASE.
6
7
Copyright 1996, 1998, 2000-2004, 2008, 2012, 2019 Free Software Foundation,
8
Inc.
9
10
This file is part of the GNU MP Library.
11
12
The GNU MP Library is free software; you can redistribute it and/or modify
13
it under the terms of either:
14
15
  * the GNU Lesser General Public License as published by the Free
16
    Software Foundation; either version 3 of the License, or (at your
17
    option) any later version.
18
19
or
20
21
  * the GNU General Public License as published by the Free Software
22
    Foundation; either version 2 of the License, or (at your option) any
23
    later version.
24
25
or both in parallel, as here.
26
27
The GNU MP Library is distributed in the hope that it will be useful, but
28
WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
29
or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
30
for more details.
31
32
You should have received copies of the GNU General Public License and the
33
GNU Lesser General Public License along with the GNU MP Library.  If not,
34
see https://www.gnu.org/licenses/.  */
35
36
#include "gmp-impl.h"
37
#include "longlong.h"
38
39
#include "mpn/generic/hgcd2-div.h"
40
41
#if GMP_NAIL_BITS != 0
42
#error Nails not implemented
43
#endif
44
45
/* Reduces a,b until |a-b| (almost) fits in one limb + 1 bit. Constructs
46
   matrix M. Returns 1 if we make progress, i.e. can perform at least
47
   one subtraction. Otherwise returns zero. */
48
49
/* FIXME: Possible optimizations:
50
51
   The div2 function starts with checking the most significant bit of
52
   the numerator. We can maintained normalized operands here, call
53
   hgcd with normalized operands only, which should make the code
54
   simpler and possibly faster.
55
56
   Experiment with table lookups on the most significant bits.
57
58
   This function is also a candidate for assembler implementation.
59
*/
60
int
61
mpn_hgcd2 (mp_limb_t ah, mp_limb_t al, mp_limb_t bh, mp_limb_t bl,
62
     struct hgcd_matrix1 *M)
63
620k
{
64
620k
  mp_limb_t u00, u01, u10, u11;
65
66
620k
  if (ah < 2 || bh < 2)
67
1.49k
    return 0;
68
69
618k
  if (ah > bh || (ah == bh && al > bl))
70
353k
    {
71
353k
      sub_ddmmss (ah, al, ah, al, bh, bl);
72
353k
      if (ah < 2)
73
3.00k
  return 0;
74
75
350k
      u00 = u01 = u11 = 1;
76
350k
      u10 = 0;
77
350k
    }
78
265k
  else
79
265k
    {
80
265k
      sub_ddmmss (bh, bl, bh, bl, ah, al);
81
265k
      if (bh < 2)
82
3.59k
  return 0;
83
84
261k
      u00 = u10 = u11 = 1;
85
261k
      u01 = 0;
86
261k
    }
87
88
612k
  if (ah < bh)
89
354k
    goto subtract_a;
90
91
258k
  for (;;)
92
5.81M
    {
93
5.81M
      ASSERT (ah >= bh);
94
5.81M
      if (ah == bh)
95
740
  goto done;
96
97
5.81M
      if (ah < (CNST_LIMB(1) << (GMP_LIMB_BITS / 2)))
98
330k
  {
99
330k
    ah = (ah << (GMP_LIMB_BITS / 2) ) + (al >> (GMP_LIMB_BITS / 2));
100
330k
    bh = (bh << (GMP_LIMB_BITS / 2) ) + (bl >> (GMP_LIMB_BITS / 2));
101
102
330k
    break;
103
330k
  }
104
105
      /* Subtract a -= q b, and multiply M from the right by (1 q ; 0
106
   1), affecting the second column of M. */
107
5.48M
      ASSERT (ah > bh);
108
5.48M
      sub_ddmmss (ah, al, ah, al, bh, bl);
109
110
5.48M
      if (ah < 2)
111
847
  goto done;
112
113
5.48M
      if (ah <= bh)
114
2.18M
  {
115
    /* Use q = 1 */
116
2.18M
    u01 += u00;
117
2.18M
    u11 += u10;
118
2.18M
  }
119
3.29M
      else
120
3.29M
  {
121
3.29M
    mp_limb_t r[2];
122
3.29M
    mp_limb_t q = div2 (r, ah, al, bh, bl);
123
3.29M
    al = r[0]; ah = r[1];
124
3.29M
    if (ah < 2)
125
1.59k
      {
126
        /* A is too small, but q is correct. */
127
1.59k
        u01 += q * u00;
128
1.59k
        u11 += q * u10;
129
1.59k
        goto done;
130
1.59k
      }
131
3.29M
    q++;
132
3.29M
    u01 += q * u00;
133
3.29M
    u11 += q * u10;
134
3.29M
  }
135
5.83M
    subtract_a:
136
5.83M
      ASSERT (bh >= ah);
137
5.83M
      if (ah == bh)
138
655
  goto done;
139
140
5.83M
      if (bh < (CNST_LIMB(1) << (GMP_LIMB_BITS / 2)))
141
275k
  {
142
275k
    ah = (ah << (GMP_LIMB_BITS / 2) ) + (al >> (GMP_LIMB_BITS / 2));
143
275k
    bh = (bh << (GMP_LIMB_BITS / 2) ) + (bl >> (GMP_LIMB_BITS / 2));
144
145
275k
    goto subtract_a1;
146
275k
  }
147
148
      /* Subtract b -= q a, and multiply M from the right by (1 0 ; q
149
   1), affecting the first column of M. */
150
5.56M
      sub_ddmmss (bh, bl, bh, bl, ah, al);
151
152
5.56M
      if (bh < 2)
153
641
  goto done;
154
155
5.55M
      if (bh <= ah)
156
2.18M
  {
157
    /* Use q = 1 */
158
2.18M
    u00 += u01;
159
2.18M
    u10 += u11;
160
2.18M
  }
161
3.37M
      else
162
3.37M
  {
163
3.37M
    mp_limb_t r[2];
164
3.37M
    mp_limb_t q = div2 (r, bh, bl, ah, al);
165
3.37M
    bl = r[0]; bh = r[1];
166
3.37M
    if (bh < 2)
167
1.56k
      {
168
        /* B is too small, but q is correct. */
169
1.56k
        u00 += q * u01;
170
1.56k
        u10 += q * u11;
171
1.56k
        goto done;
172
1.56k
      }
173
3.37M
    q++;
174
3.37M
    u00 += q * u01;
175
3.37M
    u10 += q * u11;
176
3.37M
  }
177
5.55M
    }
178
179
  /* NOTE: Since we discard the least significant half limb, we don't get a
180
     truly maximal M (corresponding to |a - b| < 2^{GMP_LIMB_BITS +1}). */
181
  /* Single precision loop */
182
330k
  for (;;)
183
5.31M
    {
184
5.31M
      ASSERT (ah >= bh);
185
186
5.31M
      ah -= bh;
187
5.31M
      if (ah < (CNST_LIMB (1) << (GMP_LIMB_BITS / 2 + 1)))
188
187k
  break;
189
190
5.12M
      if (ah <= bh)
191
2.31M
  {
192
    /* Use q = 1 */
193
2.31M
    u01 += u00;
194
2.31M
    u11 += u10;
195
2.31M
  }
196
2.80M
      else
197
2.80M
  {
198
2.80M
    mp_double_limb_t rq = div1 (ah, bh);
199
2.80M
    mp_limb_t q = rq.d1;
200
2.80M
    ah = rq.d0;
201
202
2.80M
    if (ah < (CNST_LIMB(1) << (GMP_LIMB_BITS / 2 + 1)))
203
152k
      {
204
        /* A is too small, but q is correct. */
205
152k
        u01 += q * u00;
206
152k
        u11 += q * u10;
207
152k
        break;
208
152k
      }
209
2.65M
    q++;
210
2.65M
    u01 += q * u00;
211
2.65M
    u11 += q * u10;
212
2.65M
  }
213
5.24M
    subtract_a1:
214
5.24M
      ASSERT (bh >= ah);
215
216
5.24M
      bh -= ah;
217
5.24M
      if (bh < (CNST_LIMB (1) << (GMP_LIMB_BITS / 2 + 1)))
218
113k
  break;
219
220
5.13M
      if (bh <= ah)
221
2.00M
  {
222
    /* Use q = 1 */
223
2.00M
    u00 += u01;
224
2.00M
    u10 += u11;
225
2.00M
  }
226
3.12M
      else
227
3.12M
  {
228
3.12M
    mp_double_limb_t rq = div1 (bh, ah);
229
3.12M
    mp_limb_t q = rq.d1;
230
3.12M
    bh = rq.d0;
231
232
3.12M
    if (bh < (CNST_LIMB(1) << (GMP_LIMB_BITS / 2 + 1)))
233
152k
      {
234
        /* B is too small, but q is correct. */
235
152k
        u00 += q * u01;
236
152k
        u10 += q * u11;
237
152k
        break;
238
152k
      }
239
2.97M
    q++;
240
2.97M
    u00 += q * u01;
241
2.97M
    u10 += q * u11;
242
2.97M
  }
243
5.13M
    }
244
245
612k
 done:
246
612k
  M->u[0][0] = u00; M->u[0][1] = u01;
247
612k
  M->u[1][0] = u10; M->u[1][1] = u11;
248
249
612k
  return 1;
250
330k
}
251
252
/* Sets (r;b) = (a;b) M, with M = (u00, u01; u10, u11). Vector must
253
 * have space for n + 1 limbs. Uses three buffers to avoid a copy*/
254
mp_size_t
255
mpn_hgcd_mul_matrix1_vector (const struct hgcd_matrix1 *M,
256
           mp_ptr rp, mp_srcptr ap, mp_ptr bp, mp_size_t n)
257
612k
{
258
612k
  mp_limb_t ah, bh;
259
260
  /* Compute (r,b) <-- (u00 a + u10 b, u01 a + u11 b) as
261
262
     r  = u00 * a
263
     r += u10 * b
264
     b *= u11
265
     b += u01 * a
266
  */
267
268
#if HAVE_NATIVE_mpn_addaddmul_1msb0
269
  ah = mpn_addaddmul_1msb0 (rp, ap, bp, n, M->u[0][0], M->u[1][0]);
270
  bh = mpn_addaddmul_1msb0 (bp, bp, ap, n, M->u[1][1], M->u[0][1]);
271
#else
272
612k
  ah =     mpn_mul_1 (rp, ap, n, M->u[0][0]);
273
612k
  ah += mpn_addmul_1 (rp, bp, n, M->u[1][0]);
274
275
612k
  bh =     mpn_mul_1 (bp, bp, n, M->u[1][1]);
276
612k
  bh += mpn_addmul_1 (bp, ap, n, M->u[0][1]);
277
612k
#endif
278
612k
  rp[n] = ah;
279
612k
  bp[n] = bh;
280
281
612k
  n += (ah | bh) > 0;
282
612k
  return n;
283
612k
}