Coverage Report

Created: 2026-06-08 06:48

next uncovered line (L), next uncovered region (R), next uncovered branch (B)
/src/gmp/mpn/tdiv_qr.c
Line
Count
Source
1
/* mpn_tdiv_qr -- Divide the numerator (np,nn) by the denominator (dp,dn) and
2
   write the nn-dn+1 quotient limbs at qp and the dn remainder limbs at rp.  If
3
   qxn is non-zero, generate that many fraction limbs and append them after the
4
   other quotient limbs, and update the remainder accordingly.  The input
5
   operands are unaffected.
6
7
   Preconditions:
8
   1. The most significant limb of the divisor must be non-zero.
9
   2. nn >= dn, even if qxn is non-zero.  (??? relax this ???)
10
11
   The time complexity of this is O(qn*qn+M(dn,qn)), where M(m,n) is the time
12
   complexity of multiplication.
13
14
Copyright 1997, 2000-2002, 2005, 2009, 2015 Free Software Foundation, Inc.
15
16
This file is part of the GNU MP Library.
17
18
The GNU MP Library is free software; you can redistribute it and/or modify
19
it under the terms of either:
20
21
  * the GNU Lesser General Public License as published by the Free
22
    Software Foundation; either version 3 of the License, or (at your
23
    option) any later version.
24
25
or
26
27
  * the GNU General Public License as published by the Free Software
28
    Foundation; either version 2 of the License, or (at your option) any
29
    later version.
30
31
or both in parallel, as here.
32
33
The GNU MP Library is distributed in the hope that it will be useful, but
34
WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
35
or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
36
for more details.
37
38
You should have received copies of the GNU General Public License and the
39
GNU Lesser General Public License along with the GNU MP Library.  If not,
40
see https://www.gnu.org/licenses/.  */
41
42
#include "gmp-impl.h"
43
#include "longlong.h"
44
45
46
void
47
mpn_tdiv_qr (mp_ptr qp, mp_ptr rp, mp_size_t qxn,
48
       mp_srcptr np, mp_size_t nn, mp_srcptr dp, mp_size_t dn)
49
88.8k
{
50
88.8k
  ASSERT_ALWAYS (qxn == 0);
51
52
88.8k
  ASSERT (nn >= 0);
53
88.8k
  ASSERT (dn >= 0);
54
88.8k
  ASSERT (dn == 0 || dp[dn - 1] != 0);
55
88.8k
  ASSERT (! MPN_OVERLAP_P (qp, nn - dn + 1 + qxn, np, nn));
56
88.8k
  ASSERT (! MPN_OVERLAP_P (qp, nn - dn + 1 + qxn, dp, dn));
57
58
88.8k
  switch (dn)
59
88.8k
    {
60
0
    case 0:
61
0
      DIVIDE_BY_ZERO;
62
63
8.79k
    case 1:
64
8.79k
      {
65
8.79k
  rp[0] = mpn_divrem_1 (qp, (mp_size_t) 0, np, nn, dp[0]);
66
8.79k
  return;
67
0
      }
68
69
3.93k
    case 2:
70
3.93k
      {
71
3.93k
  mp_ptr n2p;
72
3.93k
  mp_limb_t qhl, cy;
73
3.93k
  TMP_DECL;
74
3.93k
  TMP_MARK;
75
3.93k
  if ((dp[1] & GMP_NUMB_HIGHBIT) == 0)
76
3.14k
    {
77
3.14k
      int cnt;
78
3.14k
      mp_limb_t d2p[2];
79
3.14k
      count_leading_zeros (cnt, dp[1]);
80
3.14k
      cnt -= GMP_NAIL_BITS;
81
3.14k
      d2p[1] = (dp[1] << cnt) | (dp[0] >> (GMP_NUMB_BITS - cnt));
82
3.14k
      d2p[0] = (dp[0] << cnt) & GMP_NUMB_MASK;
83
3.14k
      n2p = TMP_ALLOC_LIMBS (nn + 1);
84
3.14k
      cy = mpn_lshift (n2p, np, nn, cnt);
85
3.14k
      n2p[nn] = cy;
86
3.14k
      qhl = mpn_divrem_2 (qp, 0L, n2p, nn + (cy != 0), d2p);
87
3.14k
      if (cy == 0)
88
1.14k
        qp[nn - 2] = qhl; /* always store nn-2+1 quotient limbs */
89
3.14k
      rp[0] = (n2p[0] >> cnt)
90
3.14k
        | ((n2p[1] << (GMP_NUMB_BITS - cnt)) & GMP_NUMB_MASK);
91
3.14k
      rp[1] = (n2p[1] >> cnt);
92
3.14k
    }
93
789
  else
94
789
    {
95
789
      n2p = TMP_ALLOC_LIMBS (nn);
96
789
      MPN_COPY (n2p, np, nn);
97
789
      qhl = mpn_divrem_2 (qp, 0L, n2p, nn, dp);
98
789
      qp[nn - 2] = qhl; /* always store nn-2+1 quotient limbs */
99
789
      rp[0] = n2p[0];
100
789
      rp[1] = n2p[1];
101
789
    }
102
3.93k
  TMP_FREE;
103
3.93k
  return;
104
0
      }
105
106
76.1k
    default:
107
76.1k
      {
108
76.1k
  int adjust;
109
76.1k
  gmp_pi1_t dinv;
110
76.1k
  TMP_DECL;
111
76.1k
  TMP_MARK;
112
76.1k
  adjust = np[nn - 1] >= dp[dn - 1];  /* conservative tests for quotient size */
113
76.1k
  if (nn + adjust >= 2 * dn)
114
44.0k
    {
115
44.0k
      mp_ptr n2p, d2p;
116
44.0k
      mp_limb_t cy;
117
44.0k
      int cnt;
118
119
44.0k
      qp[nn - dn] = 0;        /* zero high quotient limb */
120
44.0k
      if ((dp[dn - 1] & GMP_NUMB_HIGHBIT) == 0) /* normalize divisor */
121
5.13k
        {
122
5.13k
    count_leading_zeros (cnt, dp[dn - 1]);
123
5.13k
    cnt -= GMP_NAIL_BITS;
124
5.13k
    d2p = TMP_ALLOC_LIMBS (dn);
125
5.13k
    mpn_lshift (d2p, dp, dn, cnt);
126
5.13k
    n2p = TMP_ALLOC_LIMBS (nn + 1);
127
5.13k
    cy = mpn_lshift (n2p, np, nn, cnt);
128
5.13k
    n2p[nn] = cy;
129
5.13k
    nn += adjust;
130
5.13k
        }
131
38.9k
      else
132
38.9k
        {
133
38.9k
    cnt = 0;
134
38.9k
    d2p = (mp_ptr) dp;
135
38.9k
    n2p = TMP_ALLOC_LIMBS (nn + 1);
136
38.9k
    MPN_COPY (n2p, np, nn);
137
38.9k
    n2p[nn] = 0;
138
38.9k
    nn += adjust;
139
38.9k
        }
140
141
44.0k
      invert_pi1 (dinv, d2p[dn - 1], d2p[dn - 2]);
142
44.0k
      if (BELOW_THRESHOLD (dn, DC_DIV_QR_THRESHOLD))
143
43.8k
        mpn_sbpi1_div_qr (qp, n2p, nn, d2p, dn, dinv.inv32);
144
256
      else if (BELOW_THRESHOLD (dn, MUPI_DIV_QR_THRESHOLD) ||   /* fast condition */
145
0
         BELOW_THRESHOLD (nn, 2 * MU_DIV_QR_THRESHOLD) || /* fast condition */
146
0
         (double) (2 * (MU_DIV_QR_THRESHOLD - MUPI_DIV_QR_THRESHOLD)) * dn /* slow... */
147
0
         + (double) MUPI_DIV_QR_THRESHOLD * nn > (double) dn * nn)    /* ...condition */
148
256
        mpn_dcpi1_div_qr (qp, n2p, nn, d2p, dn, &dinv);
149
0
      else
150
0
        {
151
0
    mp_size_t itch = mpn_mu_div_qr_itch (nn, dn, 0);
152
0
    mp_ptr scratch = TMP_ALLOC_LIMBS (itch);
153
0
    mpn_mu_div_qr (qp, rp, n2p, nn, d2p, dn, scratch);
154
0
    n2p = rp;
155
0
        }
156
157
44.0k
      if (cnt != 0)
158
5.13k
        mpn_rshift (rp, n2p, dn, cnt);
159
38.9k
      else
160
38.9k
        MPN_COPY (rp, n2p, dn);
161
44.0k
      TMP_FREE;
162
44.0k
      return;
163
44.0k
    }
164
165
  /* When we come here, the numerator/partial remainder is less
166
     than twice the size of the denominator.  */
167
168
32.0k
    {
169
      /* Problem:
170
171
         Divide a numerator N with nn limbs by a denominator D with dn
172
         limbs forming a quotient of qn=nn-dn+1 limbs.  When qn is small
173
         compared to dn, conventional division algorithms perform poorly.
174
         We want an algorithm that has an expected running time that is
175
         dependent only on qn.
176
177
         Algorithm (very informally stated):
178
179
         1) Divide the 2 x qn most significant limbs from the numerator
180
      by the qn most significant limbs from the denominator.  Call
181
      the result qest.  This is either the correct quotient, but
182
      might be 1 or 2 too large.  Compute the remainder from the
183
      division.  (This step is implemented by an mpn_divrem call.)
184
185
         2) Is the most significant limb from the remainder < p, where p
186
      is the product of the most significant limb from the quotient
187
      and the next(d)?  (Next(d) denotes the next ignored limb from
188
      the denominator.)  If it is, decrement qest, and adjust the
189
      remainder accordingly.
190
191
         3) Is the remainder >= qest?  If it is, qest is the desired
192
      quotient.  The algorithm terminates.
193
194
         4) Subtract qest x next(d) from the remainder.  If there is
195
      borrow out, decrement qest, and adjust the remainder
196
      accordingly.
197
198
         5) Skip one word from the denominator (i.e., let next(d) denote
199
      the next less significant limb.  */
200
201
32.0k
      mp_size_t qn;
202
32.0k
      mp_ptr n2p, d2p;
203
32.0k
      mp_ptr tp;
204
32.0k
      mp_limb_t cy;
205
32.0k
      mp_size_t in, rn;
206
32.0k
      mp_limb_t quotient_too_large;
207
32.0k
      unsigned int cnt;
208
209
32.0k
      qn = nn - dn;
210
32.0k
      qp[qn] = 0;       /* zero high quotient limb */
211
32.0k
      qn += adjust;     /* qn cannot become bigger */
212
213
32.0k
      if (qn == 0)
214
259
        {
215
259
    MPN_COPY (rp, np, dn);
216
259
    TMP_FREE;
217
259
    return;
218
259
        }
219
220
31.7k
      in = dn - qn;   /* (at least partially) ignored # of limbs in ops */
221
      /* Normalize denominator by shifting it to the left such that its
222
         most significant bit is set.  Then shift the numerator the same
223
         amount, to mathematically preserve quotient.  */
224
31.7k
      if ((dp[dn - 1] & GMP_NUMB_HIGHBIT) == 0)
225
23.7k
        {
226
23.7k
    count_leading_zeros (cnt, dp[dn - 1]);
227
23.7k
    cnt -= GMP_NAIL_BITS;
228
229
23.7k
    d2p = TMP_ALLOC_LIMBS (qn);
230
23.7k
    mpn_lshift (d2p, dp + in, qn, cnt);
231
23.7k
    d2p[0] |= dp[in - 1] >> (GMP_NUMB_BITS - cnt);
232
233
23.7k
    n2p = TMP_ALLOC_LIMBS (2 * qn + 1);
234
23.7k
    cy = mpn_lshift (n2p, np + nn - 2 * qn, 2 * qn, cnt);
235
23.7k
    if (adjust)
236
13.5k
      {
237
13.5k
        n2p[2 * qn] = cy;
238
13.5k
        n2p++;
239
13.5k
      }
240
10.2k
    else
241
10.2k
      {
242
10.2k
        n2p[0] |= np[nn - 2 * qn - 1] >> (GMP_NUMB_BITS - cnt);
243
10.2k
      }
244
23.7k
        }
245
8.02k
      else
246
8.02k
        {
247
8.02k
    cnt = 0;
248
8.02k
    d2p = (mp_ptr) dp + in;
249
250
8.02k
    n2p = TMP_ALLOC_LIMBS (2 * qn + 1);
251
8.02k
    MPN_COPY (n2p, np + nn - 2 * qn, 2 * qn);
252
8.02k
    if (adjust)
253
858
      {
254
858
        n2p[2 * qn] = 0;
255
858
        n2p++;
256
858
      }
257
8.02k
        }
258
259
      /* Get an approximate quotient using the extracted operands.  */
260
31.7k
      if (qn == 1)
261
7.76k
        {
262
7.76k
    mp_limb_t q0, r0;
263
7.76k
    udiv_qrnnd (q0, r0, n2p[1], n2p[0] << GMP_NAIL_BITS, d2p[0] << GMP_NAIL_BITS);
264
7.76k
    n2p[0] = r0 >> GMP_NAIL_BITS;
265
7.76k
    qp[0] = q0;
266
7.76k
        }
267
24.0k
      else if (qn == 2)
268
8.24k
        mpn_divrem_2 (qp, 0L, n2p, 4L, d2p); /* FIXME: obsolete function */
269
15.7k
      else
270
15.7k
        {
271
15.7k
    invert_pi1 (dinv, d2p[qn - 1], d2p[qn - 2]);
272
15.7k
    if (BELOW_THRESHOLD (qn, DC_DIV_QR_THRESHOLD))
273
14.7k
      mpn_sbpi1_div_qr (qp, n2p, 2 * qn, d2p, qn, dinv.inv32);
274
1.07k
    else if (BELOW_THRESHOLD (qn, MU_DIV_QR_THRESHOLD))
275
1.07k
      mpn_dcpi1_div_qr (qp, n2p, 2 * qn, d2p, qn, &dinv);
276
0
    else
277
0
      {
278
0
        mp_size_t itch = mpn_mu_div_qr_itch (2 * qn, qn, 0);
279
0
        mp_ptr scratch = TMP_ALLOC_LIMBS (itch);
280
0
        mp_ptr r2p = rp;
281
0
        if (np == r2p) /* If N and R share space, put ... */
282
0
          r2p += nn - qn; /* intermediate remainder at N's upper end. */
283
0
        mpn_mu_div_qr (qp, r2p, n2p, 2 * qn, d2p, qn, scratch);
284
0
        MPN_COPY (n2p, r2p, qn);
285
0
      }
286
15.7k
        }
287
288
31.7k
      rn = qn;
289
      /* Multiply the first ignored divisor limb by the most significant
290
         quotient limb.  If that product is > the partial remainder's
291
         most significant limb, we know the quotient is too large.  This
292
         test quickly catches most cases where the quotient is too large;
293
         it catches all cases where the quotient is 2 too large.  */
294
31.7k
      {
295
31.7k
        mp_limb_t dl, x;
296
31.7k
        mp_limb_t h, dummy;
297
298
31.7k
        if (in - 2 < 0)
299
2.01k
    dl = 0;
300
29.7k
        else
301
29.7k
    dl = dp[in - 2];
302
303
31.7k
#if GMP_NAIL_BITS == 0
304
31.7k
        x = (dp[in - 1] << cnt) | ((dl >> 1) >> ((~cnt) % GMP_LIMB_BITS));
305
#else
306
        x = (dp[in - 1] << cnt) & GMP_NUMB_MASK;
307
        if (cnt != 0)
308
    x |= dl >> (GMP_NUMB_BITS - cnt);
309
#endif
310
31.7k
        umul_ppmm (h, dummy, x, qp[qn - 1] << GMP_NAIL_BITS);
311
312
31.7k
        if (n2p[qn - 1] < h)
313
2.25k
    {
314
2.25k
      mp_limb_t cy;
315
316
2.25k
      mpn_decr_u (qp, (mp_limb_t) 1);
317
2.25k
      cy = mpn_add_n (n2p, n2p, d2p, qn);
318
2.25k
      if (cy)
319
760
        {
320
          /* The partial remainder is safely large.  */
321
760
          n2p[qn] = cy;
322
760
          ++rn;
323
760
        }
324
2.25k
    }
325
31.7k
      }
326
327
31.7k
      quotient_too_large = 0;
328
31.7k
      if (cnt != 0)
329
23.7k
        {
330
23.7k
    mp_limb_t cy1, cy2;
331
332
    /* Append partially used numerator limb to partial remainder.  */
333
23.7k
    cy1 = mpn_lshift (n2p, n2p, rn, GMP_NUMB_BITS - cnt);
334
23.7k
    n2p[0] |= np[in - 1] & (GMP_NUMB_MASK >> cnt);
335
336
    /* Update partial remainder with partially used divisor limb.  */
337
23.7k
    cy2 = mpn_submul_1 (n2p, qp, qn, dp[in - 1] & (GMP_NUMB_MASK >> cnt));
338
23.7k
    if (qn != rn)
339
529
      {
340
529
        ASSERT_ALWAYS (n2p[qn] >= cy2);
341
529
        n2p[qn] -= cy2;
342
529
      }
343
23.2k
    else
344
23.2k
      {
345
23.2k
        n2p[qn] = cy1 - cy2; /* & GMP_NUMB_MASK; */
346
347
23.2k
        quotient_too_large = (cy1 < cy2);
348
23.2k
        ++rn;
349
23.2k
      }
350
23.7k
    --in;
351
23.7k
        }
352
      /* True: partial remainder now is neutral, i.e., it is not shifted up.  */
353
354
31.7k
      tp = TMP_ALLOC_LIMBS (dn);
355
356
31.7k
      if (in < qn)
357
9.34k
        {
358
9.34k
    if (in == 0)
359
1.67k
      {
360
1.67k
        MPN_COPY (rp, n2p, rn);
361
1.67k
        ASSERT_ALWAYS (rn == dn);
362
1.67k
        goto foo;
363
1.67k
      }
364
7.66k
    mpn_mul (tp, qp, qn, dp, in);
365
7.66k
        }
366
22.4k
      else
367
22.4k
        mpn_mul (tp, dp, in, qp, qn);
368
369
30.1k
      cy = mpn_sub (n2p, n2p, rn, tp + in, qn);
370
30.1k
      MPN_COPY (rp + in, n2p, dn - in);
371
30.1k
      quotient_too_large |= cy;
372
30.1k
      cy = mpn_sub_n (rp, np, tp, in);
373
30.1k
      cy = mpn_sub_1 (rp + in, rp + in, rn, cy);
374
30.1k
      quotient_too_large |= cy;
375
31.7k
    foo:
376
31.7k
      if (quotient_too_large)
377
1.63k
        {
378
1.63k
    mpn_decr_u (qp, (mp_limb_t) 1);
379
1.63k
    mpn_add_n (rp, rp, dp, dn);
380
1.63k
        }
381
31.7k
    }
382
31.7k
  TMP_FREE;
383
31.7k
  return;
384
30.1k
      }
385
88.8k
    }
386
88.8k
}