Coverage Report

Created: 2023-02-22 06:14

/src/gmp-6.2.1/mpn/toom43_mul.c
Line
Count
Source
1
/* mpn_toom43_mul -- Multiply {ap,an} and {bp,bn} where an is nominally 4/3
2
   times as large as bn.  Or more accurately, bn < an < 2 bn.
3
4
   Contributed to the GNU project by Marco Bodrato.
5
6
   The idea of applying toom to unbalanced multiplication is due to Marco
7
   Bodrato and Alberto Zanoni.
8
9
   THE FUNCTION IN THIS FILE IS INTERNAL WITH A MUTABLE INTERFACE.  IT IS ONLY
10
   SAFE TO REACH IT THROUGH DOCUMENTED INTERFACES.  IN FACT, IT IS ALMOST
11
   GUARANTEED THAT IT WILL CHANGE OR DISAPPEAR IN A FUTURE GNU MP RELEASE.
12
13
Copyright 2009 Free Software Foundation, Inc.
14
15
This file is part of the GNU MP Library.
16
17
The GNU MP Library is free software; you can redistribute it and/or modify
18
it under the terms of either:
19
20
  * the GNU Lesser General Public License as published by the Free
21
    Software Foundation; either version 3 of the License, or (at your
22
    option) any later version.
23
24
or
25
26
  * the GNU General Public License as published by the Free Software
27
    Foundation; either version 2 of the License, or (at your option) any
28
    later version.
29
30
or both in parallel, as here.
31
32
The GNU MP Library is distributed in the hope that it will be useful, but
33
WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
34
or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
35
for more details.
36
37
You should have received copies of the GNU General Public License and the
38
GNU Lesser General Public License along with the GNU MP Library.  If not,
39
see https://www.gnu.org/licenses/.  */
40
41
42
#include "gmp-impl.h"
43
44
/* Evaluate in: -2, -1, 0, +1, +2, +inf
45
46
  <-s-><--n--><--n--><--n-->
47
   ___ ______ ______ ______
48
  |a3_|___a2_|___a1_|___a0_|
49
  |_b2_|___b1_|___b0_|
50
  <-t--><--n--><--n-->
51
52
  v0  =  a0             * b0          #   A(0)*B(0)
53
  v1  = (a0+ a1+ a2+ a3)*(b0+ b1+ b2) #   A(1)*B(1)      ah  <= 3  bh <= 2
54
  vm1 = (a0- a1+ a2- a3)*(b0- b1+ b2) #  A(-1)*B(-1)    |ah| <= 1 |bh|<= 1
55
  v2  = (a0+2a1+4a2+8a3)*(b0+2b1+4b2) #   A(2)*B(2)      ah  <= 14 bh <= 6
56
  vm2 = (a0-2a1+4a2-8a3)*(b0-2b1+4b2) #  A(-2)*B(-2)    |ah| <= 9 |bh|<= 4
57
  vinf=              a3 *         b2  # A(inf)*B(inf)
58
*/
59
60
void
61
mpn_toom43_mul (mp_ptr pp,
62
    mp_srcptr ap, mp_size_t an,
63
    mp_srcptr bp, mp_size_t bn, mp_ptr scratch)
64
830
{
65
830
  mp_size_t n, s, t;
66
830
  enum toom6_flags flags;
67
830
  mp_limb_t cy;
68
69
830
#define a0  ap
70
830
#define a1  (ap + n)
71
830
#define a2  (ap + 2 * n)
72
830
#define a3  (ap + 3 * n)
73
2.49k
#define b0  bp
74
3.32k
#define b1  (bp + n)
75
2.49k
#define b2  (bp + 2 * n)
76
77
830
  n = 1 + (3 * an >= 4 * bn ? (an - 1) >> 2 : (bn - 1) / (size_t) 3);
78
79
830
  s = an - 3 * n;
80
830
  t = bn - 2 * n;
81
82
830
  ASSERT (0 < s && s <= n);
83
830
  ASSERT (0 < t && t <= n);
84
85
  /* This is true whenever an >= 25 or bn >= 19, I think. It
86
     guarantees that we can fit 5 values of size n+1 in the product
87
     area. */
88
830
  ASSERT (s+t >= 5);
89
90
830
#define v0    pp        /* 2n */
91
1.66k
#define vm1   (scratch)        /* 2n+1 */
92
830
#define v1    (pp + 2*n)      /* 2n+1 */
93
1.66k
#define vm2   (scratch + 2 * n + 1)    /* 2n+1 */
94
1.66k
#define v2    (scratch + 4 * n + 2)    /* 2n+1 */
95
830
#define vinf  (pp + 5 * n)      /* s+t */
96
2.49k
#define bs1    pp        /* n+1 */
97
12.0k
#define bsm1  (scratch + 2 * n + 2)    /* n+1 */
98
2.49k
#define asm1  (scratch + 3 * n + 3)    /* n+1 */
99
1.66k
#define asm2  (scratch + 4 * n + 4)    /* n+1 */
100
1.66k
#define bsm2  (pp + n + 1)      /* n+1 */
101
1.66k
#define bs2   (pp + 2 * n + 2)      /* n+1 */
102
1.66k
#define as2   (pp + 3 * n + 3)      /* n+1 */
103
1.66k
#define as1   (pp + 4 * n + 4)      /* n+1 */
104
105
  /* Total sccratch need is 6 * n + 3 + 1; we allocate one extra
106
     limb, because products will overwrite 2n+2 limbs. */
107
108
830
#define a0a2  scratch
109
6.64k
#define b0b2  scratch
110
830
#define a1a3  asm1
111
4.15k
#define b1d   bsm1
112
113
  /* Compute as2 and asm2.  */
114
830
  flags = (enum toom6_flags) (toom6_vm2_neg & mpn_toom_eval_dgr3_pm2 (as2, asm2, ap, n, s, a1a3));
115
116
  /* Compute bs2 and bsm2.  */
117
830
  b1d[n] = mpn_lshift (b1d, b1, n, 1);      /*       2b1      */
118
830
  cy  = mpn_lshift (b0b2, b2, t, 2);      /*  4b2           */
119
830
  cy += mpn_add_n (b0b2, b0b2, b0, t);     /*  4b2      + b0 */
120
830
  if (t != n)
121
830
    cy = mpn_add_1 (b0b2 + t, b0 + t, n - t, cy);
122
830
  b0b2[n] = cy;
123
124
#if HAVE_NATIVE_mpn_add_n_sub_n
125
  if (mpn_cmp (b0b2, b1d, n+1) < 0)
126
    {
127
      mpn_add_n_sub_n (bs2, bsm2, b1d, b0b2, n+1);
128
      flags = (enum toom6_flags) (flags ^ toom6_vm2_neg);
129
    }
130
  else
131
    {
132
      mpn_add_n_sub_n (bs2, bsm2, b0b2, b1d, n+1);
133
    }
134
#else
135
830
  mpn_add_n (bs2, b0b2, b1d, n+1);
136
830
  if (mpn_cmp (b0b2, b1d, n+1) < 0)
137
616
    {
138
616
      mpn_sub_n (bsm2, b1d, b0b2, n+1);
139
616
      flags = (enum toom6_flags) (flags ^ toom6_vm2_neg);
140
616
    }
141
214
  else
142
214
    {
143
214
      mpn_sub_n (bsm2, b0b2, b1d, n+1);
144
214
    }
145
830
#endif
146
147
  /* Compute as1 and asm1.  */
148
830
  flags = (enum toom6_flags) (flags ^ (toom6_vm1_neg & mpn_toom_eval_dgr3_pm1 (as1, asm1, ap, n, s, a0a2)));
149
150
  /* Compute bs1 and bsm1.  */
151
830
  bsm1[n] = mpn_add (bsm1, b0, n, b2, t);
152
#if HAVE_NATIVE_mpn_add_n_sub_n
153
  if (bsm1[n] == 0 && mpn_cmp (bsm1, b1, n) < 0)
154
    {
155
      cy = mpn_add_n_sub_n (bs1, bsm1, b1, bsm1, n);
156
      bs1[n] = cy >> 1;
157
      flags = (enum toom6_flags) (flags ^ toom6_vm1_neg);
158
    }
159
  else
160
    {
161
      cy = mpn_add_n_sub_n (bs1, bsm1, bsm1, b1, n);
162
      bs1[n] = bsm1[n] + (cy >> 1);
163
      bsm1[n]-= cy & 1;
164
    }
165
#else
166
830
  bs1[n] = bsm1[n] + mpn_add_n (bs1, bsm1, b1, n);
167
830
  if (bsm1[n] == 0 && mpn_cmp (bsm1, b1, n) < 0)
168
364
    {
169
364
      mpn_sub_n (bsm1, b1, bsm1, n);
170
364
      flags = (enum toom6_flags) (flags ^ toom6_vm1_neg);
171
364
    }
172
466
  else
173
466
    {
174
466
      bsm1[n] -= mpn_sub_n (bsm1, bsm1, b1, n);
175
466
    }
176
830
#endif
177
178
830
  ASSERT (as1[n] <= 3);
179
830
  ASSERT (bs1[n] <= 2);
180
830
  ASSERT (asm1[n] <= 1);
181
830
  ASSERT (bsm1[n] <= 1);
182
830
  ASSERT (as2[n] <=14);
183
830
  ASSERT (bs2[n] <= 6);
184
830
  ASSERT (asm2[n] <= 9);
185
830
  ASSERT (bsm2[n] <= 4);
186
187
  /* vm1, 2n+1 limbs */
188
830
  mpn_mul_n (vm1, asm1, bsm1, n+1);  /* W4 */
189
190
  /* vm2, 2n+1 limbs */
191
830
  mpn_mul_n (vm2, asm2, bsm2, n+1);  /* W2 */
192
193
  /* v2, 2n+1 limbs */
194
830
  mpn_mul_n (v2, as2, bs2, n+1);  /* W1 */
195
196
  /* v1, 2n+1 limbs */
197
830
  mpn_mul_n (v1, as1, bs1, n+1);  /* W3 */
198
199
  /* vinf, s+t limbs */   /* W0 */
200
830
  if (s > t)  mpn_mul (vinf, a3, s, b2, t);
201
66
  else        mpn_mul (vinf, b2, t, a3, s);
202
203
  /* v0, 2n limbs */
204
830
  mpn_mul_n (v0, ap, bp, n);  /* W5 */
205
206
830
  mpn_toom_interpolate_6pts (pp, n, flags, vm1, vm2, v2, t + s);
207
208
830
#undef v0
209
830
#undef vm1
210
830
#undef v1
211
830
#undef vm2
212
830
#undef v2
213
830
#undef vinf
214
830
#undef bs1
215
830
#undef bs2
216
830
#undef bsm1
217
830
#undef bsm2
218
830
#undef asm1
219
830
#undef asm2
220
/* #undef as1 */
221
/* #undef as2 */
222
830
#undef a0a2
223
830
#undef b0b2
224
830
#undef a1a3
225
830
#undef b1d
226
830
#undef a0
227
830
#undef a1
228
830
#undef a2
229
830
#undef a3
230
830
#undef b0
231
830
#undef b1
232
830
#undef b2
233
830
}