xref: /netbsd-src/external/lgpl3/gmp/dist/mpn/generic/mul_basecase.c (revision 72c7faa4dbb41dbb0238d6b4a109da0d4b236dd4)
1 /* mpn_mul_basecase -- Internal routine to multiply two natural numbers
2    of length m and n.
3 
4    THIS IS AN INTERNAL FUNCTION WITH A MUTABLE INTERFACE.  IT IS ONLY
5    SAFE TO REACH THIS FUNCTION THROUGH DOCUMENTED INTERFACES.
6 
7 Copyright 1991-1994, 1996, 1997, 2000-2002 Free Software Foundation, Inc.
8 
9 This file is part of the GNU MP Library.
10 
11 The GNU MP Library is free software; you can redistribute it and/or modify
12 it under the terms of either:
13 
14   * the GNU Lesser General Public License as published by the Free
15     Software Foundation; either version 3 of the License, or (at your
16     option) any later version.
17 
18 or
19 
20   * the GNU General Public License as published by the Free Software
21     Foundation; either version 2 of the License, or (at your option) any
22     later version.
23 
24 or both in parallel, as here.
25 
26 The GNU MP Library is distributed in the hope that it will be useful, but
27 WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
28 or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
29 for more details.
30 
31 You should have received copies of the GNU General Public License and the
32 GNU Lesser General Public License along with the GNU MP Library.  If not,
33 see https://www.gnu.org/licenses/.  */
34 
35 #include "gmp-impl.h"
36 
37 
38 /* Multiply {up,usize} by {vp,vsize} and write the result to
39    {prodp,usize+vsize}.  Must have usize>=vsize.
40 
41    Note that prodp gets usize+vsize limbs stored, even if the actual result
42    only needs usize+vsize-1.
43 
44    There's no good reason to call here with vsize>=MUL_TOOM22_THRESHOLD.
45    Currently this is allowed, but it might not be in the future.
46 
47    This is the most critical code for multiplication.  All multiplies rely
48    on this, both small and huge.  Small ones arrive here immediately, huge
49    ones arrive here as this is the base case for Karatsuba's recursive
50    algorithm.  */
51 
52 void
mpn_mul_basecase(mp_ptr rp,mp_srcptr up,mp_size_t un,mp_srcptr vp,mp_size_t vn)53 mpn_mul_basecase (mp_ptr rp,
54 		  mp_srcptr up, mp_size_t un,
55 		  mp_srcptr vp, mp_size_t vn)
56 {
57   ASSERT (un >= vn);
58   ASSERT (vn >= 1);
59   ASSERT (! MPN_OVERLAP_P (rp, un+vn, up, un));
60   ASSERT (! MPN_OVERLAP_P (rp, un+vn, vp, vn));
61 
62   /* We first multiply by the low order limb (or depending on optional function
63      availability, limbs).  This result can be stored, not added, to rp.  We
64      also avoid a loop for zeroing this way.  */
65 
66 #if HAVE_NATIVE_mpn_mul_2
67   if (vn >= 2)
68     {
69       rp[un + 1] = mpn_mul_2 (rp, up, un, vp);
70       rp += 2, vp += 2, vn -= 2;
71     }
72   else
73     {
74       rp[un] = mpn_mul_1 (rp, up, un, vp[0]);
75       return;
76     }
77 #else
78   rp[un] = mpn_mul_1 (rp, up, un, vp[0]);
79   rp += 1, vp += 1, vn -= 1;
80 #endif
81 
82   /* Now accumulate the product of up[] and the next higher limb (or depending
83      on optional function availability, limbs) from vp[].  */
84 
85 #define MAX_LEFT MP_SIZE_T_MAX	/* Used to simplify loops into if statements */
86 
87 
88 #if HAVE_NATIVE_mpn_addmul_6
89   while (vn >= 6)
90     {
91       rp[un + 6 - 1] = mpn_addmul_6 (rp, up, un, vp);
92       if (MAX_LEFT == 6)
93 	return;
94       rp += 6, vp += 6, vn -= 6;
95       if (MAX_LEFT < 2 * 6)
96 	break;
97     }
98 #undef MAX_LEFT
99 #define MAX_LEFT (6 - 1)
100 #endif
101 
102 #if HAVE_NATIVE_mpn_addmul_5
103   while (vn >= 5)
104     {
105       rp[un + 5 - 1] = mpn_addmul_5 (rp, up, un, vp);
106       if (MAX_LEFT == 5)
107 	return;
108       rp += 5, vp += 5, vn -= 5;
109       if (MAX_LEFT < 2 * 5)
110 	break;
111     }
112 #undef MAX_LEFT
113 #define MAX_LEFT (5 - 1)
114 #endif
115 
116 #if HAVE_NATIVE_mpn_addmul_4
117   while (vn >= 4)
118     {
119       rp[un + 4 - 1] = mpn_addmul_4 (rp, up, un, vp);
120       if (MAX_LEFT == 4)
121 	return;
122       rp += 4, vp += 4, vn -= 4;
123       if (MAX_LEFT < 2 * 4)
124 	break;
125     }
126 #undef MAX_LEFT
127 #define MAX_LEFT (4 - 1)
128 #endif
129 
130 #if HAVE_NATIVE_mpn_addmul_3
131   while (vn >= 3)
132     {
133       rp[un + 3 - 1] = mpn_addmul_3 (rp, up, un, vp);
134       if (MAX_LEFT == 3)
135 	return;
136       rp += 3, vp += 3, vn -= 3;
137       if (MAX_LEFT < 2 * 3)
138 	break;
139     }
140 #undef MAX_LEFT
141 #define MAX_LEFT (3 - 1)
142 #endif
143 
144 #if HAVE_NATIVE_mpn_addmul_2
145   while (vn >= 2)
146     {
147       rp[un + 2 - 1] = mpn_addmul_2 (rp, up, un, vp);
148       if (MAX_LEFT == 2)
149 	return;
150       rp += 2, vp += 2, vn -= 2;
151       if (MAX_LEFT < 2 * 2)
152 	break;
153     }
154 #undef MAX_LEFT
155 #define MAX_LEFT (2 - 1)
156 #endif
157 
158   while (vn >= 1)
159     {
160       rp[un] = mpn_addmul_1 (rp, up, un, vp[0]);
161       if (MAX_LEFT == 1)
162 	return;
163       rp += 1, vp += 1, vn -= 1;
164     }
165 }
166