Annotation of OpenXM_contrib/gmp/mpn/generic/addsub_n.c, Revision 1.1.1.1
1.1 maekawa 1: /* mpn_addsub_n -- Add and Subtract two limb vectors of equal, non-zero length.
2:
3: Copyright (C) 1999, 2000 Free Software Foundation, Inc.
4:
5: This file is part of the GNU MP Library.
6:
7: The GNU MP Library is free software; you can redistribute it and/or modify
8: it under the terms of the GNU Lesser General Public License as published by
9: the Free Software Foundation; either version 2.1 of the License, or (at your
10: option) any later version.
11:
12: The GNU MP Library is distributed in the hope that it will be useful, but
13: WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
14: or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
15: License for more details.
16:
17: You should have received a copy of the GNU Lesser General Public License
18: along with the GNU MP Library; see the file COPYING.LIB. If not, write to
19: the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
20: MA 02111-1307, USA. */
21:
22: #include "gmp.h"
23: #include "gmp-impl.h"
24:
25: #ifndef L1_CACHE_SIZE
26: #define L1_CACHE_SIZE 8192 /* only 68040 has less than this */
27: #endif
28:
29: #define PART_SIZE (L1_CACHE_SIZE / BYTES_PER_MP_LIMB / 6)
30:
31:
32: /* mpn_addsub_n.
33: r1[] = s1[] + s2[]
34: r2[] = s1[] - s2[]
35: All operands have n limbs.
36: In-place operations allowed. */
37: mp_limb_t
38: #if __STDC__
39: mpn_addsub_n (mp_ptr r1p, mp_ptr r2p, mp_srcptr s1p, mp_srcptr s2p, mp_size_t n)
40: #else
41: mpn_addsub_n (r1p, r2p, s1p, s2p, n)
42: mp_ptr r1p, r2p;
43: mp_srcptr s1p, s2p;
44: mp_size_t n;
45: #endif
46: {
47: mp_limb_t acyn, acyo; /* carry for add */
48: mp_limb_t scyn, scyo; /* carry for subtract */
49: mp_size_t off; /* offset in operands */
50: mp_size_t this_n; /* size of current chunk */
51:
52: /* We alternatingly add and subtract in chunks that fit into the (L1)
53: cache. Since the chunks are several hundred limbs, the function call
54: overhead is insignificant, but we get much better locality. */
55:
56: /* We have three variant of the inner loop, the proper loop is chosen
57: depending on whether r1 or r2 are the same operand as s1 or s2. */
58:
59: if (r1p != s1p && r1p != s2p)
60: {
61: /* r1 is not identical to either input operand. We can therefore write
62: to r1 directly, without using temporary storage. */
63: acyo = 0;
64: scyo = 0;
65: for (off = 0; off < n; off += PART_SIZE)
66: {
67: this_n = MIN (n - off, PART_SIZE);
68: #if HAVE_NATIVE_mpn_add_nc || !HAVE_NATIVE_mpn_add_n
69: acyo = mpn_add_nc (r1p + off, s1p + off, s2p + off, this_n, acyo);
70: #else
71: acyn = mpn_add_n (r1p + off, s1p + off, s2p + off, this_n);
72: acyo = acyn + mpn_add_1 (r1p + off, r1p + off, this_n, acyo);
73: #endif
74: #if HAVE_NATIVE_mpn_sub_nc || !HAVE_NATIVE_mpn_sub_n
75: scyo = mpn_sub_nc (r2p + off, s1p + off, s2p + off, this_n, scyo);
76: #else
77: scyn = mpn_sub_n (r2p + off, s1p + off, s2p + off, this_n);
78: scyo = scyn + mpn_sub_1 (r2p + off, r2p + off, this_n, scyo);
79: #endif
80: }
81: }
82: else if (r2p != s1p && r2p != s2p)
83: {
84: /* r2 is not identical to either input operand. We can therefore write
85: to r2 directly, without using temporary storage. */
86: acyo = 0;
87: scyo = 0;
88: for (off = 0; off < n; off += PART_SIZE)
89: {
90: this_n = MIN (n - off, PART_SIZE);
91: #if HAVE_NATIVE_mpn_sub_nc || !HAVE_NATIVE_mpn_sub_n
92: scyo = mpn_sub_nc (r2p + off, s1p + off, s2p + off, this_n, scyo);
93: #else
94: scyn = mpn_sub_n (r2p + off, s1p + off, s2p + off, this_n);
95: scyo = scyn + mpn_sub_1 (r2p + off, r2p + off, this_n, scyo);
96: #endif
97: #if HAVE_NATIVE_mpn_add_nc || !HAVE_NATIVE_mpn_add_n
98: acyo = mpn_add_nc (r1p + off, s1p + off, s2p + off, this_n, acyo);
99: #else
100: acyn = mpn_add_n (r1p + off, s1p + off, s2p + off, this_n);
101: acyo = acyn + mpn_add_1 (r1p + off, r1p + off, this_n, acyo);
102: #endif
103: }
104: }
105: else
106: {
107: /* r1 and r2 are identical to s1 and s2 (r1==s1 and r2=s2 or vice versa)
108: Need temporary storage. */
109: mp_limb_t tp[PART_SIZE];
110: acyo = 0;
111: scyo = 0;
112: for (off = 0; off < n; off += PART_SIZE)
113: {
114: this_n = MIN (n - off, PART_SIZE);
115: #if HAVE_NATIVE_mpn_add_nc || !HAVE_NATIVE_mpn_add_n
116: acyo = mpn_add_nc (tp, s1p + off, s2p + off, this_n, acyo);
117: #else
118: acyn = mpn_add_n (tp, s1p + off, s2p + off, this_n);
119: acyo = acyn + mpn_add_1 (tp, tp, this_n, acyo);
120: #endif
121: #if HAVE_NATIVE_mpn_sub_nc || !HAVE_NATIVE_mpn_sub_n
122: scyo = mpn_sub_nc (r2p + off, s1p + off, s2p + off, this_n, scyo);
123: #else
124: scyn = mpn_sub_n (r2p + off, s1p + off, s2p + off, this_n);
125: scyo = scyn + mpn_sub_1 (r2p + off, r2p + off, this_n, scyo);
126: #endif
127: MPN_COPY (r1p + off, tp, this_n);
128: }
129: }
130:
131: return 2 * acyo + scyo;
132: }
133:
134: #ifdef MAIN
135: #include <stdlib.h>
136: #include <stdio.h>
137: #include "timing.h"
138:
139: long cputime ();
140:
141: int
142: main (int argc, char **argv)
143: {
144: mp_ptr r1p, r2p, s1p, s2p;
145: double t;
146: mp_size_t n;
147:
148: n = strtol (argv[1], 0, 0);
149:
150: r1p = malloc (n * BYTES_PER_MP_LIMB);
151: r2p = malloc (n * BYTES_PER_MP_LIMB);
152: s1p = malloc (n * BYTES_PER_MP_LIMB);
153: s2p = malloc (n * BYTES_PER_MP_LIMB);
154: TIME (t,(mpn_add_n(r1p,s1p,s2p,n),mpn_sub_n(r1p,s1p,s2p,n)));
155: printf (" separate add and sub: %.3f\n", t);
156: TIME (t,mpn_addsub_n(r1p,r2p,s1p,s2p,n));
157: printf ("combined addsub separate variables: %.3f\n", t);
158: TIME (t,mpn_addsub_n(r1p,r2p,r1p,s2p,n));
159: printf (" combined addsub r1 overlap: %.3f\n", t);
160: TIME (t,mpn_addsub_n(r1p,r2p,r1p,s2p,n));
161: printf (" combined addsub r2 overlap: %.3f\n", t);
162: TIME (t,mpn_addsub_n(r1p,r2p,r1p,r2p,n));
163: printf (" combined addsub in-place: %.3f\n", t);
164:
165: return 0;
166: }
167: #endif
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>