• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /*
2  * This file derives from SFMT 1.3.3
3  * (http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/SFMT/index.html), which was
4  * released under the terms of the following license:
5  *
6  *   Copyright (c) 2006,2007 Mutsuo Saito, Makoto Matsumoto and Hiroshima
7  *   University. All rights reserved.
8  *
9  *   Redistribution and use in source and binary forms, with or without
10  *   modification, are permitted provided that the following conditions are
11  *   met:
12  *
13  *       * Redistributions of source code must retain the above copyright
14  *         notice, this list of conditions and the following disclaimer.
15  *       * Redistributions in binary form must reproduce the above
16  *         copyright notice, this list of conditions and the following
17  *         disclaimer in the documentation and/or other materials provided
18  *         with the distribution.
19  *       * Neither the name of the Hiroshima University nor the names of
20  *         its contributors may be used to endorse or promote products
21  *         derived from this software without specific prior written
22  *         permission.
23  *
24  *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
25  *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
26  *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
27  *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
28  *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
29  *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
30  *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
31  *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
32  *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
33  *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
34  *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
35  */
36 /**
37  * @file SFMT-alti.h
38  *
39  * @brief SIMD oriented Fast Mersenne Twister(SFMT)
40  * pseudorandom number generator
41  *
42  * @author Mutsuo Saito (Hiroshima University)
43  * @author Makoto Matsumoto (Hiroshima University)
44  *
45  * Copyright (C) 2007 Mutsuo Saito, Makoto Matsumoto and Hiroshima
46  * University. All rights reserved.
47  *
48  * The new BSD License is applied to this software.
49  * see LICENSE.txt
50  */
51 
52 #ifndef SFMT_ALTI_H
53 #define SFMT_ALTI_H
54 
55 /**
56  * This function represents the recursion formula in AltiVec and BIG ENDIAN.
57  * @param a a 128-bit part of the interal state array
58  * @param b a 128-bit part of the interal state array
59  * @param c a 128-bit part of the interal state array
60  * @param d a 128-bit part of the interal state array
61  * @return output
62  */
63 JEMALLOC_ALWAYS_INLINE
vec_recursion(vector unsigned int a,vector unsigned int b,vector unsigned int c,vector unsigned int d)64 vector unsigned int vec_recursion(vector unsigned int a,
65 						vector unsigned int b,
66 						vector unsigned int c,
67 						vector unsigned int d) {
68 
69     const vector unsigned int sl1 = ALTI_SL1;
70     const vector unsigned int sr1 = ALTI_SR1;
71 #ifdef ONLY64
72     const vector unsigned int mask = ALTI_MSK64;
73     const vector unsigned char perm_sl = ALTI_SL2_PERM64;
74     const vector unsigned char perm_sr = ALTI_SR2_PERM64;
75 #else
76     const vector unsigned int mask = ALTI_MSK;
77     const vector unsigned char perm_sl = ALTI_SL2_PERM;
78     const vector unsigned char perm_sr = ALTI_SR2_PERM;
79 #endif
80     vector unsigned int v, w, x, y, z;
81     x = vec_perm(a, (vector unsigned int)perm_sl, perm_sl);
82     v = a;
83     y = vec_sr(b, sr1);
84     z = vec_perm(c, (vector unsigned int)perm_sr, perm_sr);
85     w = vec_sl(d, sl1);
86     z = vec_xor(z, w);
87     y = vec_and(y, mask);
88     v = vec_xor(v, x);
89     z = vec_xor(z, y);
90     z = vec_xor(z, v);
91     return z;
92 }
93 
94 /**
95  * This function fills the internal state array with pseudorandom
96  * integers.
97  */
gen_rand_all(sfmt_t * ctx)98 JEMALLOC_INLINE void gen_rand_all(sfmt_t *ctx) {
99     int i;
100     vector unsigned int r, r1, r2;
101 
102     r1 = ctx->sfmt[N - 2].s;
103     r2 = ctx->sfmt[N - 1].s;
104     for (i = 0; i < N - POS1; i++) {
105 	r = vec_recursion(ctx->sfmt[i].s, ctx->sfmt[i + POS1].s, r1, r2);
106 	ctx->sfmt[i].s = r;
107 	r1 = r2;
108 	r2 = r;
109     }
110     for (; i < N; i++) {
111 	r = vec_recursion(ctx->sfmt[i].s, ctx->sfmt[i + POS1 - N].s, r1, r2);
112 	ctx->sfmt[i].s = r;
113 	r1 = r2;
114 	r2 = r;
115     }
116 }
117 
118 /**
119  * This function fills the user-specified array with pseudorandom
120  * integers.
121  *
122  * @param array an 128-bit array to be filled by pseudorandom numbers.
123  * @param size number of 128-bit pesudorandom numbers to be generated.
124  */
gen_rand_array(sfmt_t * ctx,w128_t * array,int size)125 JEMALLOC_INLINE void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) {
126     int i, j;
127     vector unsigned int r, r1, r2;
128 
129     r1 = ctx->sfmt[N - 2].s;
130     r2 = ctx->sfmt[N - 1].s;
131     for (i = 0; i < N - POS1; i++) {
132 	r = vec_recursion(ctx->sfmt[i].s, ctx->sfmt[i + POS1].s, r1, r2);
133 	array[i].s = r;
134 	r1 = r2;
135 	r2 = r;
136     }
137     for (; i < N; i++) {
138 	r = vec_recursion(ctx->sfmt[i].s, array[i + POS1 - N].s, r1, r2);
139 	array[i].s = r;
140 	r1 = r2;
141 	r2 = r;
142     }
143     /* main loop */
144     for (; i < size - N; i++) {
145 	r = vec_recursion(array[i - N].s, array[i + POS1 - N].s, r1, r2);
146 	array[i].s = r;
147 	r1 = r2;
148 	r2 = r;
149     }
150     for (j = 0; j < 2 * N - size; j++) {
151 	ctx->sfmt[j].s = array[j + size - N].s;
152     }
153     for (; i < size; i++) {
154 	r = vec_recursion(array[i - N].s, array[i + POS1 - N].s, r1, r2);
155 	array[i].s = r;
156 	ctx->sfmt[j++].s = r;
157 	r1 = r2;
158 	r2 = r;
159     }
160 }
161 
162 #ifndef ONLY64
163 #if defined(__APPLE__)
164 #define ALTI_SWAP (vector unsigned char) \
165 	(4, 5, 6, 7, 0, 1, 2, 3, 12, 13, 14, 15, 8, 9, 10, 11)
166 #else
167 #define ALTI_SWAP {4, 5, 6, 7, 0, 1, 2, 3, 12, 13, 14, 15, 8, 9, 10, 11}
168 #endif
169 /**
170  * This function swaps high and low 32-bit of 64-bit integers in user
171  * specified array.
172  *
173  * @param array an 128-bit array to be swaped.
174  * @param size size of 128-bit array.
175  */
swap(w128_t * array,int size)176 JEMALLOC_INLINE void swap(w128_t *array, int size) {
177     int i;
178     const vector unsigned char perm = ALTI_SWAP;
179 
180     for (i = 0; i < size; i++) {
181 	array[i].s = vec_perm(array[i].s, (vector unsigned int)perm, perm);
182     }
183 }
184 #endif
185 
186 #endif
187