1 |
207 |
jeremybenn |
/* -------------------------------------------------------------- */
|
2 |
|
|
/* (C)Copyright 2006,2008, */
|
3 |
|
|
/* International Business Machines Corporation */
|
4 |
|
|
/* All Rights Reserved. */
|
5 |
|
|
/* */
|
6 |
|
|
/* Redistribution and use in source and binary forms, with or */
|
7 |
|
|
/* without modification, are permitted provided that the */
|
8 |
|
|
/* following conditions are met: */
|
9 |
|
|
/* */
|
10 |
|
|
/* - Redistributions of source code must retain the above copyright*/
|
11 |
|
|
/* notice, this list of conditions and the following disclaimer. */
|
12 |
|
|
/* */
|
13 |
|
|
/* - Redistributions in binary form must reproduce the above */
|
14 |
|
|
/* copyright notice, this list of conditions and the following */
|
15 |
|
|
/* disclaimer in the documentation and/or other materials */
|
16 |
|
|
/* provided with the distribution. */
|
17 |
|
|
/* */
|
18 |
|
|
/* - Neither the name of IBM Corporation nor the names of its */
|
19 |
|
|
/* contributors may be used to endorse or promote products */
|
20 |
|
|
/* derived from this software without specific prior written */
|
21 |
|
|
/* permission. */
|
22 |
|
|
/* */
|
23 |
|
|
/* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND */
|
24 |
|
|
/* CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, */
|
25 |
|
|
/* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF */
|
26 |
|
|
/* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE */
|
27 |
|
|
/* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR */
|
28 |
|
|
/* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, */
|
29 |
|
|
/* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT */
|
30 |
|
|
/* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; */
|
31 |
|
|
/* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) */
|
32 |
|
|
/* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN */
|
33 |
|
|
/* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR */
|
34 |
|
|
/* OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, */
|
35 |
|
|
/* EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
|
36 |
|
|
/* -------------------------------------------------------------- */
|
37 |
|
|
/* PROLOG END TAG zYx */
|
38 |
|
|
#ifndef __SIMD_MATH_H__
|
39 |
|
|
#define __SIMD_MATH_H__ 1
|
40 |
|
|
|
41 |
|
|
#ifdef __cplusplus
|
42 |
|
|
extern "C" {
|
43 |
|
|
#endif
|
44 |
|
|
|
45 |
|
|
#if !defined(__SPU__) && !defined(__ALTIVEC__)
|
46 |
|
|
#error Bad platform
|
47 |
|
|
#else
|
48 |
|
|
|
49 |
|
|
#define SM_E 2.7182818284590452354 /* e */
|
50 |
|
|
#define SM_LOG2E 1.4426950408889634074 /* log_2 e */
|
51 |
|
|
#define SM_LOG10E 0.43429448190325182765 /* log_10 e */
|
52 |
|
|
#define SM_LN2 0.69314718055994530942 /* log_e 2 */
|
53 |
|
|
#define SM_LN10 2.30258509299404568402 /* log_e 10 */
|
54 |
|
|
#define SM_PI 3.14159265358979323846 /* pi */
|
55 |
|
|
#define SM_PI_2 1.57079632679489661923 /* pi/2 */
|
56 |
|
|
#define SM_PI_4 0.78539816339744830962 /* pi/4 */
|
57 |
|
|
#define SM_1_PI 0.31830988618379067154 /* 1/pi */
|
58 |
|
|
#define SM_2_PI 0.63661977236758134308 /* 2/pi */
|
59 |
|
|
#define SM_2_SQRTPI 1.12837916709551257390 /* 2/sqrt(pi) */
|
60 |
|
|
#define SM_SQRT2 1.41421356237309504880 /* sqrt(2) */
|
61 |
|
|
#define SM_SQRT1_2 0.70710678118654752440 /* 1/sqrt(2) */
|
62 |
|
|
|
63 |
|
|
/* Typedefs common to both SPU and PPU */
|
64 |
|
|
typedef struct divi4_s {
|
65 |
|
|
vector signed int quot;
|
66 |
|
|
vector signed int rem;
|
67 |
|
|
} divi4_t;
|
68 |
|
|
|
69 |
|
|
typedef struct divu4_s {
|
70 |
|
|
vector unsigned int quot;
|
71 |
|
|
vector unsigned int rem;
|
72 |
|
|
} divu4_t;
|
73 |
|
|
|
74 |
|
|
|
75 |
|
|
/* Functions common to both SPU and PPU */
|
76 |
|
|
vector signed int absi4(vector signed int x);
|
77 |
|
|
vector float acosf4(vector float x);
|
78 |
|
|
vector float acoshf4(vector float x);
|
79 |
|
|
vector float asinf4(vector float x);
|
80 |
|
|
vector float asinhf4(vector float x);
|
81 |
|
|
vector float atanf4(vector float x);
|
82 |
|
|
vector float atanhf4(vector float x);
|
83 |
|
|
vector float atan2f4(vector float y, vector float x);
|
84 |
|
|
vector float cbrtf4(vector float x);
|
85 |
|
|
vector float ceilf4(vector float x);
|
86 |
|
|
vector float copysignf4(vector float x, vector float y);
|
87 |
|
|
vector float cosf4(vector float x);
|
88 |
|
|
vector float coshf4(vector float x);
|
89 |
|
|
vector float divf4(vector float x, vector float y);
|
90 |
|
|
vector float divf4_fast(vector float x, vector float y);
|
91 |
|
|
divi4_t divi4(vector signed int dividend, vector signed int divisor);
|
92 |
|
|
divu4_t divu4(vector unsigned int dividend, vector unsigned int divisor);
|
93 |
|
|
vector float erff4(vector float x);
|
94 |
|
|
vector float erfcf4(vector float x);
|
95 |
|
|
vector float exp2f4(vector float x);
|
96 |
|
|
vector float expf4(vector float x);
|
97 |
|
|
vector float expm1f4(vector float x);
|
98 |
|
|
vector float fabsf4(vector float value);
|
99 |
|
|
vector float fdimf4(vector float x, vector float y);
|
100 |
|
|
vector float floorf4(vector float value);
|
101 |
|
|
vector float fmaf4(vector float x, vector float y, vector float z);
|
102 |
|
|
vector float fmaxf4(vector float x, vector float y);
|
103 |
|
|
vector float fminf4(vector float x, vector float y);
|
104 |
|
|
vector float fmodf4(vector float x, vector float y);
|
105 |
|
|
vector signed int fpclassifyf4(vector float x);
|
106 |
|
|
vector float frexpf4(vector float x, vector signed int *pexp);
|
107 |
|
|
vector float hypotf4(vector float x, vector float y);
|
108 |
|
|
vector signed int ilogbf4(vector float x);
|
109 |
|
|
vector signed int irintf4(vector float x);
|
110 |
|
|
vector signed int iroundf4(vector float x);
|
111 |
|
|
vector unsigned int is0denormf4(vector float x);
|
112 |
|
|
vector unsigned int isequalf4(vector float x, vector float y);
|
113 |
|
|
vector unsigned int isfinitef4(vector float x);
|
114 |
|
|
vector unsigned int isgreaterf4(vector float x, vector float y);
|
115 |
|
|
vector unsigned int isgreaterequalf4(vector float x, vector float y);
|
116 |
|
|
vector unsigned int isinff4(vector float x);
|
117 |
|
|
vector unsigned int islessf4(vector float x, vector float y);
|
118 |
|
|
vector unsigned int islessequalf4(vector float x, vector float y);
|
119 |
|
|
vector unsigned int islessgreaterf4(vector float x, vector float y);
|
120 |
|
|
vector unsigned int isnanf4(vector float x);
|
121 |
|
|
vector unsigned int isnormalf4(vector float x);
|
122 |
|
|
vector unsigned int isunorderedf4(vector float x, vector float y);
|
123 |
|
|
vector float ldexpf4(vector float x, vector signed int exp);
|
124 |
|
|
vector float lgammaf4(vector float x);
|
125 |
|
|
vector float log10f4(vector float x);
|
126 |
|
|
vector float log1pf4(vector float x);
|
127 |
|
|
vector float log2f4(vector float x);
|
128 |
|
|
vector float logbf4(vector float x);
|
129 |
|
|
vector float logf4(vector float x);
|
130 |
|
|
vector float modff4(vector float x, vector float *pint);
|
131 |
|
|
vector float nearbyintf4(vector float x);
|
132 |
|
|
vector float negatef4(vector float x);
|
133 |
|
|
vector signed int negatei4(vector signed int x);
|
134 |
|
|
vector float nextafterf4(vector float x, vector float y);
|
135 |
|
|
vector float powf4(vector float x, vector float y);
|
136 |
|
|
vector float recipf4(vector float a);
|
137 |
|
|
vector float remainderf4(vector float x, vector float y);
|
138 |
|
|
vector float remquof4(vector float x, vector float y, vector signed int *quo);
|
139 |
|
|
vector float rintf4(vector float x);
|
140 |
|
|
vector float roundf4(vector float x);
|
141 |
|
|
vector float rsqrtf4(vector float value);
|
142 |
|
|
vector float scalbnf4(vector float x, vector signed int exp);
|
143 |
|
|
vector unsigned int signbitf4(vector float x);
|
144 |
|
|
void sincosf4(vector float x, vector float *sx, vector float *cx);
|
145 |
|
|
vector float sinf4(vector float x);
|
146 |
|
|
vector float sinhf4(vector float x);
|
147 |
|
|
vector float sqrtf4(vector float in);
|
148 |
|
|
vector float tanf4(vector float angle);
|
149 |
|
|
vector float tanhf4(vector float x);
|
150 |
|
|
vector float tgammaf4(vector float x);
|
151 |
|
|
vector float truncf4(vector float x);
|
152 |
|
|
|
153 |
|
|
|
154 |
|
|
#ifdef __SPU__
|
155 |
|
|
/* Typedefs specific to SPU */
|
156 |
|
|
typedef struct llroundf4_s {
|
157 |
|
|
vector signed long long vll[2];
|
158 |
|
|
} llroundf4_t;
|
159 |
|
|
|
160 |
|
|
typedef struct lldivi2_s {
|
161 |
|
|
vector signed long long quot;
|
162 |
|
|
vector signed long long rem;
|
163 |
|
|
} lldivi2_t;
|
164 |
|
|
|
165 |
|
|
typedef struct lldivu2_s {
|
166 |
|
|
vector unsigned long long quot;
|
167 |
|
|
vector unsigned long long rem;
|
168 |
|
|
} lldivu2_t;
|
169 |
|
|
|
170 |
|
|
|
171 |
|
|
/* Functions specific to SPU */
|
172 |
|
|
llroundf4_t llrintf4(vector float in);
|
173 |
|
|
llroundf4_t llroundf4 (vector float x);
|
174 |
|
|
vector double acosd2(vector double x);
|
175 |
|
|
vector double acoshd2(vector double x);
|
176 |
|
|
vector double asind2(vector double x);
|
177 |
|
|
vector double asinhd2(vector double x);
|
178 |
|
|
vector double atan2d2(vector double y, vector double x);
|
179 |
|
|
vector double atand2(vector double x);
|
180 |
|
|
vector double atanhd2(vector double x);
|
181 |
|
|
vector double cbrtd2(vector double x);
|
182 |
|
|
vector double ceild2(vector double x);
|
183 |
|
|
vector float ceilf4_fast(vector float x);
|
184 |
|
|
vector double copysignd2(vector double x, vector double y);
|
185 |
|
|
vector double cosd2(vector double x);
|
186 |
|
|
vector double coshd2(vector double x);
|
187 |
|
|
vector double divd2(vector double a, vector double b);
|
188 |
|
|
vector double erfcd2(vector double x);
|
189 |
|
|
vector double erfd2(vector double x);
|
190 |
|
|
vector double exp2d2(vector double x);
|
191 |
|
|
vector double expd2(vector double x);
|
192 |
|
|
vector double expm1d2(vector double x);
|
193 |
|
|
vector double fabsd2(vector double x);
|
194 |
|
|
vector double fdimd2(vector double x, vector double y);
|
195 |
|
|
vector double floord2(vector double x);
|
196 |
|
|
vector float floorf4_fast(vector float value);
|
197 |
|
|
vector double fmad2(vector double x, vector double y, vector double z);
|
198 |
|
|
vector double fmaxd2(vector double x, vector double y);
|
199 |
|
|
vector double fmind2(vector double x, vector double y);
|
200 |
|
|
vector double fmodd2(vector double x, vector double y);
|
201 |
|
|
vector float fmodf4_fast(vector float x, vector float y);
|
202 |
|
|
vector signed long long fpclassifyd2(vector double x);
|
203 |
|
|
vector double frexpd2(vector double x, vector signed long long *pexp);
|
204 |
|
|
vector double hypotd2(vector double x, vector double y);
|
205 |
|
|
vector signed long long ilogbd2(vector double x);
|
206 |
|
|
vector unsigned long long is0denormd2(vector double x);
|
207 |
|
|
vector unsigned long long isequald2(vector double x, vector double y);
|
208 |
|
|
vector unsigned long long isfinited2(vector double x);
|
209 |
|
|
vector unsigned long long isgreaterd2(vector double x, vector double y);
|
210 |
|
|
vector unsigned long long isgreaterequald2(vector double x, vector double y);
|
211 |
|
|
vector unsigned long long isinfd2(vector double x);
|
212 |
|
|
vector unsigned long long islessd2(vector double x, vector double y);
|
213 |
|
|
vector unsigned long long islessequald2(vector double x, vector double y);
|
214 |
|
|
vector unsigned long long islessgreaterd2(vector double x, vector double y);
|
215 |
|
|
vector unsigned long long isnand2(vector double x);
|
216 |
|
|
vector unsigned long long isnormald2(vector double x);
|
217 |
|
|
vector unsigned long long isunorderedd2(vector double x, vector double y);
|
218 |
|
|
vector double ldexpd2(vector double x, vector signed long long exp);
|
219 |
|
|
vector signed long long llabsi2(vector signed long long x);
|
220 |
|
|
lldivi2_t lldivi2(vector signed long long x, vector signed long long y);
|
221 |
|
|
lldivu2_t lldivu2(vector unsigned long long x, vector unsigned long long y);
|
222 |
|
|
vector double lgammad2(vector double x);
|
223 |
|
|
vector signed long long llrintd2(vector double in);
|
224 |
|
|
vector signed long long llroundd2(vector double x);
|
225 |
|
|
vector double log10d2(vector double x);
|
226 |
|
|
vector double log1pd2(vector double x);
|
227 |
|
|
vector double log2d2(vector double x);
|
228 |
|
|
vector double logbd2(vector double x);
|
229 |
|
|
vector double logd2(vector double x);
|
230 |
|
|
vector double modfd2(vector double x, vector double* pint);
|
231 |
|
|
vector double nearbyintd2(vector double x);
|
232 |
|
|
vector double negated2(vector double x);
|
233 |
|
|
vector double nextafterd2(vector double x, vector double y);
|
234 |
|
|
vector signed long long negatell2(vector signed long long x);
|
235 |
|
|
vector double powd2(vector double x, vector double y);
|
236 |
|
|
vector double recipd2(vector double value_d);
|
237 |
|
|
vector float recipf4_fast(vector float a);
|
238 |
|
|
vector double remainderd2(vector double x, vector double y);
|
239 |
|
|
vector double remquod2(vector double x, vector double y, vector signed long long *quo);
|
240 |
|
|
vector double rintd2(vector double x);
|
241 |
|
|
vector double roundd2(vector double x);
|
242 |
|
|
vector double rsqrtd2(vector double x);
|
243 |
|
|
vector double scalbllnd2(vector double x, vector signed long long n);
|
244 |
|
|
vector unsigned long long signbitd2(vector double x);
|
245 |
|
|
void sincosd2(vector double x, vector double *sx, vector double *cx);
|
246 |
|
|
vector double sind2(vector double x);
|
247 |
|
|
vector double sinhd2(vector double x);
|
248 |
|
|
vector double sqrtd2(vector double x);
|
249 |
|
|
vector float sqrtf4_fast(vector float in);
|
250 |
|
|
vector double tand2(vector double x);
|
251 |
|
|
vector double tanhd2(vector double x);
|
252 |
|
|
vector double tgammad2(vector double x);
|
253 |
|
|
vector double truncd2(vector double x);
|
254 |
|
|
|
255 |
|
|
#endif /* __SPU__ */
|
256 |
|
|
|
257 |
|
|
/* Functions specific to PPU */
|
258 |
|
|
#ifdef __ALTIVEC__
|
259 |
|
|
#endif
|
260 |
|
|
|
261 |
|
|
#endif /* __SPU__ || __ALTIVEC__ */
|
262 |
|
|
|
263 |
|
|
#ifdef __cplusplus
|
264 |
|
|
}
|
265 |
|
|
#endif
|
266 |
|
|
|
267 |
|
|
#endif /* __SIMD_MATH_H__ */
|