summaryrefslogtreecommitdiff
path: root/src/gallium/auxiliary/util/u_math.h
blob: b0807c13392359dfea266f1d4a474ec9e6cf03d6 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
/**************************************************************************
 * 
 * Copyright 2008 Tungsten Graphics, Inc., Cedar Park, Texas.
 * All Rights Reserved.
 * 
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the
 * "Software"), to deal in the Software without restriction, including
 * without limitation the rights to use, copy, modify, merge, publish,
 * distribute, sub license, and/or sell copies of the Software, and to
 * permit persons to whom the Software is furnished to do so, subject to
 * the following conditions:
 * 
 * The above copyright notice and this permission notice (including the
 * next paragraph) shall be included in all copies or substantial portions
 * of the Software.
 * 
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
 * IN NO EVENT SHALL TUNGSTEN GRAPHICS AND/OR ITS SUPPLIERS BE LIABLE FOR
 * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
 * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
 * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 * 
 **************************************************************************/


/**
 * Math utilities and approximations for common math functions.
 * Reduced precision is usually acceptable in shaders...
 *
 * "fast" is used in the names of functions which are low-precision,
 * or at least lower-precision than the normal C lib functions.
 */


#ifndef U_MATH_H
#define U_MATH_H


#include "pipe/p_compiler.h"
#include "util/u_debug.h"


#ifdef __cplusplus
extern "C" {
#endif


#if defined(PIPE_SUBSYSTEM_WINDOWS_MINIPORT)
__inline double ceil(double val)
{
   double ceil_val;

   if ((val - (long) val) == 0) {
      ceil_val = val;
   }
   else {
      if (val > 0) {
         ceil_val = (long) val + 1;
      }
      else {
         ceil_val = (long) val;
      }
   }

   return ceil_val;
}

#ifndef PIPE_SUBSYSTEM_WINDOWS_CE_OGL
__inline double floor(double val)
{
   double floor_val;

   if ((val - (long) val) == 0) {
      floor_val = val;
   }
   else {
      if (val > 0) {
         floor_val = (long) val;
      }
      else {
         floor_val = (long) val - 1;
      }
   }

   return floor_val;
}
#endif

#pragma function(pow)
__inline double __cdecl pow(double val, double exponent)
{
   /* XXX */
   assert(0);
   return 0;
}

#pragma function(log)
__inline double __cdecl log(double val)
{
   /* XXX */
   assert(0);
   return 0;
}

#pragma function(atan2)
__inline double __cdecl atan2(double val)
{
   /* XXX */
   assert(0);
   return 0;
}
#else
#include <math.h>
#include <stdarg.h>
#endif


#if defined(_MSC_VER) 

#if _MSC_VER < 1400 && !defined(__cplusplus) || defined(PIPE_SUBSYSTEM_WINDOWS_CE)
 
static INLINE float cosf( float f ) 
{
   return (float) cos( (double) f );
}

static INLINE float sinf( float f ) 
{
   return (float) sin( (double) f );
}

static INLINE float ceilf( float f ) 
{
   return (float) ceil( (double) f );
}

static INLINE float floorf( float f ) 
{
   return (float) floor( (double) f );
}

static INLINE float powf( float f, float g ) 
{
   return (float) pow( (double) f, (double) g );
}

static INLINE float sqrtf( float f ) 
{
   return (float) sqrt( (double) f );
}

static INLINE float fabsf( float f ) 
{
   return (float) fabs( (double) f );
}

static INLINE float logf( float f ) 
{
   return (float) log( (double) f );
}

#else
/* Work-around an extra semi-colon in VS 2005 logf definition */
#ifdef logf
#undef logf
#define logf(x) ((float)log((double)(x)))
#endif /* logf */
#endif

static INLINE double log2( double x )
{
   const double invln2 = 1.442695041;
   return log( x ) * invln2;
}

#endif /* _MSC_VER */





#define POW2_TABLE_SIZE_LOG2 9
#define POW2_TABLE_SIZE (1 << POW2_TABLE_SIZE_LOG2)
#define POW2_TABLE_OFFSET (POW2_TABLE_SIZE/2)
#define POW2_TABLE_SCALE ((float)(POW2_TABLE_SIZE/2))
extern float pow2_table[POW2_TABLE_SIZE];


/**
 * Initialize math module.  This should be called before using any
 * other functions in this module.
 */
extern void
util_init_math(void);


union fi {
   float f;
   int32_t i;
   uint32_t ui;
};


/**
 * Fast version of 2^x
 * Identity: exp2(a + b) = exp2(a) * exp2(b)
 * Let ipart = int(x)
 * Let fpart = x - ipart;
 * So, exp2(x) = exp2(ipart) * exp2(fpart)
 * Compute exp2(ipart) with i << ipart
 * Compute exp2(fpart) with lookup table.
 */
static INLINE float
util_fast_exp2(float x)
{
   int32_t ipart;
   float fpart, mpart;
   union fi epart;

   if(x > 129.00000f)
      return 3.402823466e+38f;

   if (x < -126.99999f)
      return 0.0f;

   ipart = (int32_t) x;
   fpart = x - (float) ipart;

   /* same as
    *   epart.f = (float) (1 << ipart)
    * but faster and without integer overflow for ipart > 31
    */
   epart.i = (ipart + 127 ) << 23;

   mpart = pow2_table[POW2_TABLE_OFFSET + (int)(fpart * POW2_TABLE_SCALE)];

   return epart.f * mpart;
}


/**
 * Fast approximation to exp(x).
 */
static INLINE float
util_fast_exp(float x)
{
   const float k = 1.44269f; /* = log2(e) */
   return util_fast_exp2(k * x);
}


#define LOG2_TABLE_SIZE_LOG2 16
#define LOG2_TABLE_SCALE (1 << LOG2_TABLE_SIZE_LOG2)
#define LOG2_TABLE_SIZE (LOG2_TABLE_SCALE + 1)
extern float log2_table[LOG2_TABLE_SIZE];


/**
 * Fast approximation to log2(x).
 */
static INLINE float
util_fast_log2(float x)
{
   union fi num;
   float epart, mpart;
   num.f = x;
   epart = (float)(((num.i & 0x7f800000) >> 23) - 127);
   /* mpart = log2_table[mantissa*LOG2_TABLE_SCALE + 0.5] */
   mpart = log2_table[((num.i & 0x007fffff) + (1 << (22 - LOG2_TABLE_SIZE_LOG2))) >> (23 - LOG2_TABLE_SIZE_LOG2)];
   return epart + mpart;
}


/**
 * Fast approximation to x^y.
 */
static INLINE float
util_fast_pow(float x, float y)
{
   return util_fast_exp2(util_fast_log2(x) * y);
}


/**
 * Floor(x), returned as int.
 */
static INLINE int
util_ifloor(float f)
{
   int ai, bi;
   double af, bf;
   union fi u;
   af = (3 << 22) + 0.5 + (double) f;
   bf = (3 << 22) + 0.5 - (double) f;
   u.f = (float) af;  ai = u.i;
   u.f = (float) bf;  bi = u.i;
   return (ai - bi) >> 1;
}


/**
 * Round float to nearest int.
 */
static INLINE int
util_iround(float f)
{
#if defined(PIPE_CC_GCC) && defined(PIPE_ARCH_X86) 
   int r;
   __asm__ ("fistpl %0" : "=m" (r) : "t" (f) : "st");
   return r;
#elif defined(PIPE_CC_MSVC) && defined(PIPE_ARCH_X86)
   int r;
   _asm {
      fld f
      fistp r
   }
   return r;
#else
   if (f >= 0.0f)
      return (int) (f + 0.5f);
   else
      return (int) (f - 0.5f);
#endif
}



/**
 * Test if x is NaN or +/- infinity.
 */
static INLINE boolean
util_is_inf_or_nan(float x)
{
   union fi tmp;
   tmp.f = x;
   return !(int)((unsigned int)((tmp.i & 0x7fffffff)-0x7f800000) >> 31);
}


/**
 * Find first bit set in word.  Least significant bit is 1.
 * Return 0 if no bits set.
 */
#if defined(_MSC_VER) && _MSC_VER >= 1300
static INLINE
unsigned long ffs( unsigned long u )
{
   unsigned long i;
   if (_BitScanForward(&i, u))
      return i + 1;
   else
      return 0;
}
#elif defined(PIPE_CC_MSVC) && defined(PIPE_ARCH_X86)
static INLINE
unsigned ffs( unsigned u )
{
   unsigned i;

   if (u == 0) {
      return 0;
   }

   __asm bsf eax, [u]
   __asm inc eax
   __asm mov [i], eax

   return i;
}
#elif defined(__MINGW32__)
#define ffs __builtin_ffs
#endif


/* Could also binary search for the highest bit.
 */
static INLINE unsigned
util_unsigned_logbase2(unsigned n)
{
   unsigned log2 = 0;
   while (n >>= 1)
      ++log2;
   return log2;
}


/**
 * Return float bits.
 */
static INLINE unsigned
fui( float f )
{
   union fi fi;
   fi.f = f;
   return fi.ui;
}


/**
 * Convert ubyte to float in [0, 1].
 * XXX a 256-entry lookup table would be slightly faster.
 */
static INLINE float
ubyte_to_float(ubyte ub)
{
   return (float) ub * (1.0f / 255.0f);
}


/**
 * Convert float in [0,1] to ubyte in [0,255] with clamping.
 */
static INLINE ubyte
float_to_ubyte(float f)
{
   const int ieee_0996 = 0x3f7f0000;   /* 0.996 or so */
   union fi tmp;

   tmp.f = f;
   if (tmp.i < 0) {
      return (ubyte) 0;
   }
   else if (tmp.i >= ieee_0996) {
      return (ubyte) 255;
   }
   else {
      tmp.f = tmp.f * (255.0f/256.0f) + 32768.0f;
      return (ubyte) tmp.i;
   }
}


/**
 * Calc log base 2
 */
static INLINE unsigned
util_logbase2(unsigned n)
{
   unsigned log2 = 0;
   while (n >>= 1)
      ++log2;
   return log2;
}


/**
 * Clamp X to [MIN, MAX].
 * This is a macro to allow float, int, uint, etc. types.
 */
#define CLAMP( X, MIN, MAX )  ( (X)<(MIN) ? (MIN) : ((X)>(MAX) ? (MAX) : (X)) )

#define MIN2( A, B )   ( (A)<(B) ? (A) : (B) )
#define MAX2( A, B )   ( (A)>(B) ? (A) : (B) )


static INLINE int
align(int value, int alignment)
{
   return (value + alignment - 1) & ~(alignment - 1);
}

static INLINE unsigned
minify(unsigned value)
{
    return MAX2(1, value >> 1);
}

#ifndef COPY_4V
#define COPY_4V( DST, SRC )         \
do {                                \
   (DST)[0] = (SRC)[0];             \
   (DST)[1] = (SRC)[1];             \
   (DST)[2] = (SRC)[2];             \
   (DST)[3] = (SRC)[3];             \
} while (0)
#endif


#ifndef COPY_4FV
#define COPY_4FV( DST, SRC )  COPY_4V(DST, SRC)
#endif


#ifndef ASSIGN_4V
#define ASSIGN_4V( DST, V0, V1, V2, V3 ) \
do {                                     \
   (DST)[0] = (V0);                      \
   (DST)[1] = (V1);                      \
   (DST)[2] = (V2);                      \
   (DST)[3] = (V3);                      \
} while (0)
#endif


#ifdef __cplusplus
}
#endif

#endif /* U_MATH_H */