This repository has been archived by the owner on Feb 11, 2021. It is now read-only.
/
SDL_blit.h
578 lines (538 loc) · 15.9 KB
1
2
/*
SDL - Simple DirectMedia Layer
3
Copyright (C) 1997-2009 Sam Lantinga
4
5
This library is free software; you can redistribute it and/or
6
modify it under the terms of the GNU Lesser General Public
7
License as published by the Free Software Foundation; either
8
version 2.1 of the License, or (at your option) any later version.
9
10
11
12
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13
Lesser General Public License for more details.
14
15
16
17
You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free Software
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
18
19
Sam Lantinga
20
slouken@libsdl.org
21
*/
22
#include "SDL_config.h"
23
24
25
26
#ifndef _SDL_blit_h
#define _SDL_blit_h
27
28
#ifdef __MMX__
#include <mmintrin.h>
29
30
31
#endif
#ifdef __3dNOW__
#include <mm3dnow.h>
32
33
34
35
#endif
#ifdef __SSE__
#include <xmmintrin.h>
#endif
36
37
38
#ifdef __SSE2__
#include <emmintrin.h>
#endif
39
40
#include "SDL_cpuinfo.h"
41
42
#include "SDL_endian.h"
43
/* SDL blit copy flags */
44
45
46
47
48
49
50
51
52
53
54
#define SDL_COPY_MODULATE_COLOR 0x00000001
#define SDL_COPY_MODULATE_ALPHA 0x00000002
#define SDL_COPY_MASK 0x00000010
#define SDL_COPY_BLEND 0x00000020
#define SDL_COPY_ADD 0x00000040
#define SDL_COPY_MOD 0x00000080
#define SDL_COPY_COLORKEY 0x00000100
#define SDL_COPY_NEAREST 0x00000200
#define SDL_COPY_RLE_DESIRED 0x00001000
#define SDL_COPY_RLE_COLORKEY 0x00002000
#define SDL_COPY_RLE_ALPHAKEY 0x00004000
55
#define SDL_COPY_RLE_MASK (SDL_COPY_RLE_DESIRED|SDL_COPY_RLE_COLORKEY|SDL_COPY_RLE_ALPHAKEY)
56
57
/* SDL blit CPU flags */
58
59
60
61
62
63
64
#define SDL_CPU_ANY 0x00000000
#define SDL_CPU_MMX 0x00000001
#define SDL_CPU_3DNOW 0x00000002
#define SDL_CPU_SSE 0x00000004
#define SDL_CPU_SSE2 0x00000008
#define SDL_CPU_ALTIVEC_PREFETCH 0x00000010
#define SDL_CPU_ALTIVEC_NOPREFETCH 0x00000020
65
66
67
typedef struct
{
68
69
70
Uint8 *src;
int src_w, src_h;
int src_pitch;
71
int src_skip;
72
73
74
Uint8 *dst;
int dst_w, dst_h;
int dst_pitch;
75
int dst_skip;
76
77
SDL_PixelFormat *src_fmt;
SDL_PixelFormat *dst_fmt;
78
Uint8 *table;
79
80
81
int flags;
Uint32 colorkey;
Uint8 r, g, b, a;
82
83
} SDL_BlitInfo;
84
typedef void (SDLCALL * SDL_BlitFunc) (SDL_BlitInfo * info);
85
86
87
typedef struct
{
88
89
90
91
92
93
Uint32 src_format;
Uint32 dst_format;
int flags;
int cpu;
SDL_BlitFunc func;
} SDL_BlitFuncEntry;
94
95
/* Blit mapping definition */
96
97
98
99
typedef struct SDL_BlitMap
{
SDL_Surface *dst;
int identity;
100
101
SDL_blit blit;
void *data;
102
SDL_BlitInfo info;
103
104
105
106
/* the version count matches the destination; mismatch indicates
an invalid mapping */
unsigned int format_version;
107
108
109
} SDL_BlitMap;
/* Functions found in SDL_blit.c */
110
extern int SDL_CalculateBlit(SDL_Surface * surface);
111
112
113
114
115
116
/* Functions found in SDL_blit_*.c */
extern SDL_BlitFunc SDL_CalculateBlit0(SDL_Surface * surface);
extern SDL_BlitFunc SDL_CalculateBlit1(SDL_Surface * surface);
extern SDL_BlitFunc SDL_CalculateBlitN(SDL_Surface * surface);
extern SDL_BlitFunc SDL_CalculateBlitA(SDL_Surface * surface);
117
118
119
120
121
/*
* Useful macros for blitting routines
*/
122
123
124
#if defined(__GNUC__)
#define DECLARE_ALIGNED(t,v,a) t __attribute__((aligned(a))) v
#elif defined(_MSC_VER)
125
#define DECLARE_ALIGNED(t,v,a) __declspec(align(a)) t v
126
127
128
129
#else
#define DECLARE_ALIGNED(t,v,a) t v
#endif
130
131
132
133
134
135
#define FORMAT_EQUAL(A, B) \
((A)->BitsPerPixel == (B)->BitsPerPixel \
&& ((A)->Rmask == (B)->Rmask) && ((A)->Amask == (B)->Amask))
/* Load pixel of the specified format from a buffer and get its R-G-B values */
/* FIXME: rescale values to 0..255 here? */
136
#define RGB_FROM_PIXEL(Pixel, fmt, r, g, b) \
137
{ \
138
139
140
r = (((Pixel&fmt->Rmask)>>fmt->Rshift)<<fmt->Rloss); \
g = (((Pixel&fmt->Gmask)>>fmt->Gshift)<<fmt->Gloss); \
b = (((Pixel&fmt->Bmask)>>fmt->Bshift)<<fmt->Bloss); \
141
}
142
#define RGB_FROM_RGB565(Pixel, r, g, b) \
143
{ \
144
145
146
r = (((Pixel&0xF800)>>11)<<3); \
g = (((Pixel&0x07E0)>>5)<<2); \
b = ((Pixel&0x001F)<<3); \
147
}
148
#define RGB_FROM_RGB555(Pixel, r, g, b) \
149
{ \
150
151
152
r = (((Pixel&0x7C00)>>10)<<3); \
g = (((Pixel&0x03E0)>>5)<<3); \
b = ((Pixel&0x001F)<<3); \
153
}
154
#define RGB_FROM_RGB888(Pixel, r, g, b) \
155
{ \
156
157
158
r = ((Pixel&0xFF0000)>>16); \
g = ((Pixel&0xFF00)>>8); \
b = (Pixel&0xFF); \
159
}
160
#define RETRIEVE_RGB_PIXEL(buf, bpp, Pixel) \
161
162
163
do { \
switch (bpp) { \
case 2: \
164
Pixel = *((Uint16 *)(buf)); \
165
166
167
168
break; \
\
case 3: { \
Uint8 *B = (Uint8 *)(buf); \
169
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
170
Pixel = B[0] + (B[1] << 8) + (B[2] << 16); \
171
} else { \
172
Pixel = (B[0] << 16) + (B[1] << 8) + B[2]; \
173
174
175
176
177
} \
} \
break; \
\
case 4: \
178
Pixel = *((Uint32 *)(buf)); \
179
180
181
break; \
\
default: \
182
Pixel; /* stop gcc complaints */ \
183
184
break; \
} \
185
} while (0)
186
187
#define DISEMBLE_RGB(buf, bpp, fmt, Pixel, r, g, b) \
188
189
190
do { \
switch (bpp) { \
case 2: \
191
Pixel = *((Uint16 *)(buf)); \
192
RGB_FROM_PIXEL(Pixel, fmt, r, g, b); \
193
194
break; \
\
195
196
197
198
199
case 3: { \
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
r = *((buf)+fmt->Rshift/8); \
g = *((buf)+fmt->Gshift/8); \
b = *((buf)+fmt->Bshift/8); \
200
} else { \
201
202
203
r = *((buf)+2-fmt->Rshift/8); \
g = *((buf)+2-fmt->Gshift/8); \
b = *((buf)+2-fmt->Bshift/8); \
204
205
206
207
208
} \
} \
break; \
\
case 4: \
209
Pixel = *((Uint32 *)(buf)); \
210
RGB_FROM_PIXEL(Pixel, fmt, r, g, b); \
211
212
break; \
\
213
214
default: \
Pixel; /* stop gcc complaints */ \
215
216
break; \
} \
217
} while (0)
218
219
/* Assemble R-G-B values into a specified pixel format and store them */
220
#define PIXEL_FROM_RGB(Pixel, fmt, r, g, b) \
221
{ \
222
Pixel = ((r>>fmt->Rloss)<<fmt->Rshift)| \
223
224
225
((g>>fmt->Gloss)<<fmt->Gshift)| \
((b>>fmt->Bloss)<<fmt->Bshift); \
}
226
#define RGB565_FROM_RGB(Pixel, r, g, b) \
227
{ \
228
Pixel = ((r>>3)<<11)|((g>>2)<<5)|(b>>3); \
229
}
230
#define RGB555_FROM_RGB(Pixel, r, g, b) \
231
{ \
232
Pixel = ((r>>3)<<10)|((g>>3)<<5)|(b>>3); \
233
}
234
#define RGB888_FROM_RGB(Pixel, r, g, b) \
235
{ \
236
Pixel = (r<<16)|(g<<8)|b; \
237
238
239
240
241
}
#define ASSEMBLE_RGB(buf, bpp, fmt, r, g, b) \
{ \
switch (bpp) { \
case 2: { \
242
Uint16 Pixel; \
243
\
244
245
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*((Uint16 *)(buf)) = Pixel; \
246
247
248
249
} \
break; \
\
case 3: { \
250
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
251
252
253
254
255
256
257
258
259
260
261
262
*((buf)+fmt->Rshift/8) = r; \
*((buf)+fmt->Gshift/8) = g; \
*((buf)+fmt->Bshift/8) = b; \
} else { \
*((buf)+2-fmt->Rshift/8) = r; \
*((buf)+2-fmt->Gshift/8) = g; \
*((buf)+2-fmt->Bshift/8) = b; \
} \
} \
break; \
\
case 4: { \
263
Uint32 Pixel; \
264
\
265
266
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*((Uint32 *)(buf)) = Pixel; \
267
268
269
270
271
272
273
274
275
} \
break; \
} \
}
#define ASSEMBLE_RGB_AMASK(buf, bpp, fmt, r, g, b, Amask) \
{ \
switch (bpp) { \
case 2: { \
Uint16 *bufp; \
276
Uint16 Pixel; \
277
278
\
bufp = (Uint16 *)buf; \
279
280
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*bufp = Pixel | (*bufp & Amask); \
281
282
283
284
} \
break; \
\
case 3: { \
285
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
286
287
288
289
290
291
292
293
294
295
296
297
298
*((buf)+fmt->Rshift/8) = r; \
*((buf)+fmt->Gshift/8) = g; \
*((buf)+fmt->Bshift/8) = b; \
} else { \
*((buf)+2-fmt->Rshift/8) = r; \
*((buf)+2-fmt->Gshift/8) = g; \
*((buf)+2-fmt->Bshift/8) = b; \
} \
} \
break; \
\
case 4: { \
Uint32 *bufp; \
299
Uint32 Pixel; \
300
301
\
bufp = (Uint32 *)buf; \
302
303
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*bufp = Pixel | (*bufp & Amask); \
304
305
306
307
308
309
} \
break; \
} \
}
/* FIXME: Should we rescale alpha into 0..255 here? */
310
#define RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a) \
311
{ \
312
313
314
315
r = ((Pixel&fmt->Rmask)>>fmt->Rshift)<<fmt->Rloss; \
g = ((Pixel&fmt->Gmask)>>fmt->Gshift)<<fmt->Gloss; \
b = ((Pixel&fmt->Bmask)>>fmt->Bshift)<<fmt->Bloss; \
a = ((Pixel&fmt->Amask)>>fmt->Ashift)<<fmt->Aloss; \
316
}
317
#define RGBA_FROM_8888(Pixel, fmt, r, g, b, a) \
318
{ \
319
320
321
322
r = (Pixel&fmt->Rmask)>>fmt->Rshift; \
g = (Pixel&fmt->Gmask)>>fmt->Gshift; \
b = (Pixel&fmt->Bmask)>>fmt->Bshift; \
a = (Pixel&fmt->Amask)>>fmt->Ashift; \
323
}
324
#define RGBA_FROM_RGBA8888(Pixel, r, g, b, a) \
325
{ \
326
327
328
329
r = (Pixel>>24); \
g = ((Pixel>>16)&0xFF); \
b = ((Pixel>>8)&0xFF); \
a = (Pixel&0xFF); \
330
}
331
#define RGBA_FROM_ARGB8888(Pixel, r, g, b, a) \
332
{ \
333
334
335
336
r = ((Pixel>>16)&0xFF); \
g = ((Pixel>>8)&0xFF); \
b = (Pixel&0xFF); \
a = (Pixel>>24); \
337
}
338
#define RGBA_FROM_ABGR8888(Pixel, r, g, b, a) \
339
{ \
340
341
342
343
r = (Pixel&0xFF); \
g = ((Pixel>>8)&0xFF); \
b = ((Pixel>>16)&0xFF); \
a = (Pixel>>24); \
344
}
345
#define DISEMBLE_RGBA(buf, bpp, fmt, Pixel, r, g, b, a) \
346
347
348
do { \
switch (bpp) { \
case 2: \
349
Pixel = *((Uint16 *)(buf)); \
350
RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a); \
351
352
break; \
\
353
354
355
356
357
case 3: { \
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
r = *((buf)+fmt->Rshift/8); \
g = *((buf)+fmt->Gshift/8); \
b = *((buf)+fmt->Bshift/8); \
358
} else { \
359
360
361
r = *((buf)+2-fmt->Rshift/8); \
g = *((buf)+2-fmt->Gshift/8); \
b = *((buf)+2-fmt->Bshift/8); \
362
} \
363
a = 0xFF; \
364
365
366
367
} \
break; \
\
case 4: \
368
Pixel = *((Uint32 *)(buf)); \
369
RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a); \
370
371
372
break; \
\
default: \
373
Pixel; /* stop gcc complaints */ \
374
375
break; \
} \
376
} while (0)
377
378
/* FIXME: this isn't correct, especially for Alpha (maximum != 255) */
379
#define PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a) \
380
{ \
381
Pixel = ((r>>fmt->Rloss)<<fmt->Rshift)| \
382
383
((g>>fmt->Gloss)<<fmt->Gshift)| \
((b>>fmt->Bloss)<<fmt->Bshift)| \
384
((a>>fmt->Aloss)<<fmt->Ashift); \
385
386
387
388
389
}
#define ASSEMBLE_RGBA(buf, bpp, fmt, r, g, b, a) \
{ \
switch (bpp) { \
case 2: { \
390
Uint16 Pixel; \
391
\
392
393
PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a); \
*((Uint16 *)(buf)) = Pixel; \
394
395
396
} \
break; \
\
397
398
case 3: { \
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
399
400
401
402
403
404
405
406
407
408
409
410
*((buf)+fmt->Rshift/8) = r; \
*((buf)+fmt->Gshift/8) = g; \
*((buf)+fmt->Bshift/8) = b; \
} else { \
*((buf)+2-fmt->Rshift/8) = r; \
*((buf)+2-fmt->Gshift/8) = g; \
*((buf)+2-fmt->Bshift/8) = b; \
} \
} \
break; \
\
case 4: { \
411
Uint32 Pixel; \
412
\
413
414
PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a); \
*((Uint32 *)(buf)) = Pixel; \
415
416
417
418
419
} \
break; \
} \
}
420
/* Blend the RGB values of two Pixels based on a source alpha value */
421
422
423
424
425
426
427
#define ALPHA_BLEND(sR, sG, sB, A, dR, dG, dB) \
do { \
dR = (((sR-dR)*(A))>>8)+dR; \
dG = (((sG-dG)*(A))>>8)+dG; \
dB = (((sB-dB)*(A))>>8)+dB; \
} while(0)
428
/* Blend the RGB values of two Pixels based on a source alpha value */
429
430
431
432
433
434
435
436
437
438
439
440
441
#define ACCURATE_ALPHA_BLEND(sR, sG, sB, sA, dR, dG, dB) \
do { \
unsigned tR, tG, tB, tA; \
tA = 255 - sA; \
tR = 1 + (sR * sA) + (dR * tA); \
dR = (tR + (tR >> 8)) >> 8; \
tG = 1 + (sG * sA) + (dG * tA); \
dG = (tG + (tG >> 8)) >> 8; \
tB = 1 + (sB * sA) + (dB * tA); \
dB = (tB + (tB >> 8)) >> 8; \
} while(0)
442
/* This is a very useful loop for optimizing blitters */
443
444
445
#if defined(_MSC_VER) && (_MSC_VER == 1300)
/* There's a bug in the Visual C++ 7 optimizer when compiling this code */
#else
446
#define USE_DUFFS_LOOP
447
#endif
448
449
450
451
452
#ifdef USE_DUFFS_LOOP
/* 8-times unrolled loop */
#define DUFFS_LOOP8(pixel_copy_increment, width) \
{ int n = (width+7)/8; \
453
switch (width & 7) { \
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
case 0: do { pixel_copy_increment; \
case 7: pixel_copy_increment; \
case 6: pixel_copy_increment; \
case 5: pixel_copy_increment; \
case 4: pixel_copy_increment; \
case 3: pixel_copy_increment; \
case 2: pixel_copy_increment; \
case 1: pixel_copy_increment; \
} while ( --n > 0 ); \
} \
}
/* 4-times unrolled loop */
#define DUFFS_LOOP4(pixel_copy_increment, width) \
{ int n = (width+3)/4; \
469
switch (width & 3) { \
470
471
472
473
474
475
476
477
case 0: do { pixel_copy_increment; \
case 3: pixel_copy_increment; \
case 2: pixel_copy_increment; \
case 1: pixel_copy_increment; \
} while ( --n > 0 ); \
} \
}
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
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
/* 2 - times unrolled loop */
#define DUFFS_LOOP_DOUBLE2(pixel_copy_increment, \
double_pixel_copy_increment, width) \
{ int n, w = width; \
if( w & 1 ) { \
pixel_copy_increment; \
w--; \
} \
if ( w > 0 ) { \
n = ( w + 2) / 4; \
switch( w & 2 ) { \
case 0: do { double_pixel_copy_increment; \
case 2: double_pixel_copy_increment; \
} while ( --n > 0 ); \
} \
} \
}
/* 2 - times unrolled loop 4 pixels */
#define DUFFS_LOOP_QUATRO2(pixel_copy_increment, \
double_pixel_copy_increment, \
quatro_pixel_copy_increment, width) \
{ int n, w = width; \
if(w & 1) { \
pixel_copy_increment; \
w--; \
} \
if(w & 2) { \
double_pixel_copy_increment; \
w -= 2; \
} \
if ( w > 0 ) { \
n = ( w + 7 ) / 8; \
switch( w & 4 ) { \
case 0: do { quatro_pixel_copy_increment; \
case 4: quatro_pixel_copy_increment; \
} while ( --n > 0 ); \
} \
} \
}
519
520
521
522
523
524
/* Use the 8-times version of the loop by default */
#define DUFFS_LOOP(pixel_copy_increment, width) \
DUFFS_LOOP8(pixel_copy_increment, width)
#else
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
/* Don't use Duff's device to unroll loops */
#define DUFFS_LOOP_DOUBLE2(pixel_copy_increment, \
double_pixel_copy_increment, width) \
{ int n = width; \
if( n & 1 ) { \
pixel_copy_increment; \
n--; \
} \
n=n>>1; \
for(; n > 0; --n) { \
double_pixel_copy_increment; \
} \
}
/* Don't use Duff's device to unroll loops */
#define DUFFS_LOOP_QUATRO2(pixel_copy_increment, \
double_pixel_copy_increment, \
quatro_pixel_copy_increment, width) \
{ int n = width; \
if(n & 1) { \
pixel_copy_increment; \
n--; \
} \
if(n & 2) { \
double_pixel_copy_increment; \
n -= 2; \
} \
n=n>>2; \
for(; n > 0; --n) { \
quatro_pixel_copy_increment; \
} \
}
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
/* Don't use Duff's device to unroll loops */
#define DUFFS_LOOP(pixel_copy_increment, width) \
{ int n; \
for ( n=width; n > 0; --n ) { \
pixel_copy_increment; \
} \
}
#define DUFFS_LOOP8(pixel_copy_increment, width) \
DUFFS_LOOP(pixel_copy_increment, width)
#define DUFFS_LOOP4(pixel_copy_increment, width) \
DUFFS_LOOP(pixel_copy_increment, width)
#endif /* USE_DUFFS_LOOP */
/* Prevent Visual C++ 6.0 from printing out stupid warnings */
#if defined(_MSC_VER) && (_MSC_VER >= 600)
#pragma warning(disable: 4550)
#endif
#endif /* _SDL_blit_h */
578
/* vi: set ts=4 sw=4 expandtab: */