This repository has been archived by the owner on Feb 11, 2021. It is now read-only.
/
SDL_blit.h
580 lines (539 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
#include "SDL_endian.h"
42
#include "SDL_video.h"
43
44
/* SDL blit copy flags */
45
46
47
48
49
50
51
52
53
54
55
#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
56
#define SDL_COPY_RLE_MASK (SDL_COPY_RLE_DESIRED|SDL_COPY_RLE_COLORKEY|SDL_COPY_RLE_ALPHAKEY)
57
58
/* SDL blit CPU flags */
59
60
61
62
63
64
65
#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
66
67
68
typedef struct
{
69
70
71
Uint8 *src;
int src_w, src_h;
int src_pitch;
72
int src_skip;
73
74
75
Uint8 *dst;
int dst_w, dst_h;
int dst_pitch;
76
int dst_skip;
77
78
SDL_PixelFormat *src_fmt;
SDL_PixelFormat *dst_fmt;
79
Uint8 *table;
80
81
82
int flags;
Uint32 colorkey;
Uint8 r, g, b, a;
83
84
} SDL_BlitInfo;
85
typedef void (SDLCALL * SDL_BlitFunc) (SDL_BlitInfo * info);
86
87
88
typedef struct
{
89
90
91
92
93
94
Uint32 src_format;
Uint32 dst_format;
int flags;
int cpu;
SDL_BlitFunc func;
} SDL_BlitFuncEntry;
95
96
/* Blit mapping definition */
97
98
99
100
typedef struct SDL_BlitMap
{
SDL_Surface *dst;
int identity;
101
102
SDL_blit blit;
void *data;
103
SDL_BlitInfo info;
104
105
106
107
/* the version count matches the destination; mismatch indicates
an invalid mapping */
unsigned int format_version;
108
109
110
} SDL_BlitMap;
/* Functions found in SDL_blit.c */
111
extern int SDL_CalculateBlit(SDL_Surface * surface);
112
113
114
115
116
117
/* 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);
118
119
120
121
122
/*
* Useful macros for blitting routines
*/
123
124
125
#if defined(__GNUC__)
#define DECLARE_ALIGNED(t,v,a) t __attribute__((aligned(a))) v
#elif defined(_MSC_VER)
126
#define DECLARE_ALIGNED(t,v,a) __declspec(align(a)) t v
127
128
129
130
#else
#define DECLARE_ALIGNED(t,v,a) t v
#endif
131
132
133
134
135
136
#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? */
137
#define RGB_FROM_PIXEL(Pixel, fmt, r, g, b) \
138
{ \
139
140
141
r = (((Pixel&fmt->Rmask)>>fmt->Rshift)<<fmt->Rloss); \
g = (((Pixel&fmt->Gmask)>>fmt->Gshift)<<fmt->Gloss); \
b = (((Pixel&fmt->Bmask)>>fmt->Bshift)<<fmt->Bloss); \
142
}
143
#define RGB_FROM_RGB565(Pixel, r, g, b) \
144
{ \
145
146
147
r = (((Pixel&0xF800)>>11)<<3); \
g = (((Pixel&0x07E0)>>5)<<2); \
b = ((Pixel&0x001F)<<3); \
148
}
149
#define RGB_FROM_RGB555(Pixel, r, g, b) \
150
{ \
151
152
153
r = (((Pixel&0x7C00)>>10)<<3); \
g = (((Pixel&0x03E0)>>5)<<3); \
b = ((Pixel&0x001F)<<3); \
154
}
155
#define RGB_FROM_RGB888(Pixel, r, g, b) \
156
{ \
157
158
159
r = ((Pixel&0xFF0000)>>16); \
g = ((Pixel&0xFF00)>>8); \
b = (Pixel&0xFF); \
160
}
161
#define RETRIEVE_RGB_PIXEL(buf, bpp, Pixel) \
162
163
164
do { \
switch (bpp) { \
case 2: \
165
Pixel = *((Uint16 *)(buf)); \
166
167
168
169
break; \
\
case 3: { \
Uint8 *B = (Uint8 *)(buf); \
170
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
171
Pixel = B[0] + (B[1] << 8) + (B[2] << 16); \
172
} else { \
173
Pixel = (B[0] << 16) + (B[1] << 8) + B[2]; \
174
175
176
177
178
} \
} \
break; \
\
case 4: \
179
Pixel = *((Uint32 *)(buf)); \
180
181
182
break; \
\
default: \
183
Pixel; /* stop gcc complaints */ \
184
185
break; \
} \
186
} while (0)
187
188
#define DISEMBLE_RGB(buf, bpp, fmt, Pixel, r, g, b) \
189
190
191
do { \
switch (bpp) { \
case 2: \
192
Pixel = *((Uint16 *)(buf)); \
193
RGB_FROM_PIXEL(Pixel, fmt, r, g, b); \
194
195
break; \
\
196
197
198
199
200
case 3: { \
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
r = *((buf)+fmt->Rshift/8); \
g = *((buf)+fmt->Gshift/8); \
b = *((buf)+fmt->Bshift/8); \
201
} else { \
202
203
204
r = *((buf)+2-fmt->Rshift/8); \
g = *((buf)+2-fmt->Gshift/8); \
b = *((buf)+2-fmt->Bshift/8); \
205
206
207
208
209
} \
} \
break; \
\
case 4: \
210
Pixel = *((Uint32 *)(buf)); \
211
RGB_FROM_PIXEL(Pixel, fmt, r, g, b); \
212
213
break; \
\
214
215
default: \
Pixel; /* stop gcc complaints */ \
216
217
break; \
} \
218
} while (0)
219
220
/* Assemble R-G-B values into a specified pixel format and store them */
221
#define PIXEL_FROM_RGB(Pixel, fmt, r, g, b) \
222
{ \
223
Pixel = ((r>>fmt->Rloss)<<fmt->Rshift)| \
224
225
226
((g>>fmt->Gloss)<<fmt->Gshift)| \
((b>>fmt->Bloss)<<fmt->Bshift); \
}
227
#define RGB565_FROM_RGB(Pixel, r, g, b) \
228
{ \
229
Pixel = ((r>>3)<<11)|((g>>2)<<5)|(b>>3); \
230
}
231
#define RGB555_FROM_RGB(Pixel, r, g, b) \
232
{ \
233
Pixel = ((r>>3)<<10)|((g>>3)<<5)|(b>>3); \
234
}
235
#define RGB888_FROM_RGB(Pixel, r, g, b) \
236
{ \
237
Pixel = (r<<16)|(g<<8)|b; \
238
239
240
241
242
}
#define ASSEMBLE_RGB(buf, bpp, fmt, r, g, b) \
{ \
switch (bpp) { \
case 2: { \
243
Uint16 Pixel; \
244
\
245
246
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*((Uint16 *)(buf)) = Pixel; \
247
248
249
250
} \
break; \
\
case 3: { \
251
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
252
253
254
255
256
257
258
259
260
261
262
263
*((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: { \
264
Uint32 Pixel; \
265
\
266
267
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*((Uint32 *)(buf)) = Pixel; \
268
269
270
271
272
273
274
275
276
} \
break; \
} \
}
#define ASSEMBLE_RGB_AMASK(buf, bpp, fmt, r, g, b, Amask) \
{ \
switch (bpp) { \
case 2: { \
Uint16 *bufp; \
277
Uint16 Pixel; \
278
279
\
bufp = (Uint16 *)buf; \
280
281
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*bufp = Pixel | (*bufp & Amask); \
282
283
284
285
} \
break; \
\
case 3: { \
286
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
287
288
289
290
291
292
293
294
295
296
297
298
299
*((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; \
300
Uint32 Pixel; \
301
302
\
bufp = (Uint32 *)buf; \
303
304
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*bufp = Pixel | (*bufp & Amask); \
305
306
307
308
309
310
} \
break; \
} \
}
/* FIXME: Should we rescale alpha into 0..255 here? */
311
#define RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a) \
312
{ \
313
314
315
316
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; \
317
}
318
#define RGBA_FROM_8888(Pixel, fmt, r, g, b, a) \
319
{ \
320
321
322
323
r = (Pixel&fmt->Rmask)>>fmt->Rshift; \
g = (Pixel&fmt->Gmask)>>fmt->Gshift; \
b = (Pixel&fmt->Bmask)>>fmt->Bshift; \
a = (Pixel&fmt->Amask)>>fmt->Ashift; \
324
}
325
#define RGBA_FROM_RGBA8888(Pixel, r, g, b, a) \
326
{ \
327
328
329
330
r = (Pixel>>24); \
g = ((Pixel>>16)&0xFF); \
b = ((Pixel>>8)&0xFF); \
a = (Pixel&0xFF); \
331
}
332
#define RGBA_FROM_ARGB8888(Pixel, r, g, b, a) \
333
{ \
334
335
336
337
r = ((Pixel>>16)&0xFF); \
g = ((Pixel>>8)&0xFF); \
b = (Pixel&0xFF); \
a = (Pixel>>24); \
338
}
339
#define RGBA_FROM_ABGR8888(Pixel, r, g, b, a) \
340
{ \
341
342
343
344
r = (Pixel&0xFF); \
g = ((Pixel>>8)&0xFF); \
b = ((Pixel>>16)&0xFF); \
a = (Pixel>>24); \
345
}
346
#define DISEMBLE_RGBA(buf, bpp, fmt, Pixel, r, g, b, a) \
347
348
349
do { \
switch (bpp) { \
case 2: \
350
Pixel = *((Uint16 *)(buf)); \
351
RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a); \
352
353
break; \
\
354
355
356
357
358
case 3: { \
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
r = *((buf)+fmt->Rshift/8); \
g = *((buf)+fmt->Gshift/8); \
b = *((buf)+fmt->Bshift/8); \
359
} else { \
360
361
362
r = *((buf)+2-fmt->Rshift/8); \
g = *((buf)+2-fmt->Gshift/8); \
b = *((buf)+2-fmt->Bshift/8); \
363
} \
364
a = 0xFF; \
365
366
367
368
} \
break; \
\
case 4: \
369
Pixel = *((Uint32 *)(buf)); \
370
RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a); \
371
372
373
break; \
\
default: \
374
Pixel; /* stop gcc complaints */ \
375
376
break; \
} \
377
} while (0)
378
379
/* FIXME: this isn't correct, especially for Alpha (maximum != 255) */
380
#define PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a) \
381
{ \
382
Pixel = ((r>>fmt->Rloss)<<fmt->Rshift)| \
383
384
((g>>fmt->Gloss)<<fmt->Gshift)| \
((b>>fmt->Bloss)<<fmt->Bshift)| \
385
((a>>fmt->Aloss)<<fmt->Ashift); \
386
387
388
389
390
}
#define ASSEMBLE_RGBA(buf, bpp, fmt, r, g, b, a) \
{ \
switch (bpp) { \
case 2: { \
391
Uint16 Pixel; \
392
\
393
394
PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a); \
*((Uint16 *)(buf)) = Pixel; \
395
396
397
} \
break; \
\
398
399
case 3: { \
if (SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
400
401
402
403
404
405
406
407
408
409
410
411
*((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: { \
412
Uint32 Pixel; \
413
\
414
415
PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a); \
*((Uint32 *)(buf)) = Pixel; \
416
417
418
419
420
} \
break; \
} \
}
421
/* Blend the RGB values of two Pixels based on a source alpha value */
422
423
424
425
426
427
428
#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)
429
/* Blend the RGB values of two Pixels based on a source alpha value */
430
431
432
433
434
435
436
437
438
439
440
441
442
#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)
443
/* This is a very useful loop for optimizing blitters */
444
445
446
#if defined(_MSC_VER) && (_MSC_VER == 1300)
/* There's a bug in the Visual C++ 7 optimizer when compiling this code */
#else
447
#define USE_DUFFS_LOOP
448
#endif
449
450
451
452
453
#ifdef USE_DUFFS_LOOP
/* 8-times unrolled loop */
#define DUFFS_LOOP8(pixel_copy_increment, width) \
{ int n = (width+7)/8; \
454
switch (width & 7) { \
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
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; \
470
switch (width & 3) { \
471
472
473
474
475
476
477
478
case 0: do { pixel_copy_increment; \
case 3: pixel_copy_increment; \
case 2: pixel_copy_increment; \
case 1: pixel_copy_increment; \
} while ( --n > 0 ); \
} \
}
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
519
/* 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 ); \
} \
} \
}
520
521
522
523
524
525
/* Use the 8-times version of the loop by default */
#define DUFFS_LOOP(pixel_copy_increment, width) \
DUFFS_LOOP8(pixel_copy_increment, width)
#else
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
558
/* 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; \
} \
}
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
/* 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 */
579
580
/* vi: set ts=4 sw=4 expandtab: */