This repository has been archived by the owner on Feb 11, 2021. It is now read-only.
/
SDL_blit.h
550 lines (510 loc) · 15 KB
1
2
/*
SDL - Simple DirectMedia Layer
3
Copyright (C) 1997-2006 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
29
30
31
32
33
#ifdef __MMX__
#include <mmintrin.h>
#endif
#ifdef __SSE__
#include <xmmintrin.h>
#endif
34
35
36
#include "SDL_endian.h"
/* The structure passed to the low level blit functions */
37
38
39
40
41
42
43
44
45
46
47
48
49
50
typedef struct
{
Uint8 *s_pixels;
int s_width;
int s_height;
int s_skip;
Uint8 *d_pixels;
int d_width;
int d_height;
int d_skip;
void *aux_data;
SDL_PixelFormat *src;
Uint8 *table;
SDL_PixelFormat *dst;
51
52
53
} SDL_BlitInfo;
/* The type definition for the low level blit functions */
54
typedef void (*SDL_loblit) (SDL_BlitInfo * info);
55
56
/* This is the private info structure for software accelerated blits */
57
58
59
60
struct private_swaccel
{
SDL_loblit blit;
void *aux_data;
61
62
63
};
/* Blit mapping definition */
64
65
66
67
68
69
70
71
72
73
74
typedef struct SDL_BlitMap
{
SDL_Surface *dst;
int identity;
Uint8 *table;
SDL_blit sw_blit;
struct private_swaccel *sw_data;
/* the version count matches the destination; mismatch indicates
an invalid mapping */
unsigned int format_version;
75
76
} SDL_BlitMap;
77
78
79
80
81
82
83
84
85
86
87
#define SDL_BLIT_ANY 0x00000000
#define SDL_BLIT_MMX 0x00000001
#define SDL_BLIT_SSE 0x00000002
#define SDL_BLIT_ALTIVEC_PREFETCH 0x00000004
#define SDL_BLIT_ALTIVEC_NOPREFETCH 0x00000008
typedef struct SDL_BlitEntry
{
Uint32 features;
SDL_loblit blit;
} SDL_BlitEntry;
88
89
/* Functions found in SDL_blit.c */
90
extern int SDL_CalculateBlit(SDL_Surface * surface);
91
92
/* Functions found in SDL_blit_{0,1,N,A}.c */
93
94
95
96
extern SDL_loblit SDL_CalculateBlit0(SDL_Surface * surface, int complex);
extern SDL_loblit SDL_CalculateBlit1(SDL_Surface * surface, int complex);
extern SDL_loblit SDL_CalculateBlitN(SDL_Surface * surface, int complex);
extern SDL_loblit SDL_CalculateAlphaBlit(SDL_Surface * surface, int complex);
97
98
99
100
101
/*
* Useful macros for blitting routines
*/
102
103
104
105
106
107
108
109
#if defined(__GNUC__)
#define DECLARE_ALIGNED(t,v,a) t __attribute__((aligned(a))) v
#elif defined(_MSC_VER)
#define DECLARE_ALIGNED(t,v,a) t __declspec(align(a)) v
#else
#define DECLARE_ALIGNED(t,v,a) t v
#endif
110
111
112
113
114
115
#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? */
116
#define RGB_FROM_PIXEL(Pixel, fmt, r, g, b) \
117
{ \
118
119
120
r = (((Pixel&fmt->Rmask)>>fmt->Rshift)<<fmt->Rloss); \
g = (((Pixel&fmt->Gmask)>>fmt->Gshift)<<fmt->Gloss); \
b = (((Pixel&fmt->Bmask)>>fmt->Bshift)<<fmt->Bloss); \
121
}
122
#define RGB_FROM_RGB565(Pixel, r, g, b) \
123
{ \
124
125
126
r = (((Pixel&0xF800)>>11)<<3); \
g = (((Pixel&0x07E0)>>5)<<2); \
b = ((Pixel&0x001F)<<3); \
127
}
128
#define RGB_FROM_RGB555(Pixel, r, g, b) \
129
{ \
130
131
132
r = (((Pixel&0x7C00)>>10)<<3); \
g = (((Pixel&0x03E0)>>5)<<3); \
b = ((Pixel&0x001F)<<3); \
133
}
134
#define RGB_FROM_RGB888(Pixel, r, g, b) \
135
{ \
136
137
138
r = ((Pixel&0xFF0000)>>16); \
g = ((Pixel&0xFF00)>>8); \
b = (Pixel&0xFF); \
139
}
140
#define RETRIEVE_RGB_PIXEL(buf, bpp, Pixel) \
141
142
143
do { \
switch (bpp) { \
case 2: \
144
Pixel = *((Uint16 *)(buf)); \
145
146
147
148
149
break; \
\
case 3: { \
Uint8 *B = (Uint8 *)(buf); \
if(SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
150
Pixel = B[0] + (B[1] << 8) + (B[2] << 16); \
151
} else { \
152
Pixel = (B[0] << 16) + (B[1] << 8) + B[2]; \
153
154
155
156
157
} \
} \
break; \
\
case 4: \
158
Pixel = *((Uint32 *)(buf)); \
159
160
161
break; \
\
default: \
162
Pixel = 0; /* appease gcc */ \
163
164
165
166
break; \
} \
} while(0)
167
#define DISEMBLE_RGB(buf, bpp, fmt, Pixel, r, g, b) \
168
169
170
do { \
switch (bpp) { \
case 2: \
171
Pixel = *((Uint16 *)(buf)); \
172
173
174
175
176
break; \
\
case 3: { \
Uint8 *B = (Uint8 *)buf; \
if(SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
177
Pixel = B[0] + (B[1] << 8) + (B[2] << 16); \
178
} else { \
179
Pixel = (B[0] << 16) + (B[1] << 8) + B[2]; \
180
181
182
183
184
} \
} \
break; \
\
case 4: \
185
Pixel = *((Uint32 *)(buf)); \
186
187
188
break; \
\
default: \
189
Pixel = 0; /* prevent gcc from complaining */ \
190
191
break; \
} \
192
RGB_FROM_PIXEL(Pixel, fmt, r, g, b); \
193
194
195
} while(0)
/* Assemble R-G-B values into a specified pixel format and store them */
196
#define PIXEL_FROM_RGB(Pixel, fmt, r, g, b) \
197
{ \
198
Pixel = ((r>>fmt->Rloss)<<fmt->Rshift)| \
199
200
201
((g>>fmt->Gloss)<<fmt->Gshift)| \
((b>>fmt->Bloss)<<fmt->Bshift); \
}
202
#define RGB565_FROM_RGB(Pixel, r, g, b) \
203
{ \
204
Pixel = ((r>>3)<<11)|((g>>2)<<5)|(b>>3); \
205
}
206
#define RGB555_FROM_RGB(Pixel, r, g, b) \
207
{ \
208
Pixel = ((r>>3)<<10)|((g>>3)<<5)|(b>>3); \
209
}
210
#define RGB888_FROM_RGB(Pixel, r, g, b) \
211
{ \
212
Pixel = (r<<16)|(g<<8)|b; \
213
214
215
216
217
}
#define ASSEMBLE_RGB(buf, bpp, fmt, r, g, b) \
{ \
switch (bpp) { \
case 2: { \
218
Uint16 Pixel; \
219
\
220
221
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*((Uint16 *)(buf)) = Pixel; \
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
} \
break; \
\
case 3: { \
if(SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
*((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: { \
239
Uint32 Pixel; \
240
\
241
242
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*((Uint32 *)(buf)) = Pixel; \
243
244
245
246
247
248
249
250
251
} \
break; \
} \
}
#define ASSEMBLE_RGB_AMASK(buf, bpp, fmt, r, g, b, Amask) \
{ \
switch (bpp) { \
case 2: { \
Uint16 *bufp; \
252
Uint16 Pixel; \
253
254
\
bufp = (Uint16 *)buf; \
255
256
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*bufp = Pixel | (*bufp & Amask); \
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
} \
break; \
\
case 3: { \
if(SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
*((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; \
275
Uint32 Pixel; \
276
277
\
bufp = (Uint32 *)buf; \
278
279
PIXEL_FROM_RGB(Pixel, fmt, r, g, b); \
*bufp = Pixel | (*bufp & Amask); \
280
281
282
283
284
285
} \
break; \
} \
}
/* FIXME: Should we rescale alpha into 0..255 here? */
286
#define RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a) \
287
{ \
288
289
290
291
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; \
292
}
293
#define RGBA_FROM_8888(Pixel, fmt, r, g, b, a) \
294
{ \
295
296
297
298
r = (Pixel&fmt->Rmask)>>fmt->Rshift; \
g = (Pixel&fmt->Gmask)>>fmt->Gshift; \
b = (Pixel&fmt->Bmask)>>fmt->Bshift; \
a = (Pixel&fmt->Amask)>>fmt->Ashift; \
299
}
300
#define RGBA_FROM_RGBA8888(Pixel, r, g, b, a) \
301
{ \
302
303
304
305
r = (Pixel>>24); \
g = ((Pixel>>16)&0xFF); \
b = ((Pixel>>8)&0xFF); \
a = (Pixel&0xFF); \
306
}
307
#define RGBA_FROM_ARGB8888(Pixel, r, g, b, a) \
308
{ \
309
310
311
312
r = ((Pixel>>16)&0xFF); \
g = ((Pixel>>8)&0xFF); \
b = (Pixel&0xFF); \
a = (Pixel>>24); \
313
}
314
#define RGBA_FROM_ABGR8888(Pixel, r, g, b, a) \
315
{ \
316
317
318
319
r = (Pixel&0xFF); \
g = ((Pixel>>8)&0xFF); \
b = ((Pixel>>16)&0xFF); \
a = (Pixel>>24); \
320
}
321
#define DISEMBLE_RGBA(buf, bpp, fmt, Pixel, r, g, b, a) \
322
323
324
do { \
switch (bpp) { \
case 2: \
325
Pixel = *((Uint16 *)(buf)); \
326
327
328
329
330
break; \
\
case 3: {/* FIXME: broken code (no alpha) */ \
Uint8 *b = (Uint8 *)buf; \
if(SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
331
Pixel = b[0] + (b[1] << 8) + (b[2] << 16); \
332
} else { \
333
Pixel = (b[0] << 16) + (b[1] << 8) + b[2]; \
334
335
336
337
338
} \
} \
break; \
\
case 4: \
339
Pixel = *((Uint32 *)(buf)); \
340
341
342
break; \
\
default: \
343
Pixel = 0; /* stop gcc complaints */ \
344
345
break; \
} \
346
347
RGBA_FROM_PIXEL(Pixel, fmt, r, g, b, a); \
Pixel &= ~fmt->Amask; \
348
349
350
} while(0)
/* FIXME: this isn't correct, especially for Alpha (maximum != 255) */
351
#define PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a) \
352
{ \
353
Pixel = ((r>>fmt->Rloss)<<fmt->Rshift)| \
354
355
((g>>fmt->Gloss)<<fmt->Gshift)| \
((b>>fmt->Bloss)<<fmt->Bshift)| \
356
((a>>fmt->Aloss)<<fmt->Ashift); \
357
358
359
360
361
}
#define ASSEMBLE_RGBA(buf, bpp, fmt, r, g, b, a) \
{ \
switch (bpp) { \
case 2: { \
362
Uint16 Pixel; \
363
\
364
365
PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a); \
*((Uint16 *)(buf)) = Pixel; \
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
} \
break; \
\
case 3: { /* FIXME: broken code (no alpha) */ \
if(SDL_BYTEORDER == SDL_LIL_ENDIAN) { \
*((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: { \
383
Uint32 Pixel; \
384
\
385
386
PIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a); \
*((Uint32 *)(buf)) = Pixel; \
387
388
389
390
391
} \
break; \
} \
}
392
/* Blend the RGB values of two Pixels based on a source alpha value */
393
394
395
396
397
398
399
#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)
400
/* Blend the RGB values of two Pixels based on a source alpha value */
401
402
403
404
405
406
407
408
409
410
411
412
413
#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)
414
/* This is a very useful loop for optimizing blitters */
415
416
417
#if defined(_MSC_VER) && (_MSC_VER == 1300)
/* There's a bug in the Visual C++ 7 optimizer when compiling this code */
#else
418
#define USE_DUFFS_LOOP
419
#endif
420
421
422
423
424
#ifdef USE_DUFFS_LOOP
/* 8-times unrolled loop */
#define DUFFS_LOOP8(pixel_copy_increment, width) \
{ int n = (width+7)/8; \
425
switch (width & 7) { \
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
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; \
441
switch (width & 3) { \
442
443
444
445
446
447
448
449
case 0: do { pixel_copy_increment; \
case 3: pixel_copy_increment; \
case 2: pixel_copy_increment; \
case 1: pixel_copy_increment; \
} while ( --n > 0 ); \
} \
}
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
/* 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 ); \
} \
} \
}
491
492
493
494
495
496
/* Use the 8-times version of the loop by default */
#define DUFFS_LOOP(pixel_copy_increment, width) \
DUFFS_LOOP8(pixel_copy_increment, width)
#else
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
/* 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; \
} \
}
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
/* 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 */
550
/* vi: set ts=4 sw=4 expandtab: */