This repository has been archived by the owner on Feb 11, 2021. It is now read-only.
/
SDL_yuv_sw.c
1322 lines (1168 loc) · 41 KB
1
2
/*
SDL - Simple DirectMedia Layer
3
Copyright (C) 1997-2010 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
/* This is the software implementation of the YUV texture support */
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
/* This code was derived from code carrying the following copyright notices:
* Copyright (c) 1995 The Regents of the University of California.
* All rights reserved.
*
* Permission to use, copy, modify, and distribute this software and its
* documentation for any purpose, without fee, and without written agreement is
* hereby granted, provided that the above copyright notice and the following
* two paragraphs appear in all copies of this software.
*
* IN NO EVENT SHALL THE UNIVERSITY OF CALIFORNIA BE LIABLE TO ANY PARTY FOR
* DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT
* OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN IF THE UNIVERSITY OF
* CALIFORNIA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* THE UNIVERSITY OF CALIFORNIA SPECIFICALLY DISCLAIMS ANY WARRANTIES,
* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
* AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS
* ON AN "AS IS" BASIS, AND THE UNIVERSITY OF CALIFORNIA HAS NO OBLIGATION TO
* PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
* Copyright (c) 1995 Erik Corry
* All rights reserved.
*
* Permission to use, copy, modify, and distribute this software and its
* documentation for any purpose, without fee, and without written agreement is
* hereby granted, provided that the above copyright notice and the following
* two paragraphs appear in all copies of this software.
*
* IN NO EVENT SHALL ERIK CORRY BE LIABLE TO ANY PARTY FOR DIRECT, INDIRECT,
* SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OF
* THIS SOFTWARE AND ITS DOCUMENTATION, EVEN IF ERIK CORRY HAS BEEN ADVISED
* OF THE POSSIBILITY OF SUCH DAMAGE.
*
* ERIK CORRY SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
* PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS ON AN "AS IS"
* BASIS, AND ERIK CORRY HAS NO OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT,
* UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
* Portions of this software Copyright (c) 1995 Brown University.
* All rights reserved.
*
* Permission to use, copy, modify, and distribute this software and its
* documentation for any purpose, without fee, and without written agreement
* is hereby granted, provided that the above copyright notice and the
* following two paragraphs appear in all copies of this software.
*
* IN NO EVENT SHALL BROWN UNIVERSITY BE LIABLE TO ANY PARTY FOR
* DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT
* OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN IF BROWN
* UNIVERSITY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* BROWN UNIVERSITY SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
* PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS ON AN "AS IS"
* BASIS, AND BROWN UNIVERSITY HAS NO OBLIGATION TO PROVIDE MAINTENANCE,
* SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
*/
#include "SDL_video.h"
87
#include "SDL_cpuinfo.h"
88
89
90
91
92
#include "SDL_yuv_sw_c.h"
/* The colorspace conversion functions */
93
#if (__GNUC__ > 2) && defined(__i386__) && __OPTIMIZE__ && SDL_ASSEMBLY_ROUTINES
94
extern void Color565DitherYV12MMX1X(int *colortab, Uint32 * rgb_2_pix,
95
96
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
97
98
99
100
101
102
103
104
105
106
107
108
int rows, int cols, int mod);
extern void ColorRGBDitherYV12MMX1X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod);
#endif
static void
Color16DitherYV12Mod1X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
109
{
110
111
112
unsigned short *row1;
unsigned short *row2;
unsigned char *lum2;
113
114
115
116
117
118
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
119
row1 = (unsigned short *) out;
120
121
122
123
124
125
row2 = row1 + cols + mod;
lum2 = lum + cols;
mod += cols + mod;
y = rows / 2;
126
while (y--) {
127
x = cols_2;
128
while (x--) {
129
130
register int L;
131
132
133
134
135
136
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
++cr;
++cb;
137
138
L = *lum++;
139
140
141
*row1++ = (unsigned short) (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
142
143
L = *lum++;
144
145
146
*row1++ = (unsigned short) (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
147
148
149
150
151
/* Now, do second row. */
L = *lum2++;
152
153
154
*row2++ = (unsigned short) (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
155
156
L = *lum2++;
157
158
159
*row2++ = (unsigned short) (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
160
161
162
163
164
165
166
}
/*
* These values are at the start of the next line, (due
* to the ++'s above),but they need to be at the start
* of the line after that.
*/
167
lum += cols;
168
169
170
171
172
173
lum2 += cols;
row1 += mod;
row2 += mod;
}
}
174
175
176
177
178
static void
Color24DitherYV12Mod1X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
179
180
{
unsigned int value;
181
182
183
unsigned char *row1;
unsigned char *row2;
unsigned char *lum2;
184
185
186
187
188
189
190
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
row1 = out;
191
row2 = row1 + cols * 3 + mod * 3;
192
193
194
195
196
197
lum2 = lum + cols;
mod += cols + mod;
mod *= 3;
y = rows / 2;
198
while (y--) {
199
x = cols_2;
200
while (x--) {
201
202
register int L;
203
204
205
206
207
208
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
++cr;
++cb;
209
210
L = *lum++;
211
212
213
214
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
*row1++ = (value) & 0xFF;
*row1++ = (value >> 8) & 0xFF;
215
216
217
*row1++ = (value >> 16) & 0xFF;
L = *lum++;
218
219
220
221
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
*row1++ = (value) & 0xFF;
*row1++ = (value >> 8) & 0xFF;
222
223
224
225
226
227
*row1++ = (value >> 16) & 0xFF;
/* Now, do second row. */
L = *lum2++;
228
229
230
231
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
*row2++ = (value) & 0xFF;
*row2++ = (value >> 8) & 0xFF;
232
233
234
*row2++ = (value >> 16) & 0xFF;
L = *lum2++;
235
236
237
238
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
*row2++ = (value) & 0xFF;
*row2++ = (value >> 8) & 0xFF;
239
240
241
242
243
244
245
246
*row2++ = (value >> 16) & 0xFF;
}
/*
* These values are at the start of the next line, (due
* to the ++'s above),but they need to be at the start
* of the line after that.
*/
247
lum += cols;
248
249
250
251
252
253
lum2 += cols;
row1 += mod;
row2 += mod;
}
}
254
255
256
257
258
static void
Color32DitherYV12Mod1X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
259
{
260
261
262
unsigned int *row1;
unsigned int *row2;
unsigned char *lum2;
263
264
265
266
267
268
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
269
row1 = (unsigned int *) out;
270
271
272
273
274
275
row2 = row1 + cols + mod;
lum2 = lum + cols;
mod += cols + mod;
y = rows / 2;
276
while (y--) {
277
x = cols_2;
278
while (x--) {
279
280
register int L;
281
282
283
284
285
286
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
++cr;
++cb;
287
288
L = *lum++;
289
290
*row1++ = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
291
292
L = *lum++;
293
294
*row1++ = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
295
296
297
298
299
/* Now, do second row. */
L = *lum2++;
300
301
*row2++ = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
302
303
L = *lum2++;
304
305
*row2++ = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
306
307
308
309
310
311
312
}
/*
* These values are at the start of the next line, (due
* to the ++'s above),but they need to be at the start
* of the line after that.
*/
313
lum += cols;
314
315
316
317
318
319
320
321
322
323
324
lum2 += cols;
row1 += mod;
row2 += mod;
}
}
/*
* In this function I make use of a nasty trick. The tables have the lower
* 16 bits replicated in the upper 16. This means I can write ints and get
* the horisontal doubling for free (almost).
*/
325
326
327
328
329
static void
Color16DitherYV12Mod2X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
330
{
331
332
333
334
unsigned int *row1 = (unsigned int *) out;
const int next_row = cols + (mod / 2);
unsigned int *row2 = row1 + 2 * next_row;
unsigned char *lum2;
335
336
337
338
339
340
341
342
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
lum2 = lum + cols;
343
mod = (next_row * 3) + (mod / 2);
344
345
y = rows / 2;
346
while (y--) {
347
x = cols_2;
348
while (x--) {
349
350
register int L;
351
352
353
354
355
356
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
++cr;
++cb;
357
358
L = *lum++;
359
360
361
row1[0] = row1[next_row] = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
362
363
364
row1++;
L = *lum++;
365
366
367
row1[0] = row1[next_row] = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
368
369
370
371
372
373
row1++;
/* Now, do second row. */
L = *lum2++;
374
375
376
row2[0] = row2[next_row] = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
377
378
379
row2++;
L = *lum2++;
380
381
382
row2[0] = row2[next_row] = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
383
384
385
386
387
388
389
390
row2++;
}
/*
* These values are at the start of the next line, (due
* to the ++'s above),but they need to be at the start
* of the line after that.
*/
391
lum += cols;
392
393
394
395
396
397
lum2 += cols;
row1 += mod;
row2 += mod;
}
}
398
399
400
401
402
static void
Color24DitherYV12Mod2X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
403
404
{
unsigned int value;
405
406
407
408
unsigned char *row1 = out;
const int next_row = (cols * 2 + mod) * 3;
unsigned char *row2 = row1 + 2 * next_row;
unsigned char *lum2;
409
410
411
412
413
414
415
416
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
lum2 = lum + cols;
417
mod = next_row * 3 + mod * 3;
418
419
y = rows / 2;
420
while (y--) {
421
x = cols_2;
422
while (x--) {
423
424
register int L;
425
426
427
428
429
430
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
++cr;
++cb;
431
432
L = *lum++;
433
434
435
436
437
438
439
440
441
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
row1[0 + 0] = row1[3 + 0] = row1[next_row + 0] =
row1[next_row + 3 + 0] = (value) & 0xFF;
row1[0 + 1] = row1[3 + 1] = row1[next_row + 1] =
row1[next_row + 3 + 1] = (value >> 8) & 0xFF;
row1[0 + 2] = row1[3 + 2] = row1[next_row + 2] =
row1[next_row + 3 + 2] = (value >> 16) & 0xFF;
row1 += 2 * 3;
442
443
L = *lum++;
444
445
446
447
448
449
450
451
452
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
row1[0 + 0] = row1[3 + 0] = row1[next_row + 0] =
row1[next_row + 3 + 0] = (value) & 0xFF;
row1[0 + 1] = row1[3 + 1] = row1[next_row + 1] =
row1[next_row + 3 + 1] = (value >> 8) & 0xFF;
row1[0 + 2] = row1[3 + 2] = row1[next_row + 2] =
row1[next_row + 3 + 2] = (value >> 16) & 0xFF;
row1 += 2 * 3;
453
454
455
456
457
/* Now, do second row. */
L = *lum2++;
458
459
460
461
462
463
464
465
466
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
row2[0 + 0] = row2[3 + 0] = row2[next_row + 0] =
row2[next_row + 3 + 0] = (value) & 0xFF;
row2[0 + 1] = row2[3 + 1] = row2[next_row + 1] =
row2[next_row + 3 + 1] = (value >> 8) & 0xFF;
row2[0 + 2] = row2[3 + 2] = row2[next_row + 2] =
row2[next_row + 3 + 2] = (value >> 16) & 0xFF;
row2 += 2 * 3;
467
468
L = *lum2++;
469
470
471
472
473
474
475
476
477
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
row2[0 + 0] = row2[3 + 0] = row2[next_row + 0] =
row2[next_row + 3 + 0] = (value) & 0xFF;
row2[0 + 1] = row2[3 + 1] = row2[next_row + 1] =
row2[next_row + 3 + 1] = (value >> 8) & 0xFF;
row2[0 + 2] = row2[3 + 2] = row2[next_row + 2] =
row2[next_row + 3 + 2] = (value >> 16) & 0xFF;
row2 += 2 * 3;
478
479
480
481
482
483
484
}
/*
* These values are at the start of the next line, (due
* to the ++'s above),but they need to be at the start
* of the line after that.
*/
485
lum += cols;
486
487
488
489
490
491
lum2 += cols;
row1 += mod;
row2 += mod;
}
}
492
493
494
495
496
static void
Color32DitherYV12Mod2X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
497
{
498
499
500
501
unsigned int *row1 = (unsigned int *) out;
const int next_row = cols * 2 + mod;
unsigned int *row2 = row1 + 2 * next_row;
unsigned char *lum2;
502
503
504
505
506
507
508
509
510
511
512
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
lum2 = lum + cols;
mod = (next_row * 3) + mod;
y = rows / 2;
513
while (y--) {
514
x = cols_2;
515
while (x--) {
516
517
register int L;
518
519
520
521
522
523
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
++cr;
++cb;
524
525
L = *lum++;
526
527
528
row1[0] = row1[1] = row1[next_row] = row1[next_row + 1] =
(rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
529
530
531
row1 += 2;
L = *lum++;
532
533
534
row1[0] = row1[1] = row1[next_row] = row1[next_row + 1] =
(rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
535
536
537
538
539
540
row1 += 2;
/* Now, do second row. */
L = *lum2++;
541
542
543
row2[0] = row2[1] = row2[next_row] = row2[next_row + 1] =
(rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
544
545
546
row2 += 2;
L = *lum2++;
547
548
549
row2[0] = row2[1] = row2[next_row] = row2[next_row + 1] =
(rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
550
551
552
553
554
555
556
557
row2 += 2;
}
/*
* These values are at the start of the next line, (due
* to the ++'s above),but they need to be at the start
* of the line after that.
*/
558
lum += cols;
559
560
561
562
563
564
lum2 += cols;
row1 += mod;
row2 += mod;
}
}
565
566
567
568
569
static void
Color16DitherYUY2Mod1X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
570
{
571
unsigned short *row;
572
573
574
575
576
577
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
578
row = (unsigned short *) out;
579
580
y = rows;
581
while (y--) {
582
x = cols_2;
583
while (x--) {
584
585
register int L;
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
cr += 4;
cb += 4;
L = *lum;
lum += 2;
*row++ = (unsigned short) (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
L = *lum;
lum += 2;
*row++ = (unsigned short) (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
604
605
606
607
608
609
610
}
row += mod;
}
}
611
612
613
614
615
static void
Color24DitherYUY2Mod1X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
616
617
{
unsigned int value;
618
unsigned char *row;
619
620
621
622
623
624
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
625
row = (unsigned char *) out;
626
627
mod *= 3;
y = rows;
628
while (y--) {
629
x = cols_2;
630
while (x--) {
631
632
register int L;
633
634
635
636
637
638
639
640
641
642
643
644
645
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
cr += 4;
cb += 4;
L = *lum;
lum += 2;
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
*row++ = (value) & 0xFF;
*row++ = (value >> 8) & 0xFF;
646
647
*row++ = (value >> 16) & 0xFF;
648
649
650
651
652
653
L = *lum;
lum += 2;
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
*row++ = (value) & 0xFF;
*row++ = (value >> 8) & 0xFF;
654
655
656
657
658
659
660
*row++ = (value >> 16) & 0xFF;
}
row += mod;
}
}
661
662
663
664
665
static void
Color32DitherYUY2Mod1X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
666
{
667
unsigned int *row;
668
669
670
671
672
673
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
674
row = (unsigned int *) out;
675
y = rows;
676
while (y--) {
677
x = cols_2;
678
while (x--) {
679
680
register int L;
681
682
683
684
685
686
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
cr += 4;
cb += 4;
687
688
689
690
691
L = *lum;
lum += 2;
*row++ = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
692
693
694
695
696
L = *lum;
lum += 2;
*row++ = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
697
698
699
700
701
702
703
704
705
706
707
708
}
row += mod;
}
}
/*
* In this function I make use of a nasty trick. The tables have the lower
* 16 bits replicated in the upper 16. This means I can write ints and get
* the horisontal doubling for free (almost).
*/
709
710
711
712
713
static void
Color16DitherYUY2Mod2X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
714
{
715
716
unsigned int *row = (unsigned int *) out;
const int next_row = cols + (mod / 2);
717
718
719
720
721
722
723
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
y = rows;
724
while (y--) {
725
x = cols_2;
726
while (x--) {
727
728
register int L;
729
730
731
732
733
734
735
736
737
738
739
740
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
cr += 4;
cb += 4;
L = *lum;
lum += 2;
row[0] = row[next_row] = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
741
742
row++;
743
744
745
746
747
L = *lum;
lum += 2;
row[0] = row[next_row] = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] |
rgb_2_pix[L + cb_b]);
748
749
750
751
752
753
754
row++;
}
row += next_row;
}
}
755
756
757
758
759
static void
Color24DitherYUY2Mod2X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
760
761
{
unsigned int value;
762
763
unsigned char *row = out;
const int next_row = (cols * 2 + mod) * 3;
764
765
766
767
768
769
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
y = rows;
770
while (y--) {
771
x = cols_2;
772
while (x--) {
773
774
register int L;
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
cr += 4;
cb += 4;
L = *lum;
lum += 2;
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
row[0 + 0] = row[3 + 0] = row[next_row + 0] =
row[next_row + 3 + 0] = (value) & 0xFF;
row[0 + 1] = row[3 + 1] = row[next_row + 1] =
row[next_row + 3 + 1] = (value >> 8) & 0xFF;
row[0 + 2] = row[3 + 2] = row[next_row + 2] =
row[next_row + 3 + 2] = (value >> 16) & 0xFF;
row += 2 * 3;
L = *lum;
lum += 2;
value = (rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
row[0 + 0] = row[3 + 0] = row[next_row + 0] =
row[next_row + 3 + 0] = (value) & 0xFF;
row[0 + 1] = row[3 + 1] = row[next_row + 1] =
row[next_row + 3 + 1] = (value >> 8) & 0xFF;
row[0 + 2] = row[3 + 2] = row[next_row + 2] =
row[next_row + 3 + 2] = (value >> 16) & 0xFF;
row += 2 * 3;
805
806
807
808
809
810
}
row += next_row;
}
}
811
812
813
814
815
static void
Color32DitherYUY2Mod2X(int *colortab, Uint32 * rgb_2_pix,
unsigned char *lum, unsigned char *cr,
unsigned char *cb, unsigned char *out,
int rows, int cols, int mod)
816
{
817
818
unsigned int *row = (unsigned int *) out;
const int next_row = cols * 2 + mod;
819
820
821
822
823
int x, y;
int cr_r;
int crb_g;
int cb_b;
int cols_2 = cols / 2;
824
mod += mod;
825
y = rows;
826
while (y--) {
827
x = cols_2;
828
while (x--) {
829
830
register int L;
831
832
833
834
835
836
837
838
839
840
841
842
cr_r = 0 * 768 + 256 + colortab[*cr + 0 * 256];
crb_g = 1 * 768 + 256 + colortab[*cr + 1 * 256]
+ colortab[*cb + 2 * 256];
cb_b = 2 * 768 + 256 + colortab[*cb + 3 * 256];
cr += 4;
cb += 4;
L = *lum;
lum += 2;
row[0] = row[1] = row[next_row] = row[next_row + 1] =
(rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
843
844
row += 2;
845
846
847
848
849
L = *lum;
lum += 2;
row[0] = row[1] = row[next_row] = row[next_row + 1] =
(rgb_2_pix[L + cr_r] |
rgb_2_pix[L + crb_g] | rgb_2_pix[L + cb_b]);
850
851
852
853
854
855
856
857
858
859
860
861
862
row += 2;
}
row += next_row;
}
}
/*
* How many 1 bits are there in the Uint32.
* Low performance, do not call often.
*/
863
864
static int
number_of_bits_set(Uint32 a)
865
{
866
867
868
869
870
if (!a)
return 0;
if (a & 1)
return 1 + number_of_bits_set(a >> 1);
return (number_of_bits_set(a >> 1));
871
872
873
874
875
876
}
/*
* How many 0 bits are there at least significant end of Uint32.
* Low performance, do not call often.
*/
877
878
static int
free_bits_at_bottom(Uint32 a)
879
{
880
881
882
883
884
885
/* assume char is 8 bits */
if (!a)
return sizeof(Uint32) * 8;
if (((Sint32) a) & 1l)
return 0;
return 1 + free_bits_at_bottom(a >> 1);
886
887
}
888
889
static int
SDL_SW_SetupYUVDisplay(SDL_SW_YUVTexture * swdata, Uint32 target_format)
890
{
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
Uint32 *r_2_pix_alloc;
Uint32 *g_2_pix_alloc;
Uint32 *b_2_pix_alloc;
int i;
int bpp;
Uint32 Rmask, Gmask, Bmask, Amask;
if (!SDL_PixelFormatEnumToMasks
(target_format, &bpp, &Rmask, &Gmask, &Bmask, &Amask) || bpp < 15) {
SDL_SetError("Unsupported YUV destination format");
return -1;
}
swdata->target_format = target_format;
r_2_pix_alloc = &swdata->rgb_2_pix[0 * 768];
g_2_pix_alloc = &swdata->rgb_2_pix[1 * 768];
b_2_pix_alloc = &swdata->rgb_2_pix[2 * 768];
/*
* Set up entries 0-255 in rgb-to-pixel value tables.
*/
for (i = 0; i < 256; ++i) {
r_2_pix_alloc[i + 256] = i >> (8 - number_of_bits_set(Rmask));
r_2_pix_alloc[i + 256] <<= free_bits_at_bottom(Rmask);
915
r_2_pix_alloc[i + 256] |= Amask;
916
917
g_2_pix_alloc[i + 256] = i >> (8 - number_of_bits_set(Gmask));
g_2_pix_alloc[i + 256] <<= free_bits_at_bottom(Gmask);
918
g_2_pix_alloc[i + 256] |= Amask;
919
920
b_2_pix_alloc[i + 256] = i >> (8 - number_of_bits_set(Bmask));
b_2_pix_alloc[i + 256] <<= free_bits_at_bottom(Bmask);
921
b_2_pix_alloc[i + 256] |= Amask;
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
}
/*
* If we have 16-bit output depth, then we double the value
* in the top word. This means that we can write out both
* pixels in the pixel doubling mode with one op. It is
* harmless in the normal case as storing a 32-bit value
* through a short pointer will lose the top bits anyway.
*/
if (SDL_BYTESPERPIXEL(target_format) == 2) {
for (i = 0; i < 256; ++i) {
r_2_pix_alloc[i + 256] |= (r_2_pix_alloc[i + 256]) << 16;
g_2_pix_alloc[i + 256] |= (g_2_pix_alloc[i + 256]) << 16;
b_2_pix_alloc[i + 256] |= (b_2_pix_alloc[i + 256]) << 16;
}
}
/*
* Spread out the values we have to the rest of the array so that
* we do not need to check for overflow.
*/
for (i = 0; i < 256; ++i) {
r_2_pix_alloc[i] = r_2_pix_alloc[256];
r_2_pix_alloc[i + 512] = r_2_pix_alloc[511];
g_2_pix_alloc[i] = g_2_pix_alloc[256];
g_2_pix_alloc[i + 512] = g_2_pix_alloc[511];
b_2_pix_alloc[i] = b_2_pix_alloc[256];
b_2_pix_alloc[i + 512] = b_2_pix_alloc[511];
}
/* You have chosen wisely... */
953
switch (swdata->format) {
954
955
case SDL_PIXELFORMAT_YV12:
case SDL_PIXELFORMAT_IYUV:
956
if (SDL_BYTESPERPIXEL(target_format) == 2) {
957
#if (__GNUC__ > 2) && defined(__i386__) && __OPTIMIZE__ && SDL_ASSEMBLY_ROUTINES
958
959
/* inline assembly functions */
if (SDL_HasMMX() && (Rmask == 0xF800) &&
960
(Gmask == 0x07E0) && (Bmask == 0x001F)
961
&& (swdata->w & 15) == 0) {
962
/*printf("Using MMX 16-bit 565 dither\n");*/
963
964
swdata->Display1X = Color565DitherYV12MMX1X;
} else {
965
/*printf("Using C 16-bit dither\n");*/
966
967
swdata->Display1X = Color16DitherYV12Mod1X;
}
968
#else
969
swdata->Display1X = Color16DitherYV12Mod1X;
970
#endif
971
972
973
974
975
976
977
swdata->Display2X = Color16DitherYV12Mod2X;
}
if (SDL_BYTESPERPIXEL(target_format) == 3) {
swdata->Display1X = Color24DitherYV12Mod1X;
swdata->Display2X = Color24DitherYV12Mod2X;
}
if (SDL_BYTESPERPIXEL(target_format) == 4) {
978
#if (__GNUC__ > 2) && defined(__i386__) && __OPTIMIZE__ && SDL_ASSEMBLY_ROUTINES
979
980
981
/* inline assembly functions */
if (SDL_HasMMX() && (Rmask == 0x00FF0000) &&
(Gmask == 0x0000FF00) &&
982
(Bmask == 0x000000FF) && (swdata->w & 15) == 0) {
983
/*printf("Using MMX 32-bit dither\n");*/
984
985
swdata->Display1X = ColorRGBDitherYV12MMX1X;
} else {
986
/*printf("Using C 32-bit dither\n");*/
987
988
swdata->Display1X = Color32DitherYV12Mod1X;
}
989
#else
990
swdata->Display1X = Color32DitherYV12Mod1X;
991
#endif
992
993
994
swdata->Display2X = Color32DitherYV12Mod2X;
}
break;
995
996
997
case SDL_PIXELFORMAT_YUY2:
case SDL_PIXELFORMAT_UYVY:
case SDL_PIXELFORMAT_YVYU:
998
999
1000
if (SDL_BYTESPERPIXEL(target_format) == 2) {
swdata->Display1X = Color16DitherYUY2Mod1X;
swdata->Display2X = Color16DitherYUY2Mod2X;