forked from mirrorer/libbpg
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathlibbpg.c
1957 lines (1776 loc) · 54.4 KB
/
libbpg.c
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
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
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
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
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
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
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
915
916
917
918
919
920
921
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
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
* libbpg
*
* Copyright (c) 2014 Fabrice Bellard
*
* 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, sublicense, 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 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 NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS 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.
*/
#include <math.h>
#ifdef EMSCRIPTEN
#include <emscripten.h>
#endif
#include <libavutil/opt.h>
#include <libavcodec/avcodec.h>
#include <libavutil/common.h>
/* The following global defines are used:
- USE_VAR_BIT_DEPTH : support of bit depth > 8 bits
- USE_PRED : support of animations
*/
#ifndef EMSCRIPTEN
#define USE_RGB48 /* support all pixel formats */
//#define DEBUG
#endif
#if !defined(DEBUG)
#define NDEBUG
#endif
#include <assert.h>
#include "libbpg.h"
#define BPG_HEADER_MAGIC 0x425047fb
#define ITAPS2 4
#define ITAPS (2 * ITAPS2) /* number of taps of the interpolation filter */
#ifdef USE_VAR_BIT_DEPTH
typedef uint16_t PIXEL;
#else
typedef uint8_t PIXEL;
#endif
#define MAX_DATA_SIZE ((1 << 30) - 1)
typedef struct {
int c_shift;
int c_rnd;
int c_one;
int y_one, y_offset;
int c_r_cr, c_g_cb, c_g_cr, c_b_cb;
int c_center;
int bit_depth;
int limited_range;
} ColorConvertState;
typedef void ColorConvertFunc(ColorConvertState *s,
uint8_t *dst, const PIXEL *y_ptr,
const PIXEL *cb_ptr, const PIXEL *cr_ptr,
int n, int incr);
struct BPGDecoderContext {
AVCodecContext *dec_ctx;
AVCodecContext *alpha_dec_ctx;
AVFrame *frame;
AVFrame *alpha_frame;
int w, h;
BPGImageFormatEnum format;
uint8_t c_h_phase; /* only used for 422 and 420 */
uint8_t has_alpha; /* true if alpha or W plane */
uint8_t bit_depth;
uint8_t has_w_plane;
uint8_t limited_range;
uint8_t premultiplied_alpha;
uint8_t has_animation;
BPGColorSpaceEnum color_space;
uint8_t keep_extension_data; /* true if the extension data must be
kept during parsing */
uint8_t decode_animation; /* true if animation decoding is enabled */
BPGExtensionData *first_md;
/* animation */
uint16_t loop_count;
uint16_t frame_delay_num;
uint16_t frame_delay_den;
uint8_t *input_buf;
int input_buf_pos;
int input_buf_len;
/* the following is used for format conversion */
uint8_t output_inited;
BPGDecoderOutputFormat out_fmt;
uint8_t is_rgba;
uint8_t is_16bpp;
uint8_t is_cmyk;
int y; /* current line */
int w2, h2;
const uint8_t *y_buf, *cb_buf, *cr_buf, *a_buf;
int y_linesize, cb_linesize, cr_linesize, a_linesize;
PIXEL *cb_buf2, *cr_buf2, *cb_buf3[ITAPS], *cr_buf3[ITAPS];
int16_t *c_buf4;
ColorConvertState cvt;
ColorConvertFunc *cvt_func;
};
/* ffmpeg utilities */
#ifdef USE_AV_LOG
void av_log(void* avcl, int level, const char *fmt, ...)
{
#ifdef DEBUG
va_list ap;
va_start(ap, fmt);
vfprintf(stderr, fmt, ap);
va_end(ap);
#endif
}
void avpriv_report_missing_feature(void *avc, const char *msg, ...)
{
#ifdef DEBUG
va_list ap;
va_start(ap, msg);
vfprintf(stderr, msg, ap);
va_end(ap);
#endif
}
#endif /* USE_AV_LOG */
/* return < 0 if error, otherwise the consumed length */
static int get_ue32(uint32_t *pv, const uint8_t *buf, int len)
{
const uint8_t *p;
uint32_t v;
int a;
if (len <= 0)
return -1;
p = buf;
a = *p++;
len--;
if (a < 0x80) {
*pv = a;
return 1;
} else if (a == 0x80) {
/* we don't accept non canonical encodings */
return -1;
}
v = a & 0x7f;
for(;;) {
if (len <= 0)
return -1;
a = *p++;
len--;
v = (v << 7) | (a & 0x7f);
if (!(a & 0x80))
break;
}
*pv = v;
return p - buf;
}
static int get_ue(uint32_t *pv, const uint8_t *buf, int len)
{
int ret;
ret = get_ue32(pv, buf, len);
if (ret < 0)
return ret;
/* limit the maximum size to avoid overflows in buffer
computations */
if (*pv > MAX_DATA_SIZE)
return -1;
return ret;
}
static int build_msps(uint8_t **pbuf, int *pbuf_len,
const uint8_t *input_data, int input_data_len1,
int width, int height, int chroma_format_idc,
int bit_depth)
{
int input_data_len = input_data_len1;
int idx, msps_len, ret, buf_len, i;
uint32_t len;
uint8_t *buf, *msps_buf;
*pbuf = NULL;
/* build the modified SPS header to please libavcodec */
ret = get_ue(&len, input_data, input_data_len);
if (ret < 0)
return -1;
input_data += ret;
input_data_len -= ret;
if (len > input_data_len)
return -1;
msps_len = 1 + 4 + 4 + 1 + len;
msps_buf = av_malloc(msps_len);
idx = 0;
msps_buf[idx++] = chroma_format_idc;
msps_buf[idx++] = (width >> 24);
msps_buf[idx++] = (width >> 16);
msps_buf[idx++] = (width >> 8);
msps_buf[idx++] = (width >> 0);
msps_buf[idx++] = (height >> 24);
msps_buf[idx++] = (height >> 16);
msps_buf[idx++] = (height >> 8);
msps_buf[idx++] = (height >> 0);
msps_buf[idx++] = bit_depth - 8;
memcpy(msps_buf + idx, input_data, len);
idx += len;
assert(idx == msps_len);
input_data += len;
input_data_len -= len;
buf_len = 4 + 2 + msps_len * 2 + 4 + (input_data_len - len);
buf = av_malloc(buf_len);
idx = 0;
/* NAL header */
buf[idx++] = 0x00;
buf[idx++] = 0x00;
buf[idx++] = 0x00;
buf[idx++] = 0x01;
buf[idx++] = (48 << 1); /* application specific NAL unit type */
buf[idx++] = 1;
/* add the modified SPS with the correct escape codes */
i = 0;
while (i < msps_len) {
if ((i + 1) < msps_len && msps_buf[i] == 0 && msps_buf[i + 1] == 0) {
buf[idx++] = 0x00;
buf[idx++] = 0x00;
buf[idx++] = 0x03;
i += 2;
} else {
buf[idx++] = msps_buf[i++];
}
}
/* the last byte cannot be 0 */
if (idx == 0 || buf[idx - 1] == 0x00)
buf[idx++] = 0x80;
av_free(msps_buf);
*pbuf_len = idx;
*pbuf = buf;
return input_data_len1 - input_data_len;
}
/* return the position of the end of the NAL or -1 if error */
static int find_nal_end(const uint8_t *buf, int buf_len, int has_startcode)
{
int idx;
idx = 0;
if (has_startcode) {
if (buf_len >= 4 &&
buf[0] == 0 && buf[1] == 0 && buf[2] == 0 && buf[3] == 1) {
idx = 4;
} else if (buf_len >= 3 &&
buf[0] == 0 && buf[1] == 0 && buf[2] == 1) {
idx = 3;
} else {
return -1;
}
}
/* NAL header */
if (idx + 2 > buf_len)
return -1;
/* find the last byte */
for(;;) {
if (idx + 2 >= buf_len) {
idx = buf_len;
break;
}
if (buf[idx] == 0 && buf[idx + 1] == 0 && buf[idx + 2] == 1)
break;
if (idx + 3 < buf_len &&
buf[idx] == 0 && buf[idx + 1] == 0 && buf[idx + 2] == 0 && buf[idx + 3] == 1)
break;
idx++;
}
return idx;
}
typedef struct {
uint8_t *buf;
int size;
int len;
} DynBuf;
static void dyn_buf_init(DynBuf *s)
{
s->buf = NULL;
s->size = 0;
s->len = 0;
}
static int dyn_buf_resize(DynBuf *s, int size)
{
int new_size;
uint8_t *new_buf;
if (size <= s->size)
return 0;
new_size = (s->size * 3) / 2;
if (new_size < size)
new_size = size;
new_buf = av_realloc(s->buf, new_size);
if (!new_buf)
return -1;
s->buf = new_buf;
s->size = new_size;
return 0;
}
static int dyn_buf_push(DynBuf *s, const uint8_t *data, int len)
{
if (dyn_buf_resize(s, s->len + len) < 0)
return -1;
memcpy(s->buf + s->len, data, len);
s->len += len;
return 0;
}
extern AVCodec ff_hevc_decoder;
static int hevc_decode_init1(DynBuf *pbuf, AVFrame **pframe,
AVCodecContext **pc,
const uint8_t *buf, int buf_len,
int width, int height, int chroma_format_idc,
int bit_depth)
{
AVCodec *codec;
AVCodecContext *c;
AVFrame *frame;
uint8_t *nal_buf;
int nal_len, ret, ret1;
ret = build_msps(&nal_buf, &nal_len, buf, buf_len,
width, height, chroma_format_idc, bit_depth);
if (ret < 0)
return -1;
ret1 = dyn_buf_push(pbuf, nal_buf, nal_len);
av_free(nal_buf);
if (ret1 < 0)
return -1;
codec = &ff_hevc_decoder;
c = avcodec_alloc_context3(codec);
if (!c)
return -1;
frame = av_frame_alloc();
if (!frame)
return -1;
/* for testing: use the MD5 or CRC in SEI to check the decoded bit
stream. */
c->err_recognition |= AV_EF_CRCCHECK;
/* open it */
if (avcodec_open2(c, codec, NULL) < 0) {
av_frame_free(&frame);
return -1;
}
*pc = c;
*pframe = frame;
return ret;
}
static int hevc_write_frame(AVCodecContext *avctx,
AVFrame *frame,
uint8_t *buf, int buf_len)
{
AVPacket avpkt;
int len, got_frame;
av_init_packet(&avpkt);
avpkt.data = (uint8_t *)buf;
avpkt.size = buf_len;
/* avoid using uninitialized data */
memset(buf + buf_len, 0, FF_INPUT_BUFFER_PADDING_SIZE);
len = avcodec_decode_video2(avctx, frame, &got_frame, &avpkt);
if (len < 0 || !got_frame)
return -1;
else
return 0;
}
static int hevc_decode_frame_internal(BPGDecoderContext *s,
DynBuf *abuf, DynBuf *cbuf,
const uint8_t *buf, int buf_len1,
int first_nal)
{
int nal_len, start, nal_buf_len, ret, nuh_layer_id, buf_len, has_alpha;
int nut, frame_start_found[2];
DynBuf *pbuf;
uint8_t *nal_buf;
has_alpha = (s->alpha_dec_ctx != NULL);
buf_len = buf_len1;
frame_start_found[0] = 0;
frame_start_found[1] = 0;
while (buf_len > 0) {
if (buf_len < (first_nal ? 3 : 0) + 2)
goto fail;
if (first_nal)
start = 0;
else
start = 3 + (buf[2] == 0);
if (buf_len < start + 3)
goto fail;
nuh_layer_id = ((buf[start] & 1) << 5) | (buf[start + 1] >> 3);
nut = (buf[start] >> 1) & 0x3f;
#if 0
printf("nal: type=%d layer_id=%d fs=%d %d\n",
nut, nuh_layer_id, frame_start_found[0], frame_start_found[1]);
#endif
/* Note: we assume the alpha and color data are correctly
interleaved */
if ((nut >= 32 && nut <= 35) || nut == 39 || nut >= 41) {
if (frame_start_found[0] && frame_start_found[has_alpha])
break;
} else if ((nut <= 9 || (nut >= 16 && nut <= 21)) &&
start + 2 < buf_len && (buf[start + 2] & 0x80)) {
/* first slice segment */
if (frame_start_found[0] && frame_start_found[has_alpha])
break;
if (has_alpha && nuh_layer_id == 1)
frame_start_found[1] = 1;
else
frame_start_found[0] = 1;
}
nal_len = find_nal_end(buf, buf_len, !first_nal);
if (nal_len < 0)
goto fail;
nal_buf_len = nal_len - start + 3;
if (has_alpha && nuh_layer_id == 1)
pbuf = abuf;
else
pbuf = cbuf;
if (dyn_buf_resize(pbuf, pbuf->len + nal_buf_len) < 0)
goto fail;
nal_buf = pbuf->buf + pbuf->len;
nal_buf[0] = 0x00;
nal_buf[1] = 0x00;
nal_buf[2] = 0x01;
memcpy(nal_buf + 3, buf + start, nal_len - start);
if (has_alpha && nuh_layer_id == 1)
nal_buf[4] &= 0x7;
pbuf->len += nal_buf_len;
buf += nal_len;
buf_len -= nal_len;
first_nal = 0;
}
if (s->alpha_dec_ctx) {
if (dyn_buf_resize(abuf, abuf->len + FF_INPUT_BUFFER_PADDING_SIZE) < 0)
goto fail;
ret = hevc_write_frame(s->alpha_dec_ctx, s->alpha_frame, abuf->buf, abuf->len);
if (ret < 0)
goto fail;
}
if (dyn_buf_resize(cbuf, cbuf->len + FF_INPUT_BUFFER_PADDING_SIZE) < 0)
goto fail;
ret = hevc_write_frame(s->dec_ctx, s->frame, cbuf->buf, cbuf->len);
if (ret < 0)
goto fail;
ret = buf_len1 - buf_len;
done:
return ret;
fail:
ret = -1;
goto done;
}
/* decode the first frame */
static int hevc_decode_start(BPGDecoderContext *s,
const uint8_t *buf, int buf_len1,
int width, int height, int chroma_format_idc,
int bit_depth, int has_alpha)
{
int ret, buf_len;
DynBuf abuf_s, *abuf = &abuf_s;
DynBuf cbuf_s, *cbuf = &cbuf_s;
dyn_buf_init(abuf);
dyn_buf_init(cbuf);
buf_len = buf_len1;
if (has_alpha) {
ret = hevc_decode_init1(abuf, &s->alpha_frame, &s->alpha_dec_ctx,
buf, buf_len, width, height, 0, bit_depth);
if (ret < 0)
goto fail;
buf += ret;
buf_len -= ret;
}
ret = hevc_decode_init1(cbuf, &s->frame, &s->dec_ctx,
buf, buf_len, width, height, chroma_format_idc,
bit_depth);
if (ret < 0)
goto fail;
buf += ret;
buf_len -= ret;
ret = hevc_decode_frame_internal(s, abuf, cbuf, buf, buf_len, 1);
av_free(abuf->buf);
av_free(cbuf->buf);
if (ret < 0)
goto fail;
buf_len -= ret;
return buf_len1 - buf_len;
fail:
return -1;
}
#ifdef USE_PRED
static int hevc_decode_frame(BPGDecoderContext *s,
const uint8_t *buf, int buf_len)
{
int ret;
DynBuf abuf_s, *abuf = &abuf_s;
DynBuf cbuf_s, *cbuf = &cbuf_s;
dyn_buf_init(abuf);
dyn_buf_init(cbuf);
ret = hevc_decode_frame_internal(s, abuf, cbuf, buf, buf_len, 0);
av_free(abuf->buf);
av_free(cbuf->buf);
return ret;
}
#endif
static void hevc_decode_end(BPGDecoderContext *s)
{
if (s->alpha_dec_ctx) {
avcodec_close(s->alpha_dec_ctx);
av_free(s->alpha_dec_ctx);
s->alpha_dec_ctx = NULL;
}
if (s->dec_ctx) {
avcodec_close(s->dec_ctx);
av_free(s->dec_ctx);
s->dec_ctx = NULL;
}
}
uint8_t *bpg_decoder_get_data(BPGDecoderContext *img, int *pline_size, int plane)
{
int c_count;
if (img->format == BPG_FORMAT_GRAY)
c_count = 1;
else
c_count = 3;
if (plane < c_count) {
*pline_size = img->frame->linesize[plane];
return img->frame->data[plane];
} else if (img->has_alpha && plane == c_count) {
*pline_size = img->alpha_frame->linesize[0];
return img->alpha_frame->data[0];
} else {
*pline_size = 0;
return NULL;
}
}
int bpg_decoder_get_info(BPGDecoderContext *img, BPGImageInfo *p)
{
if (!img->frame)
return -1;
p->width = img->w;
p->height = img->h;
p->format = img->format;
p->has_alpha = img->has_alpha && !img->has_w_plane;
p->premultiplied_alpha = img->premultiplied_alpha;
p->has_w_plane = img->has_w_plane;
p->limited_range = img->limited_range;
p->color_space = img->color_space;
p->bit_depth = img->bit_depth;
p->has_animation = img->has_animation;
p->loop_count = img->loop_count;
return 0;
}
static inline int clamp_pix(int a, int pixel_max)
{
if (a < 0)
return 0;
else if (a > pixel_max)
return pixel_max;
else
return a;
}
static inline int clamp8(int a)
{
if (a < 0)
return 0;
else if (a > 255)
return 255;
else
return a;
}
/* 8 tap Lanczos interpolator (phase=0, symmetric) */
#define IP0C0 40
#define IP0C1 (-11)
#define IP0C2 4
#define IP0C3 (-1)
/* 7 tap Lanczos interpolator (phase=0.5) */
#define IP1C0 (-1)
#define IP1C1 4
#define IP1C2 (-10)
#define IP1C3 57
#define IP1C4 18
#define IP1C5 (-6)
#define IP1C6 2
/* interpolate by a factor of two assuming chroma is aligned with the
luma samples. */
static void interp2p0_simple(PIXEL *dst, const PIXEL *src, int n, int bit_depth)
{
int pixel_max;
pixel_max = (1 << bit_depth) - 1;
while (n >= 2) {
dst[0] = src[0];
dst[1] = clamp_pix(((src[-3] + src[4]) * IP0C3 +
(src[-2] + src[3]) * IP0C2 +
(src[-1] + src[2]) * IP0C1 +
(src[0] + src[1]) * IP0C0 + 32) >> 6, pixel_max);
dst += 2;
src++;
n -= 2;
}
if (n) {
dst[0] = src[0];
}
}
static void interp2p0_simple16(PIXEL *dst, const int16_t *src, int n, int bit_depth)
{
int shift1, offset1, shift0, offset0, pixel_max;
pixel_max = (1 << bit_depth) - 1;
shift0 = 14 - bit_depth;
offset0 = (1 << shift0) >> 1;
shift1 = 20 - bit_depth;
offset1 = 1 << (shift1 - 1);
while (n >= 2) {
dst[0] = clamp_pix((src[0] + offset0) >> shift0, pixel_max);
dst[1] = clamp_pix(((src[-3] + src[4]) * IP0C3 +
(src[-2] + src[3]) * IP0C2 +
(src[-1] + src[2]) * IP0C1 +
(src[0] + src[1]) * IP0C0 + offset1) >> shift1,
pixel_max);
dst += 2;
src++;
n -= 2;
}
if (n) {
dst[0] = clamp_pix((src[0] + offset0) >> shift0, pixel_max);
}
}
/* interpolate by a factor of two assuming chroma is between the luma
samples. */
static void interp2p1_simple(PIXEL *dst, const PIXEL *src, int n, int bit_depth)
{
int pixel_max, a0, a1, a2, a3, a4, a5, a6;
pixel_max = (1 << bit_depth) - 1;
a1 = src[-3];
a2 = src[-2];
a3 = src[-1];
a4 = src[0];
a5 = src[1];
a6 = src[2];
while (n >= 2) {
a0 = a1;
a1 = a2;
a2 = a3;
a3 = a4;
a4 = a5;
a5 = a6;
a6 = src[3];
dst[0] = clamp_pix((a0 * IP1C6 + a1 * IP1C5 + a2 * IP1C4 + a3 * IP1C3 +
a4 * IP1C2 + a5 * IP1C1 + a6 * IP1C0 + 32) >> 6,
pixel_max);
dst[1] = clamp_pix((a0 * IP1C0 + a1 * IP1C1 + a2 * IP1C2 + a3 * IP1C3 +
a4 * IP1C4 + a5 * IP1C5 + a6 * IP1C6 + 32) >> 6,
pixel_max);
dst += 2;
src++;
n -= 2;
}
if (n) {
a0 = a1;
a1 = a2;
a2 = a3;
a3 = a4;
a4 = a5;
a5 = a6;
a6 = src[3];
dst[0] = clamp_pix((a0 * IP1C6 + a1 * IP1C5 + a2 * IP1C4 + a3 * IP1C3 +
a4 * IP1C2 + a5 * IP1C1 + a6 * IP1C0 + 32) >> 6,
pixel_max);
}
}
static void interp2p1_simple16(PIXEL *dst, const int16_t *src, int n,
int bit_depth)
{
int shift, offset, pixel_max, a0, a1, a2, a3, a4, a5, a6;
pixel_max = (1 << bit_depth) - 1;
shift = 20 - bit_depth;
offset = 1 << (shift - 1);
a1 = src[-3];
a2 = src[-2];
a3 = src[-1];
a4 = src[0];
a5 = src[1];
a6 = src[2];
while (n >= 2) {
a0 = a1;
a1 = a2;
a2 = a3;
a3 = a4;
a4 = a5;
a5 = a6;
a6 = src[3];
dst[0] = clamp_pix((a0 * IP1C6 + a1 * IP1C5 + a2 * IP1C4 + a3 * IP1C3 +
a4 * IP1C2 + a5 * IP1C1 + a6 * IP1C0 + offset) >> shift,
pixel_max);
dst[1] = clamp_pix((a0 * IP1C0 + a1 * IP1C1 + a2 * IP1C2 + a3 * IP1C3 +
a4 * IP1C4 + a5 * IP1C5 + a6 * IP1C6 + offset) >> shift,
pixel_max);
dst += 2;
src++;
n -= 2;
}
if (n) {
a0 = a1;
a1 = a2;
a2 = a3;
a3 = a4;
a4 = a5;
a5 = a6;
a6 = src[3];
dst[0] = clamp_pix((a0 * IP1C6 + a1 * IP1C5 + a2 * IP1C4 + a3 * IP1C3 +
a4 * IP1C2 + a5 * IP1C1 + a6 * IP1C0 + offset) >> shift,
pixel_max);
}
}
/* tmp_buf is a temporary buffer of length (n2 + 2 * ITAPS2 - 1) */
static void interp2_h(PIXEL *dst, const PIXEL *src, int n, int bit_depth,
int phase, PIXEL *tmp_buf)
{
PIXEL *src1 = tmp_buf, v;
int i, n2;
/* add extra pixels and do the interpolation (XXX: could go faster) */
n2 = (n + 1) / 2;
memcpy(src1 + ITAPS2 - 1, src, n2 * sizeof(PIXEL));
v = src[0];
for(i = 0; i < ITAPS2 - 1; i++)
src1[i] = v;
v = src[n2 - 1];
for(i = 0; i < ITAPS2; i++)
src1[ITAPS2 - 1 + n2 + i] = v;
if (phase == 0)
interp2p0_simple(dst, src1 + ITAPS2 - 1, n, bit_depth);
else
interp2p1_simple(dst, src1 + ITAPS2 - 1, n, bit_depth);
}
/* y_pos is the position of the sample '0' in the 'src' circular
buffer. tmp_buf is a temporary buffer of length (n2 + 2 * ITAPS2 - 1) */
static void interp2_vh(PIXEL *dst, PIXEL **src, int n, int y_pos,
int16_t *tmp_buf, int bit_depth, int frac_pos,
int c_h_phase)
{
const PIXEL *src0, *src1, *src2, *src3, *src4, *src5, *src6;
int i, n2, shift, rnd;
int16_t v;
src0 = src[(y_pos - 3) & 7];
src1 = src[(y_pos - 2) & 7];
src2 = src[(y_pos - 1) & 7];
src3 = src[(y_pos + 0) & 7];
src4 = src[(y_pos + 1) & 7];
src5 = src[(y_pos + 2) & 7];
src6 = src[(y_pos + 3) & 7];
/* vertical interpolation first */
shift = bit_depth - 8;
rnd = (1 << shift) >> 1;
n2 = (n + 1) / 2;
if (frac_pos == 0) {
for(i = 0; i < n2; i++) {
tmp_buf[ITAPS2 - 1 + i] =
(src0[i] * IP1C6 + src1[i] * IP1C5 +
src2[i] * IP1C4 + src3[i] * IP1C3 +
src4[i] * IP1C2 + src5[i] * IP1C1 +
src6[i] * IP1C0 + rnd) >> shift;
}
} else {
for(i = 0; i < n2; i++) {
tmp_buf[ITAPS2 - 1 + i] =
(src0[i] * IP1C0 + src1[i] * IP1C1 +
src2[i] * IP1C2 + src3[i] * IP1C3 +
src4[i] * IP1C4 + src5[i] * IP1C5 +
src6[i] * IP1C6 + rnd) >> shift;
}
}
/* then horizontal interpolation */
v = tmp_buf[ITAPS2 - 1];
for(i = 0; i < ITAPS2 - 1; i++)
tmp_buf[i] = v;
v = tmp_buf[ITAPS2 - 1 + n2 - 1];
for(i = 0; i < ITAPS2; i++)
tmp_buf[ITAPS2 - 1 + n2 + i] = v;
if (c_h_phase == 0)
interp2p0_simple16(dst, tmp_buf + ITAPS2 - 1, n, bit_depth);
else
interp2p1_simple16(dst, tmp_buf + ITAPS2 - 1, n, bit_depth);
}
static void ycc_to_rgb24(ColorConvertState *s, uint8_t *dst, const PIXEL *y_ptr,
const PIXEL *cb_ptr, const PIXEL *cr_ptr,
int n, int incr)
{
uint8_t *q = dst;
int y_val, cb_val, cr_val, x;
int c_r_cr, c_g_cb, c_g_cr, c_b_cb, rnd, shift, center, c_one;
c_r_cr = s->c_r_cr;
c_g_cb = s->c_g_cb;
c_g_cr = s->c_g_cr;
c_b_cb = s->c_b_cb;
c_one = s->y_one;
rnd = s->y_offset;
shift = s->c_shift;
center = s->c_center;
for(x = 0; x < n; x++) {
y_val = y_ptr[x] * c_one;
cb_val = cb_ptr[x] - center;
cr_val = cr_ptr[x] - center;
q[0] = clamp8((y_val + c_r_cr * cr_val + rnd) >> shift);
q[1] = clamp8((y_val - c_g_cb * cb_val - c_g_cr * cr_val + rnd) >> shift);
q[2] = clamp8((y_val + c_b_cb * cb_val + rnd) >> shift);
q += incr;
}
}
static void ycgco_to_rgb24(ColorConvertState *s,
uint8_t *dst, const PIXEL *y_ptr,
const PIXEL *cb_ptr, const PIXEL *cr_ptr,
int n, int incr)
{
uint8_t *q = dst;
int y_val, cb_val, cr_val, x;
int rnd, shift, center, c_one;
c_one = s->y_one;
rnd = s->y_offset;
shift = s->c_shift;
center = s->c_center;
for(x = 0; x < n; x++) {
y_val = y_ptr[x];
cb_val = cb_ptr[x] - center;
cr_val = cr_ptr[x] - center;
q[0] = clamp8(((y_val - cb_val + cr_val) * c_one + rnd) >> shift);
q[1] = clamp8(((y_val + cb_val) * c_one + rnd) >> shift);
q[2] = clamp8(((y_val - cb_val - cr_val) * c_one + rnd) >> shift);
q += incr;
}
}
/* c = c * alpha */
static void alpha_combine8(ColorConvertState *s,
uint8_t *dst, const PIXEL *a_ptr, int n, int incr)
{
uint8_t *q = dst;
int x, a_val, shift, rnd;
shift = s->bit_depth;
rnd = 1 << (shift - 1);
for(x = 0; x < n; x++) {
a_val = a_ptr[x];
/* XXX: not accurate enough */
q[0] = (q[0] * a_val + rnd) >> shift;
q[1] = (q[1] * a_val + rnd) >> shift;
q[2] = (q[2] * a_val + rnd) >> shift;
q += incr;
}
}
static uint32_t divide8_table[256];
#define DIV8_BITS 16
static void alpha_divide8_init(void)
{
int i;
for(i = 1; i < 256; i++) {
/* Note: the 128 is added to have 100% correct results for all
the values */
divide8_table[i] = ((255 << DIV8_BITS) + (i / 2) + 128) / i;
}
}
static inline unsigned int comp_divide8(unsigned int val, unsigned int alpha,
unsigned int alpha_inv)
{
if (val >= alpha)
return 255;
return (val * alpha_inv + (1 << (DIV8_BITS - 1))) >> DIV8_BITS;
}
/* c = c / alpha */
static void alpha_divide8(uint8_t *dst, int n)
{
static int inited;
uint8_t *q = dst;
int x;
unsigned int a_val, a_inv;
if (!inited) {
inited = 1;
alpha_divide8_init();
}
for(x = 0; x < n; x++) {
a_val = q[3];
if (a_val == 0) {
q[0] = 255;
q[1] = 255;
q[2] = 255;
} else {
a_inv = divide8_table[a_val];
q[0] = comp_divide8(q[0], a_val, a_inv);
q[1] = comp_divide8(q[1], a_val, a_inv);
q[2] = comp_divide8(q[2], a_val, a_inv);
}
q += 4;
}
}
static void gray_to_rgb24(ColorConvertState *s,
uint8_t *dst, const PIXEL *y_ptr,
const PIXEL *cb_ptr, const PIXEL *cr_ptr,
int n, int incr)
{
uint8_t *q = dst;
int x, y_val, c, rnd, shift;
if (s->bit_depth == 8 && !s->limited_range) {
for(x = 0; x < n; x++) {
y_val = y_ptr[x];
q[0] = y_val;
q[1] = y_val;
q[2] = y_val;
q += incr;
}
} else {
c = s->y_one;