summaryrefslogtreecommitdiffstats
path: root/libavcodec/mpegvideo.h
blob: ac077b3269c00785a9d0cd146edd693d9f1e43ee (plain)
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
/*
 * Generic DCT based hybrid video encoder
 * Copyright (c) 2000, 2001, 2002 Fabrice Bellard
 * Copyright (c) 2002-2004 Michael Niedermayer
 *
 * This file is part of Libav.
 *
 * Libav is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * Libav 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
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with Libav; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * mpegvideo header.
 */

#ifndef AVCODEC_MPEGVIDEO_H
#define AVCODEC_MPEGVIDEO_H

#include <float.h>

#include "avcodec.h"
#include "blockdsp.h"
#include "error_resilience.h"
#include "fdctdsp.h"
#include "get_bits.h"
#include "h263dsp.h"
#include "hpeldsp.h"
#include "idctdsp.h"
#include "me_cmp.h"
#include "mpegvideodsp.h"
#include "mpegvideoencdsp.h"
#include "pixblockdsp.h"
#include "put_bits.h"
#include "ratecontrol.h"
#include "parser.h"
#include "mpeg12data.h"
#include "qpeldsp.h"
#include "rl.h"
#include "thread.h"
#include "videodsp.h"

#include "libavutil/opt.h"

#define FRAME_SKIPPED 100 ///< return value for header parsers if frame is not coded

enum OutputFormat {
    FMT_MPEG1,
    FMT_H261,
    FMT_H263,
    FMT_MJPEG,
};

#define MAX_FCODE 7
#define MAX_MV 2048

#define MAX_THREADS 16

#define MAX_PICTURE_COUNT 32

#define MAX_B_FRAMES 16

#define ME_MAP_SIZE 64

#define MAX_MB_BYTES (30*16*16*3/8 + 120)

#define INPLACE_OFFSET 16

#define EDGE_WIDTH 16

/* Start codes. */
#define SEQ_END_CODE            0x000001b7
#define SEQ_START_CODE          0x000001b3
#define GOP_START_CODE          0x000001b8
#define PICTURE_START_CODE      0x00000100
#define SLICE_MIN_START_CODE    0x00000101
#define SLICE_MAX_START_CODE    0x000001af
#define EXT_START_CODE          0x000001b5
#define USER_START_CODE         0x000001b2

/* encoding scans */
extern const uint8_t ff_alternate_horizontal_scan[64];
extern const uint8_t ff_alternate_vertical_scan[64];

struct MpegEncContext;

/**
 * Picture.
 */
typedef struct Picture{
    struct AVFrame *f;
    ThreadFrame tf;

    AVBufferRef *qscale_table_buf;
    int8_t *qscale_table;

    AVBufferRef *motion_val_buf[2];
    int16_t (*motion_val[2])[2];

    AVBufferRef *mb_type_buf;
    uint32_t *mb_type;          ///< types and macros are defined in mpegutils.h

    AVBufferRef *mbskip_table_buf;
    uint8_t *mbskip_table;

    AVBufferRef *ref_index_buf[2];
    int8_t *ref_index[2];

    AVBufferRef *mb_var_buf;
    uint16_t *mb_var;           ///< Table for MB variances

    AVBufferRef *mc_mb_var_buf;
    uint16_t *mc_mb_var;        ///< Table for motion compensated MB variances

    AVBufferRef *mb_mean_buf;
    uint8_t *mb_mean;           ///< Table for MB luminance

    AVBufferRef *hwaccel_priv_buf;
    /**
     * hardware accelerator private data
     */
    void *hwaccel_picture_private;

    int field_picture;          ///< whether or not the picture was encoded in separate fields

    int mb_var_sum;             ///< sum of MB variance for current frame
    int mc_mb_var_sum;          ///< motion compensated MB variance for current frame

    int b_frame_score;          /* */
    int needs_realloc;          ///< Picture needs to be reallocated (eg due to a frame size change)

    int reference;
    int shared;
} Picture;

/**
 * Motion estimation context.
 */
typedef struct MotionEstContext{
    AVCodecContext *avctx;
    int skip;                          ///< set if ME is skipped for the current MB
    int co_located_mv[4][2];           ///< mv from last P-frame for direct mode ME
    int direct_basis_mv[4][2];
    uint8_t *scratchpad;               ///< data area for the ME algo, so that the ME does not need to malloc/free
    uint8_t *best_mb;
    uint8_t *temp_mb[2];
    uint8_t *temp;
    int best_bits;
    uint32_t *map;                     ///< map to avoid duplicate evaluations
    uint32_t *score_map;               ///< map to store the scores
    unsigned map_generation;
    int pre_penalty_factor;
    int penalty_factor;                /**< an estimate of the bits required to
                                        code a given mv value, e.g. (1,0) takes
                                        more bits than (0,0). We have to
                                        estimate whether any reduction in
                                        residual is worth the extra bits. */
    int sub_penalty_factor;
    int mb_penalty_factor;
    int flags;
    int sub_flags;
    int mb_flags;
    int pre_pass;                      ///< = 1 for the pre pass
    int dia_size;
    int xmin;
    int xmax;
    int ymin;
    int ymax;
    int pred_x;
    int pred_y;
    uint8_t *src[4][4];
    uint8_t *ref[4][4];
    int stride;
    int uvstride;
    /* temp variables for picture complexity calculation */
    int mc_mb_var_sum_temp;
    int mb_var_sum_temp;
    int scene_change_score;
/*    cmp, chroma_cmp;*/
    op_pixels_func (*hpel_put)[4];
    op_pixels_func (*hpel_avg)[4];
    qpel_mc_func (*qpel_put)[16];
    qpel_mc_func (*qpel_avg)[16];
    uint8_t (*mv_penalty)[MAX_MV*2+1];  ///< amount of bits needed to encode a MV
    uint8_t *current_mv_penalty;
    int (*sub_motion_search)(struct MpegEncContext * s,
                                  int *mx_ptr, int *my_ptr, int dmin,
                                  int src_index, int ref_index,
                                  int size, int h);
}MotionEstContext;

/**
 * MpegEncContext.
 */
typedef struct MpegEncContext {
    AVClass *class;

    int y_dc_scale, c_dc_scale;
    int ac_pred;
    int block_last_index[12];  ///< last non zero coefficient in block
    int h263_aic;              ///< Advanded INTRA Coding (AIC)

    /* scantables */
    ScanTable inter_scantable; ///< if inter == intra then intra should be used to reduce tha cache usage
    ScanTable intra_scantable;
    ScanTable intra_h_scantable;
    ScanTable intra_v_scantable;

    /* WARNING: changes above this line require updates to hardcoded
     *          offsets used in asm. */

    struct AVCodecContext *avctx;
    /* the following parameters must be initialized before encoding */
    int width, height;///< picture size. must be a multiple of 16
    int gop_size;
    int intra_only;   ///< if true, only intra pictures are generated
    int bit_rate;     ///< wanted bit rate
    enum OutputFormat out_format; ///< output format
    int h263_pred;    ///< use mpeg4/h263 ac/dc predictions
    int pb_frame;     ///< PB frame mode (0 = none, 1 = base, 2 = improved)

/* the following codec id fields are deprecated in favor of codec_id */
    int h263_plus;    ///< h263 plus headers
    int h263_flv;     ///< use flv h263 header

    enum AVCodecID codec_id;     /* see AV_CODEC_ID_xxx */
    int fixed_qscale; ///< fixed qscale if non zero
    int encoding;     ///< true if we are encoding (vs decoding)
    int flags;        ///< AVCodecContext.flags (HQ, MV4, ...)
    int flags2;       ///< AVCodecContext.flags2
    int max_b_frames; ///< max number of b-frames for encoding
    int luma_elim_threshold;
    int chroma_elim_threshold;
    int strict_std_compliance; ///< strictly follow the std (MPEG4, ...)
    int workaround_bugs;       ///< workaround bugs in encoders which cannot be detected automatically
    int codec_tag;             ///< internal codec_tag upper case converted from avctx codec_tag
    int stream_codec_tag;      ///< internal stream_codec_tag upper case converted from avctx stream_codec_tag
    /* the following fields are managed internally by the encoder */

    /* sequence parameters */
    int context_initialized;
    int input_picture_number;  ///< used to set pic->display_picture_number, should not be used for/by anything else
    int coded_picture_number;  ///< used to set pic->coded_picture_number, should not be used for/by anything else
    int picture_number;       //FIXME remove, unclear definition
    int picture_in_gop_number; ///< 0-> first pic in gop, ...
    int mb_width, mb_height;   ///< number of MBs horizontally & vertically
    int mb_stride;             ///< mb_width+1 used for some arrays to allow simple addressing of left & top MBs without sig11
    int b8_stride;             ///< 2*mb_width+1 used for some 8x8 block arrays to allow simple addressing
    int h_edge_pos, v_edge_pos;///< horizontal / vertical position of the right/bottom edge (pixel replication)
    int mb_num;                ///< number of MBs of a picture
    ptrdiff_t linesize;        ///< line size, in bytes, may be different from width
    ptrdiff_t uvlinesize;      ///< line size, for chroma in bytes, may be different from width
    Picture *picture;          ///< main picture buffer
    Picture **input_picture;   ///< next pictures on display order for encoding
    Picture **reordered_input_picture; ///< pointer to the next pictures in codedorder for encoding

    int64_t user_specified_pts; ///< last non-zero pts from AVFrame which was passed into avcodec_encode_video2()
    /**
     * pts difference between the first and second input frame, used for
     * calculating dts of the first frame when there's a delay */
    int64_t dts_delta;
    /**
     * reordered pts to be used as dts for the next output frame when there's
     * a delay */
    int64_t reordered_pts;

    /** bit output */
    PutBitContext pb;

    int start_mb_y;            ///< start mb_y of this thread (so current thread should process start_mb_y <= row < end_mb_y)
    int end_mb_y;              ///< end   mb_y of this thread (so current thread should process start_mb_y <= row < end_mb_y)
    struct MpegEncContext *thread_context[MAX_THREADS];
    int slice_context_count;   ///< number of used thread_contexts

    /**
     * copy of the previous picture structure.
     * note, linesize & data, might not match the previous picture (for field pictures)
     */
    Picture last_picture;

    /**
     * copy of the next picture structure.
     * note, linesize & data, might not match the next picture (for field pictures)
     */
    Picture next_picture;

    /**
     * copy of the source picture structure for encoding.
     * note, linesize & data, might not match the source picture (for field pictures)
     */
    Picture new_picture;

    /**
     * copy of the current picture structure.
     * note, linesize & data, might not match the current picture (for field pictures)
     */
    Picture current_picture;    ///< buffer to store the decompressed current picture

    Picture *last_picture_ptr;     ///< pointer to the previous picture.
    Picture *next_picture_ptr;     ///< pointer to the next picture (for bidir pred)
    Picture *current_picture_ptr;  ///< pointer to the current picture
    int last_dc[3];                ///< last DC values for MPEG1
    int16_t *dc_val_base;
    int16_t *dc_val[3];            ///< used for mpeg4 DC prediction, all 3 arrays must be continuous
    const uint8_t *y_dc_scale_table;     ///< qscale -> y_dc_scale table
    const uint8_t *c_dc_scale_table;     ///< qscale -> c_dc_scale table
    const uint8_t *chroma_qscale_table;  ///< qscale -> chroma_qscale (h263)
    uint8_t *coded_block_base;
    uint8_t *coded_block;          ///< used for coded block pattern prediction (msmpeg4v3, wmv1)
    int16_t (*ac_val_base)[16];
    int16_t (*ac_val[3])[16];      ///< used for for mpeg4 AC prediction, all 3 arrays must be continuous
    int mb_skipped;                ///< MUST BE SET only during DECODING
    uint8_t *mbskip_table;        /**< used to avoid copy if macroblock skipped (for black regions for example)
                                   and used for b-frame encoding & decoding (contains skip table of next P Frame) */
    uint8_t *mbintra_table;       ///< used to avoid setting {ac, dc, cbp}-pred stuff to zero on inter MB decoding
    uint8_t *cbp_table;           ///< used to store cbp, ac_pred for partitioned decoding
    uint8_t *pred_dir_table;      ///< used to store pred_dir for partitioned decoding
    uint8_t *edge_emu_buffer;     ///< temporary buffer for if MVs point to out-of-frame data
    uint8_t *rd_scratchpad;       ///< scratchpad for rate distortion mb decision
    uint8_t *obmc_scratchpad;
    uint8_t *b_scratchpad;        ///< scratchpad used for writing into write only buffers

    int qscale;                 ///< QP
    int chroma_qscale;          ///< chroma QP
    unsigned int lambda;        ///< lagrange multipler used in rate distortion
    unsigned int lambda2;       ///< (lambda*lambda) >> FF_LAMBDA_SHIFT
    int *lambda_table;
    int adaptive_quant;         ///< use adaptive quantization
    int dquant;                 ///< qscale difference to prev qscale
    int pict_type;              ///< AV_PICTURE_TYPE_I, AV_PICTURE_TYPE_P, AV_PICTURE_TYPE_B, ...
    int last_pict_type; //FIXME removes
    int last_non_b_pict_type;   ///< used for mpeg4 gmc b-frames & ratecontrol
    int droppable;
    int frame_rate_index;
    int last_lambda_for[5];     ///< last lambda for a specific pict type
    int skipdct;                ///< skip dct and code zero residual

    /* motion compensation */
    int unrestricted_mv;        ///< mv can point outside of the coded picture
    int h263_long_vectors;      ///< use horrible h263v1 long vector mode

    BlockDSPContext bdsp;
    FDCTDSPContext fdsp;
    HpelDSPContext hdsp;
    IDCTDSPContext idsp;
    MECmpContext mecc;
    MpegVideoDSPContext mdsp;
    MpegvideoEncDSPContext mpvencdsp;
    PixblockDSPContext pdsp;
    QpelDSPContext qdsp;
    VideoDSPContext vdsp;
    H263DSPContext h263dsp;
    int f_code;                 ///< forward MV resolution
    int b_code;                 ///< backward MV resolution for B Frames (mpeg4)
    int16_t (*p_mv_table_base)[2];
    int16_t (*b_forw_mv_table_base)[2];
    int16_t (*b_back_mv_table_base)[2];
    int16_t (*b_bidir_forw_mv_table_base)[2];
    int16_t (*b_bidir_back_mv_table_base)[2];
    int16_t (*b_direct_mv_table_base)[2];
    int16_t (*p_field_mv_table_base[2][2])[2];
    int16_t (*b_field_mv_table_base[2][2][2])[2];
    int16_t (*p_mv_table)[2];            ///< MV table (1MV per MB) p-frame encoding
    int16_t (*b_forw_mv_table)[2];       ///< MV table (1MV per MB) forward mode b-frame encoding
    int16_t (*b_back_mv_table)[2];       ///< MV table (1MV per MB) backward mode b-frame encoding
    int16_t (*b_bidir_forw_mv_table)[2]; ///< MV table (1MV per MB) bidir mode b-frame encoding
    int16_t (*b_bidir_back_mv_table)[2]; ///< MV table (1MV per MB) bidir mode b-frame encoding
    int16_t (*b_direct_mv_table)[2];     ///< MV table (1MV per MB) direct mode b-frame encoding
    int16_t (*p_field_mv_table[2][2])[2];   ///< MV table (2MV per MB) interlaced p-frame encoding
    int16_t (*b_field_mv_table[2][2][2])[2];///< MV table (4MV per MB) interlaced b-frame encoding
    uint8_t (*p_field_select_table[2]);
    uint8_t (*b_field_select_table[2][2]);
    int me_method;                       ///< ME algorithm
    int mv_dir;
#define MV_DIR_FORWARD   1
#define MV_DIR_BACKWARD  2
#define MV_DIRECT        4 ///< bidirectional mode where the difference equals the MV of the last P/S/I-Frame (mpeg4)
    int mv_type;
#define MV_TYPE_16X16       0   ///< 1 vector for the whole mb
#define MV_TYPE_8X8         1   ///< 4 vectors (h263, mpeg4 4MV)
#define MV_TYPE_16X8        2   ///< 2 vectors, one per 16x8 block
#define MV_TYPE_FIELD       3   ///< 2 vectors, one per field
#define MV_TYPE_DMV         4   ///< 2 vectors, special mpeg2 Dual Prime Vectors
    /**motion vectors for a macroblock
       first coordinate : 0 = forward 1 = backward
       second "         : depend on type
       third  "         : 0 = x, 1 = y
    */
    int mv[2][4][2];
    int field_select[2][2];
    int last_mv[2][2][2];             ///< last MV, used for MV prediction in MPEG1 & B-frame MPEG4
    uint8_t *fcode_tab;               ///< smallest fcode needed for each MV
    int16_t direct_scale_mv[2][64];   ///< precomputed to avoid divisions in ff_mpeg4_set_direct_mv

    MotionEstContext me;

    int no_rounding;  /**< apply no rounding to motion compensation (MPEG4, msmpeg4, ...)
                        for b-frames rounding mode is always 0 */

    /* macroblock layer */
    int mb_x, mb_y;
    int mb_skip_run;
    int mb_intra;
    uint16_t *mb_type;  ///< Table for candidate MB types for encoding (defines in mpegutils.h)

    int block_index[6]; ///< index to current MB in block based arrays with edges
    int block_wrap[6];
    uint8_t *dest[3];

    int *mb_index2xy;        ///< mb_index -> mb_x + mb_y*mb_stride

    /** matrix transmitted in the bitstream */
    uint16_t intra_matrix[64];
    uint16_t chroma_intra_matrix[64];
    uint16_t inter_matrix[64];
    uint16_t chroma_inter_matrix[64];

    int intra_quant_bias;    ///< bias for the quantizer
    int inter_quant_bias;    ///< bias for the quantizer
    int min_qcoeff;          ///< minimum encodable coefficient
    int max_qcoeff;          ///< maximum encodable coefficient
    int ac_esc_length;       ///< num of bits needed to encode the longest esc
    uint8_t *intra_ac_vlc_length;
    uint8_t *intra_ac_vlc_last_length;
    uint8_t *inter_ac_vlc_length;
    uint8_t *inter_ac_vlc_last_length;
    uint8_t *luma_dc_vlc_length;
#define UNI_AC_ENC_INDEX(run,level) ((run)*128 + (level))

    int coded_score[8];

    /** precomputed matrix (combine qscale and DCT renorm) */
    int (*q_intra_matrix)[64];
    int (*q_inter_matrix)[64];
    /** identical to the above but for MMX & these are not permutated, second 64 entries are bias*/
    uint16_t (*q_intra_matrix16)[2][64];
    uint16_t (*q_inter_matrix16)[2][64];

    /* noise reduction */
    int (*dct_error_sum)[64];
    int dct_count[2];
    uint16_t (*dct_offset)[64];

    /* bit rate control */
    int64_t total_bits;
    int frame_bits;                ///< bits used for the current frame
    int next_lambda;               ///< next lambda used for retrying to encode a frame
    RateControlContext rc_context; ///< contains stuff only accessed in ratecontrol.c

    /* statistics, used for 2-pass encoding */
    int mv_bits;
    int header_bits;
    int i_tex_bits;
    int p_tex_bits;
    int i_count;
    int f_count;
    int b_count;
    int skip_count;
    int misc_bits; ///< cbp, mb_type
    int last_bits; ///< temp var used for calculating the above vars

    /* error concealment / resync */
    int resync_mb_x;                 ///< x position of last resync marker
    int resync_mb_y;                 ///< y position of last resync marker
    GetBitContext last_resync_gb;    ///< used to search for the next resync marker
    int mb_num_left;                 ///< number of MBs left in this video packet (for partitioned Slices only)
    int next_p_frame_damaged;        ///< set if the next p frame is damaged, to avoid showing trashed b frames
    int err_recognition;

    ParseContext parse_context;

    /* H.263 specific */
    int gob_index;
    int obmc;                       ///< overlapped block motion compensation
    int mb_info;                    ///< interval for outputting info about mb offsets as side data
    int prev_mb_info, last_mb_info;
    uint8_t *mb_info_ptr;
    int mb_info_size;

    /* H.263+ specific */
    int umvplus;                    ///< == H263+ && unrestricted_mv
    int h263_aic_dir;               ///< AIC direction: 0 = left, 1 = top
    int h263_slice_structured;
    int alt_inter_vlc;              ///< alternative inter vlc
    int modified_quant;
    int loop_filter;
    int custom_pcf;

    /* mpeg4 specific */
    ///< number of bits to represent the fractional part of time (encoder only)
    int time_increment_bits;
    int last_time_base;
    int time_base;                  ///< time in seconds of last I,P,S Frame
    int64_t time;                   ///< time of current frame
    int64_t last_non_b_time;
    uint16_t pp_time;               ///< time distance between the last 2 p,s,i frames
    uint16_t pb_time;               ///< time distance between the last b and p,s,i frame
    uint16_t pp_field_time;
    uint16_t pb_field_time;         ///< like above, just for interlaced
    int real_sprite_warping_points;
    int sprite_offset[2][2];         ///< sprite offset[isChroma][isMVY]
    int sprite_delta[2][2];          ///< sprite_delta [isY][isMVY]
    int mcsel;
    int quant_precision;
    int quarter_sample;              ///< 1->qpel, 0->half pel ME/MC
    int aspect_ratio_info; //FIXME remove
    int sprite_warping_accuracy;
    int data_partitioning;           ///< data partitioning flag from header
    int partitioned_frame;           ///< is current frame partitioned
    int low_delay;                   ///< no reordering needed / has no b-frames
    int vo_type;
    PutBitContext tex_pb;            ///< used for data partitioned VOPs
    PutBitContext pb2;               ///< used for data partitioned VOPs
    int mpeg_quant;
    int padding_bug_score;             ///< used to detect the VERY common padding bug in MPEG4

    /* divx specific, used to workaround (many) bugs in divx5 */
    int divx_packed;
    uint8_t *bitstream_buffer; //Divx 5.01 puts several frames in a single one, this is used to reorder them
    int bitstream_buffer_size;
    unsigned int allocated_bitstream_buffer_size;

    /* RV10 specific */
    int rv10_version; ///< RV10 version: 0 or 3
    int rv10_first_dc_coded[3];

    /* MJPEG specific */
    struct MJpegContext *mjpeg_ctx;

    /* MSMPEG4 specific */
    int mv_table_index;
    int rl_table_index;
    int rl_chroma_table_index;
    int dc_table_index;
    int use_skip_mb_code;
    int slice_height;      ///< in macroblocks
    int first_slice_line;  ///< used in mpeg4 too to handle resync markers
    int flipflop_rounding;
    int msmpeg4_version;   ///< 0=not msmpeg4, 1=mp41, 2=mp42, 3=mp43/divx3 4=wmv1/7 5=wmv2/8
    int per_mb_rl_table;
    int esc3_level_length;
    int esc3_run_length;
    /** [mb_intra][isChroma][level][run][last] */
    int (*ac_stats)[2][MAX_LEVEL+1][MAX_RUN+1][2];
    int inter_intra_pred;
    int mspel;

    /* decompression specific */
    GetBitContext gb;

    /* Mpeg1 specific */
    int gop_picture_number;  ///< index of the first picture of a GOP based on fake_pic_num & mpeg1 specific
    int last_mv_dir;         ///< last mv_dir, used for b frame encoding
    uint8_t *vbv_delay_ptr;  ///< pointer to vbv_delay in the bitstream

    /* MPEG-2-specific - I wished not to have to support this mess. */
    int progressive_sequence;
    int mpeg_f_code[2][2];

    // picture structure defines are loaded from mpegutils.h
    int picture_structure;

    int intra_dc_precision;
    int frame_pred_frame_dct;
    int top_field_first;
    int concealment_motion_vectors;
    int q_scale_type;
    int intra_vlc_format;
    int alternate_scan;
    int repeat_first_field;
    int chroma_420_type;
    int chroma_format;
#define CHROMA_420 1
#define CHROMA_422 2
    int chroma_x_shift;//depend on pix_format, that depend on chroma_format
    int chroma_y_shift;

    int progressive_frame;
    int full_pel[2];
    int interlaced_dct;
    int first_field;         ///< is 1 for the first field of a field picture 0 otherwise
    int drop_frame_timecode; ///< timecode is in drop frame format.
    int scan_offset;         ///< reserve space for SVCD scan offset user data.

    /* RTP specific */
    int rtp_mode;

    uint8_t *ptr_lastgob;
    int16_t (*pblocks[12])[64];

    int16_t (*block)[64]; ///< points to one of the following blocks
    int16_t (*blocks)[12][64]; // for HQ mode we need to keep the best block
    int (*decode_mb)(struct MpegEncContext *s, int16_t block[6][64]); // used by some codecs to avoid a switch()
#define SLICE_OK         0
#define SLICE_ERROR     -1
#define SLICE_END       -2 ///<end marker found
#define SLICE_NOEND     -3 ///<no end marker or error found but mb count exceeded

    void (*dct_unquantize_mpeg1_intra)(struct MpegEncContext *s,
                           int16_t *block/*align 16*/, int n, int qscale);
    void (*dct_unquantize_mpeg1_inter)(struct MpegEncContext *s,
                           int16_t *block/*align 16*/, int n, int qscale);
    void (*dct_unquantize_mpeg2_intra)(struct MpegEncContext *s,
                           int16_t *block/*align 16*/, int n, int qscale);
    void (*dct_unquantize_mpeg2_inter)(struct MpegEncContext *s,
                           int16_t *block/*align 16*/, int n, int qscale);
    void (*dct_unquantize_h263_intra)(struct MpegEncContext *s,
                           int16_t *block/*align 16*/, int n, int qscale);
    void (*dct_unquantize_h263_inter)(struct MpegEncContext *s,
                           int16_t *block/*align 16*/, int n, int qscale);
    void (*dct_unquantize_intra)(struct MpegEncContext *s, // unquantizer to use (mpeg4 can use both)
                           int16_t *block/*align 16*/, int n, int qscale);
    void (*dct_unquantize_inter)(struct MpegEncContext *s, // unquantizer to use (mpeg4 can use both)
                           int16_t *block/*align 16*/, int n, int qscale);
    int (*dct_quantize)(struct MpegEncContext *s, int16_t *block/*align 16*/, int n, int qscale, int *overflow);
    int (*fast_dct_quantize)(struct MpegEncContext *s, int16_t *block/*align 16*/, int n, int qscale, int *overflow);
    void (*denoise_dct)(struct MpegEncContext *s, int16_t *block);

    int mpv_flags;      ///< flags set by private options
    int quantizer_noise_shaping;

    /**
     * ratecontrol qmin qmax limiting method
     * 0-> clipping, 1-> use a nice continuous function to limit qscale wthin qmin/qmax.
     */
    float rc_qsquish;
    float rc_qmod_amp;
    int   rc_qmod_freq;
    float rc_initial_cplx;
    float rc_buffer_aggressivity;
    float border_masking;
    int lmin, lmax;

    char *rc_eq;

    /* temp buffers for rate control */
    float *cplx_tab, *bits_tab;

    /* flag to indicate a reinitialization is required, e.g. after
     * a frame size change */
    int context_reinit;

    ERContext er;

    int error_rate;

    /* temporary frames used by b_frame_strategy = 2 */
    AVFrame *tmp_frames[MAX_B_FRAMES + 2];
} MpegEncContext;

/* mpegvideo_enc common options */
#define FF_MPV_FLAG_SKIP_RD      0x0001
#define FF_MPV_FLAG_STRICT_GOP   0x0002
#define FF_MPV_FLAG_QP_RD        0x0004
#define FF_MPV_FLAG_CBP_RD       0x0008
#define FF_MPV_FLAG_NAQ          0x0010
#define FF_MPV_FLAG_MV0          0x0020

#define FF_MPV_OFFSET(x) offsetof(MpegEncContext, x)
#define FF_MPV_OPT_FLAGS (AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_ENCODING_PARAM)
#define FF_MPV_COMMON_OPTS \
{ "mpv_flags",      "Flags common for all mpegvideo-based encoders.", FF_MPV_OFFSET(mpv_flags), AV_OPT_TYPE_FLAGS, { .i64 = 0 }, INT_MIN, INT_MAX, FF_MPV_OPT_FLAGS, "mpv_flags" },\
{ "skip_rd",        "RD optimal MB level residual skipping", 0, AV_OPT_TYPE_CONST, { .i64 = FF_MPV_FLAG_SKIP_RD },    0, 0, FF_MPV_OPT_FLAGS, "mpv_flags" },\
{ "strict_gop",     "Strictly enforce gop size",             0, AV_OPT_TYPE_CONST, { .i64 = FF_MPV_FLAG_STRICT_GOP }, 0, 0, FF_MPV_OPT_FLAGS, "mpv_flags" },\
{ "qp_rd",          "Use rate distortion optimization for qp selection", 0, AV_OPT_TYPE_CONST, { .i64 = FF_MPV_FLAG_QP_RD },  0, 0, FF_MPV_OPT_FLAGS, "mpv_flags" },\
{ "cbp_rd",         "use rate distortion optimization for CBP",          0, AV_OPT_TYPE_CONST, { .i64 = FF_MPV_FLAG_CBP_RD }, 0, 0, FF_MPV_OPT_FLAGS, "mpv_flags" },\
{ "naq",            "normalize adaptive quantization",                   0, AV_OPT_TYPE_CONST, { .i64 = FF_MPV_FLAG_NAQ },    0, 0, FF_MPV_OPT_FLAGS, "mpv_flags" },\
{ "mv0",            "always try a mb with mv=<0,0>",                     0, AV_OPT_TYPE_CONST, { .i64 = FF_MPV_FLAG_MV0 },    0, 0, FF_MPV_OPT_FLAGS, "mpv_flags" },\
{ "luma_elim_threshold",   "single coefficient elimination threshold for luminance (negative values also consider dc coefficient)",\
                                                                      FF_MPV_OFFSET(luma_elim_threshold), AV_OPT_TYPE_INT, { .i64 = 0 }, INT_MIN, INT_MAX, FF_MPV_OPT_FLAGS },\
{ "chroma_elim_threshold", "single coefficient elimination threshold for chrominance (negative values also consider dc coefficient)",\
                                                                      FF_MPV_OFFSET(chroma_elim_threshold), AV_OPT_TYPE_INT, { .i64 = 0 }, INT_MIN, INT_MAX, FF_MPV_OPT_FLAGS },\
{ "quantizer_noise_shaping", NULL,                                  FF_MPV_OFFSET(quantizer_noise_shaping), AV_OPT_TYPE_INT, { .i64 = 0 },       0, INT_MAX, FF_MPV_OPT_FLAGS },\
{ "error_rate", "Simulate errors in the bitstream to test error concealment.",                                                                                                  \
                                                                    FF_MPV_OFFSET(error_rate),              AV_OPT_TYPE_INT, { .i64 = 0 },       0, INT_MAX, FF_MPV_OPT_FLAGS },\
{"qsquish", "how to keep quantizer between qmin and qmax (0 = clip, 1 = use differentiable function)",                                                                          \
                                                                    FF_MPV_OFFSET(rc_qsquish), AV_OPT_TYPE_FLOAT, {.dbl = 0 }, 0, 99, FF_MPV_OPT_FLAGS},                        \
{"rc_qmod_amp", "experimental quantizer modulation",                FF_MPV_OFFSET(rc_qmod_amp), AV_OPT_TYPE_FLOAT, {.dbl = 0 }, -FLT_MAX, FLT_MAX, FF_MPV_OPT_FLAGS},           \
{"rc_qmod_freq", "experimental quantizer modulation",               FF_MPV_OFFSET(rc_qmod_freq), AV_OPT_TYPE_INT, {.i64 = 0 }, INT_MIN, INT_MAX, FF_MPV_OPT_FLAGS},             \
{"rc_eq", "Set rate control equation. When computing the expression, besides the standard functions "                                                                           \
          "defined in the section 'Expression Evaluation', the following functions are available: "                                                                             \
          "bits2qp(bits), qp2bits(qp). Also the following constants are available: iTex pTex tex mv "                                                                           \
          "fCode iCount mcVar var isI isP isB avgQP qComp avgIITex avgPITex avgPPTex avgBPTex avgTex.",                                                                         \
                                                                    FF_MPV_OFFSET(rc_eq), AV_OPT_TYPE_STRING,                           .flags = FF_MPV_OPT_FLAGS },            \
{"rc_init_cplx", "initial complexity for 1-pass encoding",          FF_MPV_OFFSET(rc_initial_cplx), AV_OPT_TYPE_FLOAT, {.dbl = 0 }, -FLT_MAX, FLT_MAX, FF_MPV_OPT_FLAGS},       \
{"rc_buf_aggressivity", "currently useless",                        FF_MPV_OFFSET(rc_buffer_aggressivity), AV_OPT_TYPE_FLOAT, {.dbl = 1.0 }, -FLT_MAX, FLT_MAX, FF_MPV_OPT_FLAGS}, \
{"border_mask", "increase the quantizer for macroblocks close to borders", FF_MPV_OFFSET(border_masking), AV_OPT_TYPE_FLOAT, {.dbl = 0 }, -FLT_MAX, FLT_MAX, FF_MPV_OPT_FLAGS},    \
{"lmin", "minimum Lagrange factor (VBR)",                           FF_MPV_OFFSET(lmin), AV_OPT_TYPE_INT, {.i64 =  2*FF_QP2LAMBDA }, 0, INT_MAX, FF_MPV_OPT_FLAGS },            \
{"lmax", "maximum Lagrange factor (VBR)",                           FF_MPV_OFFSET(lmax), AV_OPT_TYPE_INT, {.i64 = 31*FF_QP2LAMBDA }, 0, INT_MAX, FF_MPV_OPT_FLAGS },            \

extern const AVOption ff_mpv_generic_options[];

#define FF_MPV_GENERIC_CLASS(name) \
static const AVClass name ## _class = {\
    .class_name = #name " encoder",\
    .item_name  = av_default_item_name,\
    .option     = ff_mpv_generic_options,\
    .version    = LIBAVUTIL_VERSION_INT,\
};

/**
 * Set the given MpegEncContext to common defaults (same for encoding
 * and decoding).  The changed fields will not depend upon the prior
 * state of the MpegEncContext.
 */
void ff_mpv_common_defaults(MpegEncContext *s);

int ff_mpv_common_init(MpegEncContext *s);
void ff_mpv_common_init_arm(MpegEncContext *s);
void ff_mpv_common_init_neon(MpegEncContext *s);
void ff_mpv_common_init_ppc(MpegEncContext *s);
void ff_mpv_common_init_x86(MpegEncContext *s);

int ff_mpv_common_frame_size_change(MpegEncContext *s);
void ff_mpv_common_end(MpegEncContext *s);

void ff_mpv_decode_defaults(MpegEncContext *s);
void ff_mpv_decode_mb(MpegEncContext *s, int16_t block[12][64]);
void ff_mpv_report_decode_progress(MpegEncContext *s);

int ff_mpv_frame_start(MpegEncContext *s, AVCodecContext *avctx);
void ff_mpv_frame_end(MpegEncContext *s);

int ff_mpv_lowest_referenced_row(MpegEncContext *s, int dir);

int ff_mpv_encode_init(AVCodecContext *avctx);
void ff_mpv_encode_init_x86(MpegEncContext *s);

int ff_mpv_encode_end(AVCodecContext *avctx);
int ff_mpv_encode_picture(AVCodecContext *avctx, AVPacket *pkt,
                          const AVFrame *frame, int *got_packet);

void ff_clean_intra_table_entries(MpegEncContext *s);
void ff_mpeg_draw_horiz_band(MpegEncContext *s, int y, int h);
void ff_mpeg_flush(AVCodecContext *avctx);
void ff_print_debug_info(MpegEncContext *s, Picture *p);
void ff_write_quant_matrix(PutBitContext *pb, uint16_t *matrix);
int ff_find_unused_picture(MpegEncContext *s, int shared);
int ff_update_duplicate_context(MpegEncContext *dst, MpegEncContext *src);
int ff_mpeg_update_thread_context(AVCodecContext *dst, const AVCodecContext *src);
void ff_set_qscale(MpegEncContext * s, int qscale);

void ff_mpv_idct_init(MpegEncContext *s);
void ff_convert_matrix(MpegEncContext *s, int (*qmat)[64], uint16_t (*qmat16)[2][64],
                       const uint16_t *quant_matrix, int bias, int qmin, int qmax, int intra);
int ff_dct_quantize_c(MpegEncContext *s, int16_t *block, int n, int qscale, int *overflow);

void ff_init_block_index(MpegEncContext *s);

void ff_mpv_motion(MpegEncContext *s,
                   uint8_t *dest_y, uint8_t *dest_cb,
                   uint8_t *dest_cr, int dir,
                   uint8_t **ref_picture,
                   op_pixels_func (*pix_op)[4],
                   qpel_mc_func (*qpix_op)[16]);

/**
 * Allocate a Picture.
 * The pixels are allocated/set by calling get_buffer() if shared = 0.
 */
int ff_alloc_picture(MpegEncContext *s, Picture *pic, int shared);

/**
 * permute block according to permuatation.
 * @param last last non zero element in scantable order
 */
void ff_block_permute(int16_t *block, uint8_t *permutation, const uint8_t *scantable, int last);

static inline void ff_update_block_index(MpegEncContext *s){
    const int block_size = 8;

    s->block_index[0]+=2;
    s->block_index[1]+=2;
    s->block_index[2]+=2;
    s->block_index[3]+=2;
    s->block_index[4]++;
    s->block_index[5]++;
    s->dest[0]+= 2*block_size;
    s->dest[1]+= block_size;
    s->dest[2]+= block_size;
}

static inline int get_bits_diff(MpegEncContext *s){
    const int bits= put_bits_count(&s->pb);
    const int last= s->last_bits;

    s->last_bits = bits;

    return bits - last;
}

static inline int ff_h263_round_chroma(int x){
    static const uint8_t h263_chroma_roundtab[16] = {
    //  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15
        0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1,
    };
    return h263_chroma_roundtab[x & 0xf] + (x >> 3);
}

/* motion_est.c */
void ff_estimate_p_frame_motion(MpegEncContext * s,
                             int mb_x, int mb_y);
void ff_estimate_b_frame_motion(MpegEncContext * s,
                             int mb_x, int mb_y);
int ff_get_best_fcode(MpegEncContext * s, int16_t (*mv_table)[2], int type);
void ff_fix_long_p_mvs(MpegEncContext * s);
void ff_fix_long_mvs(MpegEncContext * s, uint8_t *field_select_table, int field_select,
                     int16_t (*mv_table)[2], int f_code, int type, int truncate);
int ff_init_me(MpegEncContext *s);
int ff_pre_estimate_p_frame_motion(MpegEncContext * s, int mb_x, int mb_y);
int ff_epzs_motion_search(MpegEncContext * s, int *mx_ptr, int *my_ptr,
                             int P[10][2], int src_index, int ref_index, int16_t (*last_mv)[2],
                             int ref_mv_scale, int size, int h);
int ff_get_mb_score(MpegEncContext * s, int mx, int my, int src_index,
                               int ref_index, int size, int h, int add_rate);

/* mpeg12.c */
extern const uint8_t ff_mpeg1_dc_scale_table[128];
extern const uint8_t * const ff_mpeg2_dc_scale_table[4];

void ff_mpeg1_encode_picture_header(MpegEncContext *s, int picture_number);
void ff_mpeg1_encode_mb(MpegEncContext *s,
                        int16_t block[8][64],
                        int motion_x, int motion_y);
void ff_mpeg1_encode_init(MpegEncContext *s);
void ff_mpeg1_encode_slice_header(MpegEncContext *s);

extern const uint8_t ff_aic_dc_scale_table[32];
extern const uint8_t ff_h263_chroma_qscale_table[32];

/* rv10.c */
void ff_rv10_encode_picture_header(MpegEncContext *s, int picture_number);
int ff_rv_decode_dc(MpegEncContext *s, int n);
void ff_rv20_encode_picture_header(MpegEncContext *s, int picture_number);


/* msmpeg4.c */
void ff_msmpeg4_encode_picture_header(MpegEncContext * s, int picture_number);
void ff_msmpeg4_encode_ext_header(MpegEncContext * s);
void ff_msmpeg4_encode_mb(MpegEncContext * s,
                          int16_t block[6][64],
                          int motion_x, int motion_y);
int ff_msmpeg4_decode_picture_header(MpegEncContext * s);
int ff_msmpeg4_decode_ext_header(MpegEncContext * s, int buf_size);
int ff_msmpeg4_decode_init(AVCodecContext *avctx);
void ff_msmpeg4_encode_init(MpegEncContext *s);
int ff_wmv2_decode_picture_header(MpegEncContext * s);
int ff_wmv2_decode_secondary_picture_header(MpegEncContext * s);
void ff_wmv2_add_mb(MpegEncContext *s, int16_t block[6][64], uint8_t *dest_y, uint8_t *dest_cb, uint8_t *dest_cr);
void ff_mspel_motion(MpegEncContext *s,
                               uint8_t *dest_y, uint8_t *dest_cb, uint8_t *dest_cr,
                               uint8_t **ref_picture, op_pixels_func (*pix_op)[4],
                               int motion_x, int motion_y, int h);
int ff_wmv2_encode_picture_header(MpegEncContext * s, int picture_number);
void ff_wmv2_encode_mb(MpegEncContext * s,
                       int16_t block[6][64],
                       int motion_x, int motion_y);

int ff_mpeg_ref_picture(MpegEncContext *s, Picture *dst, Picture *src);
void ff_mpeg_unref_picture(MpegEncContext *s, Picture *picture);
void ff_free_picture_tables(Picture *pic);

#endif /* AVCODEC_MPEGVIDEO_H */
OpenPOWER on IntegriCloud