FFmpeg
iamf_parse.c
Go to the documentation of this file.
1 /*
2  * Immersive Audio Model and Formats parsing
3  * Copyright (c) 2023 James Almer <jamrial@gmail.com>
4  *
5  * This file is part of FFmpeg.
6  *
7  * FFmpeg is free software; you can redistribute it and/or
8  * modify it under the terms of the GNU Lesser General Public
9  * License as published by the Free Software Foundation; either
10  * version 2.1 of the License, or (at your option) any later version.
11  *
12  * FFmpeg is distributed in the hope that it will be useful,
13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15  * Lesser General Public License for more details.
16  *
17  * You should have received a copy of the GNU Lesser General Public
18  * License along with FFmpeg; if not, write to the Free Software
19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20  */
21 
22 #include "libavutil/avassert.h"
23 #include "libavutil/iamf.h"
24 #include "libavutil/intreadwrite.h"
25 #include "libavutil/log.h"
26 #include "libavutil/mem.h"
27 #include "libavcodec/get_bits.h"
28 #include "libavcodec/flac.h"
29 #include "libavcodec/leb.h"
30 #include "libavcodec/mpeg4audio.h"
31 #include "libavcodec/put_bits.h"
32 #include "avio_internal.h"
33 #include "iamf_parse.h"
34 #include "isom.h"
35 
36 static int opus_decoder_config(IAMFCodecConfig *codec_config,
37  AVIOContext *pb, int len)
38 {
39  int ret, left = len - avio_tell(pb);
40 
41  if (left < 11 || codec_config->audio_roll_distance >= 0)
42  return AVERROR_INVALIDDATA;
43 
44  codec_config->extradata = av_malloc(left + 8);
45  if (!codec_config->extradata)
46  return AVERROR(ENOMEM);
47 
48  AV_WB32A(codec_config->extradata, MKBETAG('O','p','u','s'));
49  AV_WB32A(codec_config->extradata + 4, MKBETAG('H','e','a','d'));
50  ret = ffio_read_size(pb, codec_config->extradata + 8, left);
51  if (ret < 0)
52  return ret;
53 
54  codec_config->extradata_size = left + 8;
55  codec_config->sample_rate = 48000;
56 
57  return 0;
58 }
59 
60 static int aac_decoder_config(IAMFCodecConfig *codec_config,
61  AVIOContext *pb, int len, void *logctx)
62 {
63  MPEG4AudioConfig cfg = { 0 };
64  int object_type_id, codec_id, stream_type;
65  int ret, tag, left;
66 
67  if (codec_config->audio_roll_distance >= 0)
68  return AVERROR_INVALIDDATA;
69 
70  ff_mp4_read_descr(logctx, pb, &tag);
71  if (tag != MP4DecConfigDescrTag)
72  return AVERROR_INVALIDDATA;
73 
74  object_type_id = avio_r8(pb);
75  if (object_type_id != 0x40)
76  return AVERROR_INVALIDDATA;
77 
78  stream_type = avio_r8(pb);
79  if (((stream_type >> 2) != 5) || ((stream_type >> 1) & 1))
80  return AVERROR_INVALIDDATA;
81 
82  avio_skip(pb, 3); // buffer size db
83  avio_skip(pb, 4); // rc_max_rate
84  avio_skip(pb, 4); // avg bitrate
85 
86  codec_id = ff_codec_get_id(ff_mp4_obj_type, object_type_id);
87  if (codec_id && codec_id != codec_config->codec_id)
88  return AVERROR_INVALIDDATA;
89 
90  left = ff_mp4_read_descr(logctx, pb, &tag);
91  if (tag != MP4DecSpecificDescrTag ||
92  !left || left > (len - avio_tell(pb)))
93  return AVERROR_INVALIDDATA;
94 
95  // We pad extradata here because avpriv_mpeg4audio_get_config2() needs it.
96  codec_config->extradata = av_malloc((size_t)left + AV_INPUT_BUFFER_PADDING_SIZE);
97  if (!codec_config->extradata)
98  return AVERROR(ENOMEM);
99 
100  ret = ffio_read_size(pb, codec_config->extradata, left);
101  if (ret < 0)
102  return ret;
103  codec_config->extradata_size = left;
104  memset(codec_config->extradata + codec_config->extradata_size, 0,
105  AV_INPUT_BUFFER_PADDING_SIZE);
106 
107  ret = avpriv_mpeg4audio_get_config2(&cfg, codec_config->extradata,
108  codec_config->extradata_size, 1, logctx);
109  if (ret < 0)
110  return ret;
111 
112  codec_config->sample_rate = cfg.sample_rate;
113 
114  return 0;
115 }
116 
117 static int flac_decoder_config(IAMFCodecConfig *codec_config,
118  AVIOContext *pb, int len)
119 {
120  int ret, left;
121 
122  if (codec_config->audio_roll_distance)
123  return AVERROR_INVALIDDATA;
124 
125  avio_skip(pb, 4); // METADATA_BLOCK_HEADER
126 
127  left = len - avio_tell(pb);
128  if (left < FLAC_STREAMINFO_SIZE)
129  return AVERROR_INVALIDDATA;
130 
131  codec_config->extradata = av_malloc(left);
132  if (!codec_config->extradata)
133  return AVERROR(ENOMEM);
134 
135  ret = ffio_read_size(pb, codec_config->extradata, left);
136  if (ret < 0)
137  return ret;
138 
139  codec_config->extradata_size = left;
140  codec_config->sample_rate = AV_RB24(codec_config->extradata + 10) >> 4;
141 
142  return 0;
143 }
144 
145 static int ipcm_decoder_config(IAMFCodecConfig *codec_config,
146  AVIOContext *pb, int len)
147 {
148  static const enum AVCodecID sample_fmt[2][3] = {
149  { AV_CODEC_ID_PCM_S16BE, AV_CODEC_ID_PCM_S24BE, AV_CODEC_ID_PCM_S32BE },
150  { AV_CODEC_ID_PCM_S16LE, AV_CODEC_ID_PCM_S24LE, AV_CODEC_ID_PCM_S32LE },
151  };
152  int sample_format = avio_r8(pb); // 0 = BE, 1 = LE
153  int sample_size = (avio_r8(pb) / 8 - 2); // 16, 24, 32
154  if (sample_format > 1 || sample_size > 2U || codec_config->audio_roll_distance)
155  return AVERROR_INVALIDDATA;
156 
157  codec_config->codec_id = sample_fmt[sample_format][sample_size];
158  codec_config->sample_rate = avio_rb32(pb);
159 
160  if (len - avio_tell(pb))
161  return AVERROR_INVALIDDATA;
162 
163  return 0;
164 }
165 
166 static int codec_config_obu(void *s, IAMFContext *c, AVIOContext *pb, int len)
167 {
168  IAMFCodecConfig **tmp, *codec_config = NULL;
169  FFIOContext b;
170  AVIOContext *pbc;
171  uint8_t *buf;
172  enum AVCodecID avcodec_id;
173  unsigned codec_config_id, nb_samples, codec_id;
174  int16_t audio_roll_distance;
175  int ret;
176 
177  buf = av_malloc(len);
178  if (!buf)
179  return AVERROR(ENOMEM);
180 
181  ret = ffio_read_size(pb, buf, len);
182  if (ret < 0)
183  goto fail;
184 
185  ffio_init_context(&b, buf, len, 0, NULL, NULL, NULL, NULL);
186  pbc = &b.pub;
187 
188  codec_config_id = ffio_read_leb(pbc);
189  codec_id = avio_rb32(pbc);
190  nb_samples = ffio_read_leb(pbc);
191  audio_roll_distance = avio_rb16(pbc);
192 
193  switch(codec_id) {
194  case MKBETAG('O','p','u','s'):
195  avcodec_id = AV_CODEC_ID_OPUS;
196  break;
197  case MKBETAG('m','p','4','a'):
198  avcodec_id = AV_CODEC_ID_AAC;
199  break;
200  case MKBETAG('f','L','a','C'):
201  avcodec_id = AV_CODEC_ID_FLAC;
202  break;
203  default:
204  avcodec_id = AV_CODEC_ID_NONE;
205  break;
206  }
207 
208  for (int i = 0; i < c->nb_codec_configs; i++)
209  if (c->codec_configs[i]->codec_config_id == codec_config_id) {
210  ret = AVERROR_INVALIDDATA;
211  goto fail;
212  }
213 
214  tmp = av_realloc_array(c->codec_configs, c->nb_codec_configs + 1, sizeof(*c->codec_configs));
215  if (!tmp) {
216  ret = AVERROR(ENOMEM);
217  goto fail;
218  }
219  c->codec_configs = tmp;
220 
221  codec_config = av_mallocz(sizeof(*codec_config));
222  if (!codec_config) {
223  ret = AVERROR(ENOMEM);
224  goto fail;
225  }
226 
227  codec_config->codec_config_id = codec_config_id;
228  codec_config->codec_id = avcodec_id;
229  codec_config->nb_samples = nb_samples;
230  codec_config->audio_roll_distance = audio_roll_distance;
231 
232  switch(codec_id) {
233  case MKBETAG('O','p','u','s'):
234  ret = opus_decoder_config(codec_config, pbc, len);
235  break;
236  case MKBETAG('m','p','4','a'):
237  ret = aac_decoder_config(codec_config, pbc, len, s);
238  break;
239  case MKBETAG('f','L','a','C'):
240  ret = flac_decoder_config(codec_config, pbc, len);
241  break;
242  case MKBETAG('i','p','c','m'):
243  ret = ipcm_decoder_config(codec_config, pbc, len);
244  break;
245  default:
246  break;
247  }
248  if (ret < 0)
249  goto fail;
250 
251  if ((codec_config->nb_samples > INT_MAX) || codec_config->nb_samples <= 0 ||
252  (-codec_config->audio_roll_distance > INT_MAX / codec_config->nb_samples)) {
253  ret = AVERROR_INVALIDDATA;
254  goto fail;
255  }
256 
257  c->codec_configs[c->nb_codec_configs++] = codec_config;
258 
259  len -= avio_tell(pbc);
260  if (len)
261  av_log(s, AV_LOG_WARNING, "Underread in codec_config_obu. %d bytes left at the end\n", len);
262 
263  ret = 0;
264 fail:
265  av_free(buf);
266  if (ret < 0) {
267  if (codec_config)
268  av_free(codec_config->extradata);
269  av_free(codec_config);
270  }
271  return ret;
272 }
273 
274 static int update_extradata(AVCodecParameters *codecpar)
275 {
276  GetBitContext gb;
277  PutBitContext pb;
278  int ret;
279 
280  switch(codecpar->codec_id) {
281  case AV_CODEC_ID_OPUS:
282  AV_WB8(codecpar->extradata + 9, codecpar->ch_layout.nb_channels);
283  AV_WL16A(codecpar->extradata + 10, AV_RB16A(codecpar->extradata + 10)); // Byte swap pre-skip
284  AV_WL32A(codecpar->extradata + 12, AV_RB32A(codecpar->extradata + 12)); // Byte swap sample rate
285  AV_WL16A(codecpar->extradata + 16, AV_RB16A(codecpar->extradata + 16)); // Byte swap Output Gain
286  break;
287  case AV_CODEC_ID_AAC: {
288  uint8_t buf[6];
289  int size = FFMIN(codecpar->extradata_size, sizeof(buf));
290 
291  init_put_bits(&pb, buf, sizeof(buf));
292  ret = init_get_bits8(&gb, codecpar->extradata, size);
293  if (ret < 0)
294  return ret;
295 
296  ret = get_bits(&gb, 5);
297  put_bits(&pb, 5, ret);
298  if (ret == AOT_ESCAPE) // violates section 3.11.2, but better check for it
299  put_bits(&pb, 6, get_bits(&gb, 6));
300  ret = get_bits(&gb, 4);
301  put_bits(&pb, 4, ret);
302  if (ret == 0x0f)
303  put_bits(&pb, 24, get_bits(&gb, 24));
304 
305  skip_bits(&gb, 4);
306  put_bits(&pb, 4, codecpar->ch_layout.nb_channels); // set channel config
307  ret = get_bits_left(&gb);
308  if (ret < 0)
309  return AVERROR_INVALIDDATA;
310  ret = FFMIN(ret, put_bits_left(&pb));
311  while (ret >= 32) {
312  put_bits32(&pb, get_bits_long(&gb, 32));
313  ret -= 32;
314  }
315  put_bits(&pb, ret, get_bits_long(&gb, ret));
316  flush_put_bits(&pb);
317 
318  memcpy(codecpar->extradata, buf, put_bytes_output(&pb));
319  break;
320  }
321  case AV_CODEC_ID_FLAC: {
322  uint8_t buf[13];
323  int size = FFMIN(codecpar->extradata_size, sizeof(buf));
324 
325  init_put_bits(&pb, buf, sizeof(buf));
326  ret = init_get_bits8(&gb, codecpar->extradata, size);
327  if (ret < 0)
328  return ret;
329 
330  put_bits32(&pb, get_bits_long(&gb, 32)); // min/max blocksize
331  put_bits64(&pb, 48, get_bits64(&gb, 48)); // min/max framesize
332  put_bits(&pb, 20, get_bits(&gb, 20)); // samplerate
333  skip_bits(&gb, 3);
334  put_bits(&pb, 3, codecpar->ch_layout.nb_channels - 1);
335  ret = get_bits_left(&gb);
336  if (ret < 0)
337  return AVERROR_INVALIDDATA;
338  ret = FFMIN(ret, put_bits_left(&pb));
339  put_bits(&pb, ret, get_bits(&gb, ret));
340  flush_put_bits(&pb);
341 
342  memcpy(codecpar->extradata, buf, put_bytes_output(&pb));
343  break;
344  }
345  }
346 
347  return 0;
348 }
349 
350 static int scalable_channel_layout_config(void *s, AVIOContext *pb,
351  IAMFAudioElement *audio_element,
352  const IAMFCodecConfig *codec_config)
353 {
354  int nb_layers, k = 0;
355 
356  nb_layers = avio_r8(pb) >> 5; // get_bits(&gb, 3);
357  // skip_bits(&gb, 5); //reserved
358 
359  if (nb_layers > 6 || nb_layers == 0)
360  return AVERROR_INVALIDDATA;
361 
362  audio_element->layers = av_calloc(nb_layers, sizeof(*audio_element->layers));
363  if (!audio_element->layers)
364  return AVERROR(ENOMEM);
365 
366  audio_element->nb_layers = nb_layers;
367  for (int i = 0; i < nb_layers; i++) {
368  AVIAMFLayer *layer;
369  int loudspeaker_layout, output_gain_is_present_flag;
370  int substream_count, coupled_substream_count;
371  int ret, byte = avio_r8(pb);
372 
373  layer = av_iamf_audio_element_add_layer(audio_element->element);
374  if (!layer)
375  return AVERROR(ENOMEM);
376 
377  loudspeaker_layout = byte >> 4; // get_bits(&gb, 4);
378  output_gain_is_present_flag = (byte >> 3) & 1; //get_bits1(&gb);
379  if ((byte >> 2) & 1)
380  layer->flags |= AV_IAMF_LAYER_FLAG_RECON_GAIN;
381  substream_count = avio_r8(pb);
382  coupled_substream_count = avio_r8(pb);
383 
384  if (substream_count + k > audio_element->nb_substreams)
385  return AVERROR_INVALIDDATA;
386 
387  audio_element->layers[i].substream_count = substream_count;
388  audio_element->layers[i].coupled_substream_count = coupled_substream_count;
389  if (output_gain_is_present_flag) {
390  layer->output_gain_flags = avio_r8(pb) >> 2; // get_bits(&gb, 6);
391  layer->output_gain = av_make_q(sign_extend(avio_rb16(pb), 16), 1 << 8);
392  }
393 
394  if (loudspeaker_layout < 10)
395  av_channel_layout_copy(&layer->ch_layout, &ff_iamf_scalable_ch_layouts[loudspeaker_layout]);
396  else
397  layer->ch_layout = (AVChannelLayout){ .order = AV_CHANNEL_ORDER_UNSPEC,
398  .nb_channels = substream_count +
399  coupled_substream_count };
400 
401  for (int j = 0; j < substream_count; j++) {
402  IAMFSubStream *substream = &audio_element->substreams[k++];
403 
404  substream->codecpar->ch_layout = coupled_substream_count-- > 0 ? (AVChannelLayout)AV_CHANNEL_LAYOUT_STEREO :
405  (AVChannelLayout)AV_CHANNEL_LAYOUT_MONO;
406 
407  ret = update_extradata(substream->codecpar);
408  if (ret < 0)
409  return ret;
410  }
411 
412  }
413 
414  return 0;
415 }
416 
417 static int ambisonics_config(void *s, AVIOContext *pb,
418  IAMFAudioElement *audio_element,
419  const IAMFCodecConfig *codec_config)
420 {
421  AVIAMFLayer *layer;
422  unsigned ambisonics_mode;
423  int output_channel_count, substream_count, order;
424  int ret;
425 
426  ambisonics_mode = ffio_read_leb(pb);
427  if (ambisonics_mode > 1)
428  return AVERROR_INVALIDDATA;
429 
430  output_channel_count = avio_r8(pb); // C
431  substream_count = avio_r8(pb); // N
432  if (audio_element->nb_substreams != substream_count || output_channel_count == 0)
433  return AVERROR_INVALIDDATA;
434 
435  order = floor(sqrt(output_channel_count - 1));
436  /* incomplete order - some harmonics are missing */
437  if ((order + 1) * (order + 1) != output_channel_count)
438  return AVERROR_INVALIDDATA;
439 
440  audio_element->layers = av_mallocz(sizeof(*audio_element->layers));
441  if (!audio_element->layers)
442  return AVERROR(ENOMEM);
443 
444  audio_element->nb_layers = 1;
445  audio_element->layers->substream_count = substream_count;
446 
447  layer = av_iamf_audio_element_add_layer(audio_element->element);
448  if (!layer)
449  return AVERROR(ENOMEM);
450 
451  layer->ambisonics_mode = ambisonics_mode;
452  if (ambisonics_mode == 0) {
453  for (int i = 0; i < substream_count; i++) {
454  IAMFSubStream *substream = &audio_element->substreams[i];
455 
456  substream->codecpar->ch_layout = (AVChannelLayout)AV_CHANNEL_LAYOUT_MONO;
457 
458  ret = update_extradata(substream->codecpar);
459  if (ret < 0)
460  return ret;
461  }
462 
463  layer->ch_layout.order = AV_CHANNEL_ORDER_CUSTOM;
464  layer->ch_layout.nb_channels = output_channel_count;
465  layer->ch_layout.u.map = av_calloc(output_channel_count, sizeof(*layer->ch_layout.u.map));
466  if (!layer->ch_layout.u.map)
467  return AVERROR(ENOMEM);
468 
469  for (int i = 0; i < output_channel_count; i++)
470  layer->ch_layout.u.map[i].id = avio_r8(pb) + AV_CHAN_AMBISONIC_BASE;
471  } else {
472  int coupled_substream_count = avio_r8(pb); // M
473  int nb_demixing_matrix = substream_count + coupled_substream_count;
474  int demixing_matrix_size = nb_demixing_matrix * output_channel_count;
475 
476  audio_element->layers->coupled_substream_count = coupled_substream_count;
477 
478  layer->ch_layout = (AVChannelLayout){ .order = AV_CHANNEL_ORDER_AMBISONIC, .nb_channels = output_channel_count };
479  layer->demixing_matrix = av_malloc_array(demixing_matrix_size, sizeof(*layer->demixing_matrix));
480  if (!layer->demixing_matrix)
481  return AVERROR(ENOMEM);
482 
483  for (int i = 0; i < demixing_matrix_size; i++)
484  layer->demixing_matrix[i] = av_make_q(sign_extend(avio_rb16(pb), 16), 1 << 8);
485 
486  for (int i = 0; i < substream_count; i++) {
487  IAMFSubStream *substream = &audio_element->substreams[i];
488 
489  substream->codecpar->ch_layout = coupled_substream_count-- > 0 ? (AVChannelLayout)AV_CHANNEL_LAYOUT_STEREO :
490  (AVChannelLayout)AV_CHANNEL_LAYOUT_MONO;
491 
492 
493  ret = update_extradata(substream->codecpar);
494  if (ret < 0)
495  return ret;
496  }
497  }
498 
499  return 0;
500 }
501 
502 static int param_parse(void *s, IAMFContext *c, AVIOContext *pb,
503  unsigned int type,
504  const IAMFAudioElement *audio_element,
505  AVIAMFParamDefinition **out_param_definition)
506 {
507  IAMFParamDefinition *param_definition = NULL;
508  AVIAMFParamDefinition *param;
509  unsigned int parameter_id, parameter_rate, mode;
510  unsigned int duration = 0, constant_subblock_duration = 0, nb_subblocks = 0;
511  unsigned int total_duration = 0;
512  size_t param_size;
513 
514  parameter_id = ffio_read_leb(pb);
515 
516  for (int i = 0; i < c->nb_param_definitions; i++)
517  if (c->param_definitions[i]->param->parameter_id == parameter_id) {
518  param_definition = c->param_definitions[i];
519  break;
520  }
521 
522  parameter_rate = ffio_read_leb(pb);
523  mode = avio_r8(pb) >> 7;
524 
525  if (mode == 0) {
526  duration = ffio_read_leb(pb);
527  if (!duration)
528  return AVERROR_INVALIDDATA;
529  constant_subblock_duration = ffio_read_leb(pb);
530  if (constant_subblock_duration == 0)
531  nb_subblocks = ffio_read_leb(pb);
532  else {
533  nb_subblocks = duration / constant_subblock_duration;
534  total_duration = duration;
535  }
536  }
537 
538  param = av_iamf_param_definition_alloc(type, nb_subblocks, &param_size);
539  if (!param)
540  return AVERROR(ENOMEM);
541 
542  for (int i = 0; i < nb_subblocks; i++) {
543  void *subblock = av_iamf_param_definition_get_subblock(param, i);
544  unsigned int subblock_duration = constant_subblock_duration;
545 
546  if (constant_subblock_duration == 0) {
547  subblock_duration = ffio_read_leb(pb);
548  total_duration += subblock_duration;
549  } else if (i == nb_subblocks - 1)
550  subblock_duration = duration - i * constant_subblock_duration;
551 
552  switch (type) {
553  case AV_IAMF_PARAMETER_DEFINITION_MIX_GAIN: {
554  AVIAMFMixGain *mix = subblock;
555  mix->subblock_duration = subblock_duration;
556  break;
557  }
558  case AV_IAMF_PARAMETER_DEFINITION_DEMIXING: {
559  AVIAMFDemixingInfo *demix = subblock;
560  demix->subblock_duration = subblock_duration;
561  // DefaultDemixingInfoParameterData
562  av_assert0(audio_element);
563  demix->dmixp_mode = avio_r8(pb) >> 5;
564  audio_element->element->default_w = avio_r8(pb) >> 4;
565  break;
566  }
567  case AV_IAMF_PARAMETER_DEFINITION_RECON_GAIN: {
568  AVIAMFReconGain *recon = subblock;
569  recon->subblock_duration = subblock_duration;
570  break;
571  }
572  default:
573  av_free(param);
574  return AVERROR_INVALIDDATA;
575  }
576  }
577 
578  if (!mode && !constant_subblock_duration && total_duration != duration) {
579  av_log(s, AV_LOG_ERROR, "Invalid subblock durations in parameter_id %u\n", parameter_id);
580  av_free(param);
581  return AVERROR_INVALIDDATA;
582  }
583 
584  param->parameter_id = parameter_id;
585  param->parameter_rate = parameter_rate;
586  param->duration = duration;
587  param->constant_subblock_duration = constant_subblock_duration;
588  param->nb_subblocks = nb_subblocks;
589 
590  if (param_definition) {
591  if (param_definition->param_size != param_size || memcmp(param_definition->param, param, param_size)) {
592  av_log(s, AV_LOG_ERROR, "Incosistent parameters for parameter_id %u\n", parameter_id);
593  av_free(param);
594  return AVERROR_INVALIDDATA;
595  }
596  } else {
597  IAMFParamDefinition **tmp = av_realloc_array(c->param_definitions, c->nb_param_definitions + 1,
598  sizeof(*c->param_definitions));
599  if (!tmp) {
600  av_free(param);
601  return AVERROR(ENOMEM);
602  }
603  c->param_definitions = tmp;
604 
605  param_definition = av_mallocz(sizeof(*param_definition));
606  if (!param_definition) {
607  av_free(param);
608  return AVERROR(ENOMEM);
609  }
610  param_definition->param = param;
611  param_definition->mode = !mode;
612  param_definition->param_size = param_size;
613  param_definition->audio_element = audio_element;
614 
615  c->param_definitions[c->nb_param_definitions++] = param_definition;
616  }
617 
618  av_assert0(out_param_definition);
619  *out_param_definition = param;
620 
621  return 0;
622 }
623 
624 static int audio_element_obu(void *s, IAMFContext *c, AVIOContext *pb, int len)
625 {
626  const IAMFCodecConfig *codec_config;
627  AVIAMFAudioElement *element;
628  IAMFAudioElement **tmp, *audio_element = NULL;
629  FFIOContext b;
630  AVIOContext *pbc;
631  uint8_t *buf;
632  unsigned audio_element_id, nb_substreams, codec_config_id, num_parameters;
633  int audio_element_type, ret;
634 
635  buf = av_malloc(len);
636  if (!buf)
637  return AVERROR(ENOMEM);
638 
639  ret = ffio_read_size(pb, buf, len);
640  if (ret < 0)
641  goto fail;
642 
643  ffio_init_context(&b, buf, len, 0, NULL, NULL, NULL, NULL);
644  pbc = &b.pub;
645 
646  audio_element_id = ffio_read_leb(pbc);
647 
648  for (int i = 0; i < c->nb_audio_elements; i++)
649  if (c->audio_elements[i]->audio_element_id == audio_element_id) {
650  av_log(s, AV_LOG_ERROR, "Duplicate audio_element_id %d\n", audio_element_id);
651  ret = AVERROR_INVALIDDATA;
652  goto fail;
653  }
654 
655  audio_element_type = avio_r8(pbc) >> 5;
656  if (audio_element_type > AV_IAMF_AUDIO_ELEMENT_TYPE_SCENE) {
657  av_log(s, AV_LOG_DEBUG, "Unknown audio_element_type referenced in an audio element. Ignoring\n");
658  ret = 0;
659  goto fail;
660  }
661 
662  codec_config_id = ffio_read_leb(pbc);
663 
664  codec_config = ff_iamf_get_codec_config(c, codec_config_id);
665  if (!codec_config) {
666  av_log(s, AV_LOG_ERROR, "Non existant codec config id %d referenced in an audio element\n", codec_config_id);
667  ret = AVERROR_INVALIDDATA;
668  goto fail;
669  }
670 
671  if (codec_config->codec_id == AV_CODEC_ID_NONE) {
672  av_log(s, AV_LOG_DEBUG, "Unknown codec id referenced in an audio element. Ignoring\n");
673  ret = 0;
674  goto fail;
675  }
676 
677  tmp = av_realloc_array(c->audio_elements, c->nb_audio_elements + 1, sizeof(*c->audio_elements));
678  if (!tmp) {
679  ret = AVERROR(ENOMEM);
680  goto fail;
681  }
682  c->audio_elements = tmp;
683 
684  audio_element = av_mallocz(sizeof(*audio_element));
685  if (!audio_element) {
686  ret = AVERROR(ENOMEM);
687  goto fail;
688  }
689 
690  nb_substreams = ffio_read_leb(pbc);
691  audio_element->codec_config_id = codec_config_id;
692  audio_element->audio_element_id = audio_element_id;
693  audio_element->substreams = av_calloc(nb_substreams, sizeof(*audio_element->substreams));
694  if (!audio_element->substreams) {
695  ret = AVERROR(ENOMEM);
696  goto fail;
697  }
698  audio_element->nb_substreams = nb_substreams;
699 
700  element = audio_element->element = av_iamf_audio_element_alloc();
701  if (!element) {
702  ret = AVERROR(ENOMEM);
703  goto fail;
704  }
705  audio_element->celement = element;
706 
707  element->audio_element_type = audio_element_type;
708 
709  for (int i = 0; i < audio_element->nb_substreams; i++) {
710  IAMFSubStream *substream = &audio_element->substreams[i];
711 
712  substream->codecpar = avcodec_parameters_alloc();
713  if (!substream->codecpar) {
714  ret = AVERROR(ENOMEM);
715  goto fail;
716  }
717 
718  substream->audio_substream_id = ffio_read_leb(pbc);
719 
720  substream->codecpar->codec_type = AVMEDIA_TYPE_AUDIO;
721  substream->codecpar->codec_id = codec_config->codec_id;
722  substream->codecpar->frame_size = codec_config->nb_samples;
723  substream->codecpar->sample_rate = codec_config->sample_rate;
724  substream->codecpar->seek_preroll = -codec_config->audio_roll_distance * codec_config->nb_samples;
725 
726  switch(substream->codecpar->codec_id) {
727  case AV_CODEC_ID_AAC:
728  case AV_CODEC_ID_FLAC:
729  case AV_CODEC_ID_OPUS:
730  substream->codecpar->extradata = av_malloc(codec_config->extradata_size + AV_INPUT_BUFFER_PADDING_SIZE);
731  if (!substream->codecpar->extradata) {
732  ret = AVERROR(ENOMEM);
733  goto fail;
734  }
735  memcpy(substream->codecpar->extradata, codec_config->extradata, codec_config->extradata_size);
736  memset(substream->codecpar->extradata + codec_config->extradata_size, 0, AV_INPUT_BUFFER_PADDING_SIZE);
737  substream->codecpar->extradata_size = codec_config->extradata_size;
738  break;
739  }
740  }
741 
742  num_parameters = ffio_read_leb(pbc);
743  if (num_parameters > 2 && audio_element_type == 0) {
744  av_log(s, AV_LOG_ERROR, "Audio Element parameter count %u is invalid"
745  " for Channel representations\n", num_parameters);
746  ret = AVERROR_INVALIDDATA;
747  goto fail;
748  }
749  if (num_parameters && audio_element_type != 0) {
750  av_log(s, AV_LOG_ERROR, "Audio Element parameter count %u is invalid"
751  " for Scene representations\n", num_parameters);
752  ret = AVERROR_INVALIDDATA;
753  goto fail;
754  }
755 
756  for (int i = 0; i < num_parameters; i++) {
757  unsigned type;
758 
759  type = ffio_read_leb(pbc);
760  if (type == AV_IAMF_PARAMETER_DEFINITION_MIX_GAIN)
761  ret = AVERROR_INVALIDDATA;
762  else if (type == AV_IAMF_PARAMETER_DEFINITION_DEMIXING) {
763  if (element->demixing_info) {
764  ret = AVERROR_INVALIDDATA;
765  goto fail;
766  }
767  ret = param_parse(s, c, pbc, type, audio_element, &element->demixing_info);
768  } else if (type == AV_IAMF_PARAMETER_DEFINITION_RECON_GAIN) {
769  if (element->recon_gain_info) {
770  ret = AVERROR_INVALIDDATA;
771  goto fail;
772  }
773  ret = param_parse(s, c, pbc, type, audio_element, &element->recon_gain_info);
774  } else {
775  unsigned param_definition_size = ffio_read_leb(pbc);
776  avio_skip(pbc, param_definition_size);
777  }
778  if (ret < 0)
779  goto fail;
780  }
781 
782  if (audio_element_type == AV_IAMF_AUDIO_ELEMENT_TYPE_CHANNEL) {
783  ret = scalable_channel_layout_config(s, pbc, audio_element, codec_config);
784  if (ret < 0)
785  goto fail;
786  } else if (audio_element_type == AV_IAMF_AUDIO_ELEMENT_TYPE_SCENE) {
787  ret = ambisonics_config(s, pbc, audio_element, codec_config);
788  if (ret < 0)
789  goto fail;
790  } else {
791  av_assert0(0);
792  }
793 
794  c->audio_elements[c->nb_audio_elements++] = audio_element;
795 
796  len -= avio_tell(pbc);
797  if (len)
798  av_log(s, AV_LOG_WARNING, "Underread in audio_element_obu. %d bytes left at the end\n", len);
799 
800  ret = 0;
801 fail:
802  av_free(buf);
803  if (ret < 0)
804  ff_iamf_free_audio_element(&audio_element);
805  return ret;
806 }
807 
808 static int label_string(AVIOContext *pb, char **label)
809 {
810  uint8_t buf[128];
811 
812  avio_get_str(pb, sizeof(buf), buf, sizeof(buf));
813 
814  if (pb->error)
815  return pb->error;
816  if (pb->eof_reached)
817  return AVERROR_INVALIDDATA;
818  *label = av_strdup(buf);
819  if (!*label)
820  return AVERROR(ENOMEM);
821 
822  return 0;
823 }
824 
825 static int mix_presentation_obu(void *s, IAMFContext *c, AVIOContext *pb, int len)
826 {
827  AVIAMFMixPresentation *mix;
828  IAMFMixPresentation **tmp, *mix_presentation = NULL;
829  FFIOContext b;
830  AVIOContext *pbc;
831  uint8_t *buf;
832  unsigned nb_submixes, mix_presentation_id;
833  int ret;
834 
835  buf = av_malloc(len);
836  if (!buf)
837  return AVERROR(ENOMEM);
838 
839  ret = ffio_read_size(pb, buf, len);
840  if (ret < 0)
841  goto fail;
842 
843  ffio_init_context(&b, buf, len, 0, NULL, NULL, NULL, NULL);
844  pbc = &b.pub;
845 
846  mix_presentation_id = ffio_read_leb(pbc);
847 
848  for (int i = 0; i < c->nb_mix_presentations; i++)
849  if (c->mix_presentations[i]->mix_presentation_id == mix_presentation_id) {
850  av_log(s, AV_LOG_ERROR, "Duplicate mix_presentation_id %d\n", mix_presentation_id);
851  ret = AVERROR_INVALIDDATA;
852  goto fail;
853  }
854 
855  tmp = av_realloc_array(c->mix_presentations, c->nb_mix_presentations + 1, sizeof(*c->mix_presentations));
856  if (!tmp) {
857  ret = AVERROR(ENOMEM);
858  goto fail;
859  }
860  c->mix_presentations = tmp;
861 
862  mix_presentation = av_mallocz(sizeof(*mix_presentation));
863  if (!mix_presentation) {
864  ret = AVERROR(ENOMEM);
865  goto fail;
866  }
867 
868  mix_presentation->mix_presentation_id = mix_presentation_id;
869  mix = mix_presentation->mix = av_iamf_mix_presentation_alloc();
870  if (!mix) {
871  ret = AVERROR(ENOMEM);
872  goto fail;
873  }
874  mix_presentation->cmix = mix;
875 
876  mix_presentation->count_label = ffio_read_leb(pbc);
877  mix_presentation->language_label = av_calloc(mix_presentation->count_label,
878  sizeof(*mix_presentation->language_label));
879  if (!mix_presentation->language_label) {
880  mix_presentation->count_label = 0;
881  ret = AVERROR(ENOMEM);
882  goto fail;
883  }
884 
885  for (int i = 0; i < mix_presentation->count_label; i++) {
886  ret = label_string(pbc, &mix_presentation->language_label[i]);
887  if (ret < 0)
888  goto fail;
889  }
890 
891  for (int i = 0; i < mix_presentation->count_label; i++) {
892  char *annotation = NULL;
893  ret = label_string(pbc, &annotation);
894  if (ret < 0)
895  goto fail;
896  ret = av_dict_set(&mix->annotations, mix_presentation->language_label[i], annotation,
897  AV_DICT_DONT_STRDUP_VAL | AV_DICT_DONT_OVERWRITE);
898  if (ret < 0)
899  goto fail;
900  }
901 
902  nb_submixes = ffio_read_leb(pbc);
903  for (int i = 0; i < nb_submixes; i++) {
904  AVIAMFSubmix *sub_mix;
905  unsigned nb_elements, nb_layouts;
906 
907  sub_mix = av_iamf_mix_presentation_add_submix(mix);
908  if (!sub_mix) {
909  ret = AVERROR(ENOMEM);
910  goto fail;
911  }
912 
913  nb_elements = ffio_read_leb(pbc);
914  for (int j = 0; j < nb_elements; j++) {
915  AVIAMFSubmixElement *submix_element;
916  IAMFAudioElement *audio_element = NULL;
917  unsigned int rendering_config_extension_size;
918 
919  submix_element = av_iamf_submix_add_element(sub_mix);
920  if (!submix_element) {
921  ret = AVERROR(ENOMEM);
922  goto fail;
923  }
924 
925  submix_element->audio_element_id = ffio_read_leb(pbc);
926 
927  for (int k = 0; k < c->nb_audio_elements; k++)
928  if (c->audio_elements[k]->audio_element_id == submix_element->audio_element_id) {
929  audio_element = c->audio_elements[k];
930  break;
931  }
932 
933  if (!audio_element) {
934  av_log(s, AV_LOG_ERROR, "Invalid Audio Element with id %u referenced by Mix Parameters %u\n",
935  submix_element->audio_element_id, mix_presentation_id);
936  ret = AVERROR_INVALIDDATA;
937  goto fail;
938  }
939 
940  for (int k = 0; k < mix_presentation->count_label; k++) {
941  char *annotation = NULL;
942  ret = label_string(pbc, &annotation);
943  if (ret < 0)
944  goto fail;
945  ret = av_dict_set(&submix_element->annotations, mix_presentation->language_label[k], annotation,
946  AV_DICT_DONT_STRDUP_VAL | AV_DICT_DONT_OVERWRITE);
947  if (ret < 0)
948  goto fail;
949  }
950 
951  submix_element->headphones_rendering_mode = avio_r8(pbc) >> 6;
952 
953  rendering_config_extension_size = ffio_read_leb(pbc);
954  avio_skip(pbc, rendering_config_extension_size);
955 
956  ret = param_parse(s, c, pbc, AV_IAMF_PARAMETER_DEFINITION_MIX_GAIN,
957  NULL,
958  &submix_element->element_mix_config);
959  if (ret < 0)
960  goto fail;
961  submix_element->default_mix_gain = av_make_q(sign_extend(avio_rb16(pbc), 16), 1 << 8);
962  }
963 
964  ret = param_parse(s, c, pbc, AV_IAMF_PARAMETER_DEFINITION_MIX_GAIN, NULL, &sub_mix->output_mix_config);
965  if (ret < 0)
966  goto fail;
967  sub_mix->default_mix_gain = av_make_q(sign_extend(avio_rb16(pbc), 16), 1 << 8);
968 
969  nb_layouts = ffio_read_leb(pbc);
970  for (int j = 0; j < nb_layouts; j++) {
971  AVIAMFSubmixLayout *submix_layout;
972  int info_type;
973  int byte = avio_r8(pbc);
974 
975  submix_layout = av_iamf_submix_add_layout(sub_mix);
976  if (!submix_layout) {
977  ret = AVERROR(ENOMEM);
978  goto fail;
979  }
980 
981  submix_layout->layout_type = byte >> 6;
982  if (submix_layout->layout_type < AV_IAMF_SUBMIX_LAYOUT_TYPE_LOUDSPEAKERS ||
983  submix_layout->layout_type > AV_IAMF_SUBMIX_LAYOUT_TYPE_BINAURAL) {
984  av_log(s, AV_LOG_ERROR, "Invalid Layout type %u in a submix from Mix Presentation %u\n",
985  submix_layout->layout_type, mix_presentation_id);
986  ret = AVERROR_INVALIDDATA;
987  goto fail;
988  }
989  if (submix_layout->layout_type == 2) {
990  int sound_system;
991  sound_system = (byte >> 2) & 0xF;
992  if (sound_system >= FF_ARRAY_ELEMS(ff_iamf_sound_system_map)) {
993  ret = AVERROR_INVALIDDATA;
994  goto fail;
995  }
996  av_channel_layout_copy(&submix_layout->sound_system, &ff_iamf_sound_system_map[sound_system].layout);
997  }
998 
999  info_type = avio_r8(pbc);
1000  submix_layout->integrated_loudness = av_make_q(sign_extend(avio_rb16(pbc), 16), 1 << 8);
1001  submix_layout->digital_peak = av_make_q(sign_extend(avio_rb16(pbc), 16), 1 << 8);
1002 
1003  if (info_type & 1)
1004  submix_layout->true_peak = av_make_q(sign_extend(avio_rb16(pbc), 16), 1 << 8);
1005  if (info_type & 2) {
1006  unsigned int num_anchored_loudness = avio_r8(pbc);
1007 
1008  for (int k = 0; k < num_anchored_loudness; k++) {
1009  unsigned int anchor_element = avio_r8(pbc);
1010  AVRational anchored_loudness = av_make_q(sign_extend(avio_rb16(pbc), 16), 1 << 8);
1011  if (anchor_element == IAMF_ANCHOR_ELEMENT_DIALOGUE)
1012  submix_layout->dialogue_anchored_loudness = anchored_loudness;
1013  else if (anchor_element <= IAMF_ANCHOR_ELEMENT_ALBUM)
1014  submix_layout->album_anchored_loudness = anchored_loudness;
1015  else
1016  av_log(s, AV_LOG_DEBUG, "Unknown anchor_element. Ignoring\n");
1017  }
1018  }
1019 
1020  if (info_type & 0xFC) {
1021  unsigned int info_type_size = ffio_read_leb(pbc);
1022  avio_skip(pbc, info_type_size);
1023  }
1024  }
1025  }
1026 
1027  c->mix_presentations[c->nb_mix_presentations++] = mix_presentation;
1028 
1029  len -= avio_tell(pbc);
1030  if (len)
1031  av_log(s, AV_LOG_WARNING, "Underread in mix_presentation_obu. %d bytes left at the end\n", len);
1032 
1033  ret = 0;
1034 fail:
1035  av_free(buf);
1036  if (ret < 0)
1037  ff_iamf_free_mix_presentation(&mix_presentation);
1038  return ret;
1039 }
1040 
1041 int ff_iamf_parse_obu_header(const uint8_t *buf, int buf_size,
1042  unsigned *obu_size, int *start_pos, enum IAMF_OBU_Type *type,
1043  unsigned *skip_samples, unsigned *discard_padding)
1044 {
1045  GetBitContext gb;
1046  int ret, extension_flag, trimming, start;
1047  unsigned skip = 0, discard = 0;
1048  unsigned size;
1049 
1050  ret = init_get_bits8(&gb, buf, FFMIN(buf_size, MAX_IAMF_OBU_HEADER_SIZE));
1051  if (ret < 0)
1052  return ret;
1053 
1054  *type = get_bits(&gb, 5);
1055  /*redundant =*/ get_bits1(&gb);
1056  trimming = get_bits1(&gb);
1057  extension_flag = get_bits1(&gb);
1058 
1059  *obu_size = get_leb(&gb);
1060  if (*obu_size > INT_MAX)
1061  return AVERROR_INVALIDDATA;
1062 
1063  start = get_bits_count(&gb) / 8;
1064 
1065  if (trimming) {
1066  discard = get_leb(&gb); // num_samples_to_trim_at_end
1067  skip = get_leb(&gb); // num_samples_to_trim_at_start
1068  }
1069 
1070  if (skip_samples)
1071  *skip_samples = skip;
1072  if (discard_padding)
1073  *discard_padding = discard;
1074 
1075  if (extension_flag) {
1076  unsigned int extension_bytes;
1077  extension_bytes = get_leb(&gb);
1078  if (extension_bytes > INT_MAX / 8)
1079  return AVERROR_INVALIDDATA;
1080  skip_bits_long(&gb, extension_bytes * 8);
1081  }
1082 
1083  if (get_bits_left(&gb) < 0)
1084  return AVERROR_INVALIDDATA;
1085 
1086  size = *obu_size + start;
1087  if (size > INT_MAX)
1088  return AVERROR_INVALIDDATA;
1089 
1090  *obu_size -= get_bits_count(&gb) / 8 - start;
1091  *start_pos = size - *obu_size;
1092 
1093  return size;
1094 }
1095 
1096 int ff_iamfdec_read_descriptors(IAMFContext *c, AVIOContext *pb,
1097  int max_size, void *log_ctx)
1098 {
1099  uint8_t header[MAX_IAMF_OBU_HEADER_SIZE + AV_INPUT_BUFFER_PADDING_SIZE];
1100  int ret;
1101 
1102  while (1) {
1103  unsigned obu_size;
1104  enum IAMF_OBU_Type type;
1105  int start_pos, len, size;
1106 
1107  if ((ret = ffio_ensure_seekback(pb, FFMIN(MAX_IAMF_OBU_HEADER_SIZE, max_size))) < 0)
1108  return ret;
1109  size = avio_read(pb, header, FFMIN(MAX_IAMF_OBU_HEADER_SIZE, max_size));
1110  if (size < 0)
1111  return size;
1112  memset(header + size, 0, AV_INPUT_BUFFER_PADDING_SIZE);
1113 
1114  len = ff_iamf_parse_obu_header(header, size, &obu_size, &start_pos, &type, NULL, NULL);
1115  if (len < 0 || obu_size > max_size) {
1116  av_log(log_ctx, AV_LOG_ERROR, "Failed to read obu header\n");
1117  avio_seek(pb, -size, SEEK_CUR);
1118  return len;
1119  }
1120 
1121  if (type >= IAMF_OBU_IA_PARAMETER_BLOCK && type < IAMF_OBU_IA_SEQUENCE_HEADER) {
1122  avio_seek(pb, -size, SEEK_CUR);
1123  break;
1124  }
1125 
1126  avio_seek(pb, -(size - start_pos), SEEK_CUR);
1127  switch (type) {
1128  case IAMF_OBU_IA_CODEC_CONFIG:
1129  ret = codec_config_obu(log_ctx, c, pb, obu_size);
1130  break;
1131  case IAMF_OBU_IA_AUDIO_ELEMENT:
1132  ret = audio_element_obu(log_ctx, c, pb, obu_size);
1133  break;
1134  case IAMF_OBU_IA_MIX_PRESENTATION:
1135  ret = mix_presentation_obu(log_ctx, c, pb, obu_size);
1136  break;
1137  default: {
1138  int64_t offset = avio_skip(pb, obu_size);
1139  if (offset < 0)
1140  ret = offset;
1141  break;
1142  }
1143  }
1144  if (ret < 0) {
1145  av_log(log_ctx, AV_LOG_ERROR, "Failed to read obu type %d\n", type);
1146  return ret;
1147  }
1148  max_size -= obu_size + start_pos;
1149  if (max_size < 0)
1150  return AVERROR_INVALIDDATA;
1151  if (!max_size)
1152  break;
1153  }
1154 
1155  return 0;
1156 }
update_extradata
static int update_extradata(AVCodecParameters *codecpar)
Definition: iamf_parse.c:274
AV_CODEC_ID_PCM_S16LE
@ AV_CODEC_ID_PCM_S16LE
Definition: codec_id.h:328
iamf.h
ff_iamf_free_mix_presentation
void ff_iamf_free_mix_presentation(IAMFMixPresentation **pmix_presentation)
Definition: iamf.c:85
skip_bits_long
static void skip_bits_long(GetBitContext *s, int n)
Skips the specified number of bits.
Definition: get_bits.h:278
AV_IAMF_SUBMIX_LAYOUT_TYPE_LOUDSPEAKERS
@ AV_IAMF_SUBMIX_LAYOUT_TYPE_LOUDSPEAKERS
The layout follows the loudspeaker sound system convention of ITU-2051-3.
Definition: iamf.h:497
AV_LOG_WARNING
#define AV_LOG_WARNING
Something somehow does not look correct.
Definition: log.h:186
AVCodecParameters::extradata
uint8_t * extradata
Extra binary data needed for initializing the decoder, codec-dependent.
Definition: codec_par.h:69
MP4DecConfigDescrTag
#define MP4DecConfigDescrTag
Definition: isom.h:375
ffio_init_context
void ffio_init_context(FFIOContext *s, unsigned char *buffer, int buffer_size, int write_flag, void *opaque, int(*read_packet)(void *opaque, uint8_t *buf, int buf_size), int(*write_packet)(void *opaque, const uint8_t *buf, int buf_size), int64_t(*seek)(void *opaque, int64_t offset, int whence))
Definition: aviobuf.c:50
mix
static int mix(int c0, int c1)
Definition: 4xm.c:716
get_bits_left
static int get_bits_left(GetBitContext *gb)
Definition: get_bits.h:695
AVERROR
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
put_bits32
static void av_unused put_bits32(PutBitContext *s, uint32_t value)
Write exactly 32 bits into a bitstream.
Definition: put_bits.h:291
AVCodecParameters::codec_type
enum AVMediaType codec_type
General type of the encoded data.
Definition: codec_par.h:51
AV_CHANNEL_LAYOUT_STEREO
#define AV_CHANNEL_LAYOUT_STEREO
Definition: channel_layout.h:387
put_bytes_output
static int put_bytes_output(const PutBitContext *s)
Definition: put_bits.h:89
IAMF_OBU_IA_SEQUENCE_HEADER
@ IAMF_OBU_IA_SEQUENCE_HEADER
Definition: iamf.h:63
ff_mp4_obj_type
const AVCodecTag ff_mp4_obj_type[]
Definition: isom.c:34
AVCodecParameters
This struct describes the properties of an encoded stream.
Definition: codec_par.h:47
AVIAMFAudioElement::default_w
unsigned int default_w
Default weight value as defined in section 3.6 of IAMF.
Definition: iamf.h:393
IAMFAudioElement::nb_substreams
unsigned int nb_substreams
Definition: iamf.h:99
av_iamf_param_definition_alloc
AVIAMFParamDefinition * av_iamf_param_definition_alloc(enum AVIAMFParamDefinitionType type, unsigned int nb_subblocks, size_t *out_size)
Allocates memory for AVIAMFParamDefinition, plus an array of.
Definition: iamf.c:159
get_bits_long
static unsigned int get_bits_long(GetBitContext *s, int n)
Read 0-32 bits.
Definition: get_bits.h:421
int64_t
long long int64_t
Definition: coverity.c:34
init_put_bits
static void init_put_bits(PutBitContext *s, uint8_t *buffer, int buffer_size)
Initialize the PutBitContext s.
Definition: put_bits.h:62
get_bits_count
static int get_bits_count(const GetBitContext *s)
Definition: get_bits.h:266
AVChannelLayout::map
AVChannelCustom * map
This member must be used when the channel order is AV_CHANNEL_ORDER_CUSTOM.
Definition: channel_layout.h:362
AV_RB32A
#define AV_RB32A(p)
Definition: intreadwrite.h:575
put_bits
static void put_bits(Jpeg2000EncoderContext *s, int val, int n)
put n times val bit
Definition: j2kenc.c:223
tmp
static uint8_t tmp[11]
Definition: aes_ctr.c:28
ff_iamf_sound_system_map
const struct IAMFSoundSystemMap ff_iamf_sound_system_map[13]
Definition: iamf.c:48
put_bits64
static void put_bits64(PutBitContext *s, int n, uint64_t value)
Write up to 64 bits into a bitstream.
Definition: put_bits.h:334
AVCodecParameters::seek_preroll
int seek_preroll
Audio only.
Definition: codec_par.h:214
b
#define b
Definition: input.c:41
AVIOContext::error
int error
contains the error code or 0 if no error happened
Definition: avio.h:239
AVIAMFSubmixLayout::layout_type
enum AVIAMFSubmixLayoutType layout_type
Definition: iamf.h:513
AVIAMFParamDefinition
Parameters as defined in section 3.6.1 of IAMF.
Definition: iamf.h:193
AVIAMFSubmixElement::default_mix_gain
AVRational default_mix_gain
Default mix gain value to apply when there are no AVIAMFParamDefinition with element_mix_config's par...
Definition: iamf.h:469
AV_CODEC_ID_FLAC
@ AV_CODEC_ID_FLAC
Definition: codec_id.h:452
AVChannelLayout::order
enum AVChannelOrder order
Channel order used in this layout.
Definition: channel_layout.h:316
iamf_parse.h
ambisonics_config
static int ambisonics_config(void *s, AVIOContext *pb, IAMFAudioElement *audio_element, const IAMFCodecConfig *codec_config)
Definition: iamf_parse.c:417
AVChannelLayout::nb_channels
int nb_channels
Number of channels in this layout.
Definition: channel_layout.h:321
FFIOContext
Definition: avio_internal.h:28
flac_decoder_config
static int flac_decoder_config(IAMFCodecConfig *codec_config, AVIOContext *pb, int len)
Definition: iamf_parse.c:117
av_malloc
#define av_malloc(s)
Definition: tableprint_vlc.h:30
IAMFMixPresentation::cmix
const AVIAMFMixPresentation * cmix
Definition: iamf.h:108
MPEG4AudioConfig
Definition: mpeg4audio.h:29
AVIAMFSubmixLayout::digital_peak
AVRational digital_peak
The digital (sampled) peak value of the audio signal, as defined in ITU-1770-4.
Definition: iamf.h:531
skip_bits
static void skip_bits(GetBitContext *s, int n)
Definition: get_bits.h:381
get_bits
static unsigned int get_bits(GetBitContext *s, int n)
Read 1-25 bits.
Definition: get_bits.h:335
mpeg4audio.h
AVIAMFSubmixLayout::integrated_loudness
AVRational integrated_loudness
The program integrated loudness information, as defined in ITU-1770-4.
Definition: iamf.h:526
av_iamf_mix_presentation_add_submix
AVIAMFSubmix * av_iamf_mix_presentation_add_submix(AVIAMFMixPresentation *mix_presentation)
Allocate a submix and add it to a given AVIAMFMixPresentation.
IAMFCodecConfig::extradata
uint8_t * extradata
Definition: iamf.h:74
MP4DecSpecificDescrTag
#define MP4DecSpecificDescrTag
Definition: isom.h:376
AOT_ESCAPE
@ AOT_ESCAPE
Y Escape Value.
Definition: mpeg4audio.h:102
AV_CODEC_ID_PCM_S16BE
@ AV_CODEC_ID_PCM_S16BE
Definition: codec_id.h:329
IAMFParamDefinition
Definition: iamf.h:121
fail
#define fail()
Definition: checkasm.h:188
GetBitContext
Definition: get_bits.h:108
AVIAMFSubmixLayout
Submix layout as defined in section 3.7.6 of IAMF.
Definition: iamf.h:510
IAMF_ANCHOR_ELEMENT_ALBUM
@ IAMF_ANCHOR_ELEMENT_ALBUM
Definition: iamf.h:142
put_bits_left
static int put_bits_left(PutBitContext *s)
Definition: put_bits.h:125
avio_tell
static av_always_inline int64_t avio_tell(AVIOContext *s)
ftell() equivalent for AVIOContext.
Definition: avio.h:494
av_iamf_audio_element_alloc
AVIAMFAudioElement * av_iamf_audio_element_alloc(void)
Allocates a AVIAMFAudioElement, and initializes its fields with default values.
Definition: iamf.c:322
type
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf type
Definition: writing_filters.txt:86
IAMF_OBU_IA_PARAMETER_BLOCK
@ IAMF_OBU_IA_PARAMETER_BLOCK
Definition: iamf.h:41
AVIAMFAudioElement::audio_element_type
enum AVIAMFAudioElementType audio_element_type
Audio element type as defined in section 3.6 of IAMF.
Definition: iamf.h:388
opus_decoder_config
static int opus_decoder_config(IAMFCodecConfig *codec_config, AVIOContext *pb, int len)
Definition: iamf_parse.c:36
AVIAMFReconGain
Recon Gain Info Parameter Data as defined in section 3.8.3 of IAMF.
Definition: iamf.h:148
param_parse
static int param_parse(void *s, IAMFContext *c, AVIOContext *pb, unsigned int type, const IAMFAudioElement *audio_element, AVIAMFParamDefinition **out_param_definition)
Definition: iamf_parse.c:502
codec_config_obu
static int codec_config_obu(void *s, IAMFContext *c, AVIOContext *pb, int len)
Definition: iamf_parse.c:166
AV_DICT_DONT_STRDUP_VAL
#define AV_DICT_DONT_STRDUP_VAL
Take ownership of a value that's been allocated with av_malloc() or another memory allocation functio...
Definition: dict.h:79
IAMFAudioElement::element
AVIAMFAudioElement * element
element backs celement iff the AVIAMFAudioElement is owned by this structure.
Definition: iamf.h:95
AV_IAMF_PARAMETER_DEFINITION_RECON_GAIN
@ AV_IAMF_PARAMETER_DEFINITION_RECON_GAIN
Subblocks are of struct type AVIAMFReconGain.
Definition: iamf.h:181
AVIAMFSubmixElement::annotations
AVDictionary * annotations
A dictionary of strings describing the submix in different languages.
Definition: iamf.h:490
avassert.h
avio_rb32
unsigned int avio_rb32(AVIOContext *s)
Definition: aviobuf.c:761
AV_LOG_ERROR
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:180
FF_ARRAY_ELEMS
#define FF_ARRAY_ELEMS(a)
Definition: sinewin_tablegen.c:29
init_get_bits8
static int init_get_bits8(GetBitContext *s, const uint8_t *buffer, int byte_size)
Initialize GetBitContext.
Definition: get_bits.h:545
IAMFCodecConfig::sample_rate
int sample_rate
Definition: iamf.h:72
IAMF_OBU_IA_MIX_PRESENTATION
@ IAMF_OBU_IA_MIX_PRESENTATION
Definition: iamf.h:40
duration
int64_t duration
Definition: movenc.c:65
AVCodecParameters::frame_size
int frame_size
Audio only.
Definition: codec_par.h:195
av_iamf_submix_add_layout
AVIAMFSubmixLayout * av_iamf_submix_add_layout(AVIAMFSubmix *submix)
Allocate a submix layout and add it to a given AVIAMFSubmix.
avpriv_mpeg4audio_get_config2
int avpriv_mpeg4audio_get_config2(MPEG4AudioConfig *c, const uint8_t *buf, int size, int sync_extension, void *logctx)
Parse MPEG-4 systems extradata from a raw buffer to retrieve audio configuration.
Definition: mpeg4audio.c:165
intreadwrite.h
s
#define s(width, name)
Definition: cbs_vp9.c:198
IAMFSubStream::audio_substream_id
unsigned int audio_substream_id
Definition: iamf.h:83
IAMFLayer::substream_count
unsigned int substream_count
Definition: iamf.h:78
AVIAMFSubmixLayout::dialogue_anchored_loudness
AVRational dialogue_anchored_loudness
The Dialogue loudness information, as defined in ITU-1770-4.
Definition: iamf.h:539
av_realloc_array
void * av_realloc_array(void *ptr, size_t nmemb, size_t size)
Definition: mem.c:217
floor
static __device__ float floor(float a)
Definition: cuda_runtime.h:173
AVMEDIA_TYPE_AUDIO
@ AVMEDIA_TYPE_AUDIO
Definition: avutil.h:202
av_iamf_param_definition_get_subblock
static av_always_inline void * av_iamf_param_definition_get_subblock(const AVIAMFParamDefinition *par, unsigned int idx)
Get the subblock at the specified.
Definition: iamf.h:260
AV_CHANNEL_ORDER_UNSPEC
@ AV_CHANNEL_ORDER_UNSPEC
Only the channel count is specified, without any further information about the channel order.
Definition: channel_layout.h:116
av_assert0
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:40
AVIAMFSubmixElement::headphones_rendering_mode
enum AVIAMFHeadphonesMode headphones_rendering_mode
A value that indicates whether the referenced channel-based Audio Element shall be rendered to stereo...
Definition: iamf.h:478
ffio_read_leb
unsigned int ffio_read_leb(AVIOContext *s)
Read a unsigned integer coded as a variable number of up to eight little-endian bytes,...
Definition: aviobuf.c:927
AV_LOG_DEBUG
#define AV_LOG_DEBUG
Stuff which is only useful for libav* developers.
Definition: log.h:201
get_bits.h
AVIAMFLayer::ch_layout
AVChannelLayout ch_layout
Definition: iamf.h:297
IAMFAudioElement::nb_layers
unsigned int nb_layers
Definition: iamf.h:104
ff_iamfdec_read_descriptors
int ff_iamfdec_read_descriptors(IAMFContext *c, AVIOContext *pb, int max_size, void *log_ctx)
Definition: iamf_parse.c:1096
codec_id
enum AVCodecID codec_id
Definition: vaapi_decode.c:394
PutBitContext
Definition: put_bits.h:50
IAMFAudioElement::audio_element_id
unsigned int audio_element_id
Definition: iamf.h:96
AVIAMFDemixingInfo
Demixing Info Parameter Data as defined in section 3.8.2 of IAMF.
Definition: iamf.h:128
IAMFSoundSystemMap::layout
AVChannelLayout layout
Definition: iamf.h:163
AV_CHANNEL_ORDER_AMBISONIC
@ AV_CHANNEL_ORDER_AMBISONIC
The audio is represented as the decomposition of the sound field into spherical harmonics.
Definition: channel_layout.h:152
NULL
#define NULL
Definition: coverity.c:32
get_leb
static unsigned get_leb(GetBitContext *s)
Read a unsigned integer coded as a variable number of up to eight little-endian bytes,...
Definition: leb.h:35
isom.h
IAMF_OBU_IA_AUDIO_ELEMENT
@ IAMF_OBU_IA_AUDIO_ELEMENT
Definition: iamf.h:39
AVRational
Rational number (pair of numerator and denominator).
Definition: rational.h:58
AVIAMFParamDefinition::duration
unsigned int duration
The accumulated duration of all blocks in this parameter definition, in units of 1 / parameter_rate.
Definition: iamf.h:231
ff_iamf_get_codec_config
static IAMFCodecConfig * ff_iamf_get_codec_config(const IAMFContext *c, unsigned int codec_config_id)
Definition: iamf.h:170
get_bits1
static unsigned int get_bits1(GetBitContext *s)
Definition: get_bits.h:388
audio_element_obu
static int audio_element_obu(void *s, IAMFContext *c, AVIOContext *pb, int len)
Definition: iamf_parse.c:624
AVIAMFLayer
A layer defining a Channel Layout in the Audio Element.
Definition: iamf.h:294
IAMF_OBU_IA_CODEC_CONFIG
@ IAMF_OBU_IA_CODEC_CONFIG
Definition: iamf.h:38
IAMFSubStream
Definition: iamf.h:82
IAMFAudioElement::layers
IAMFLayer * layers
Definition: iamf.h:103
FLAC_STREAMINFO_SIZE
#define FLAC_STREAMINFO_SIZE
Definition: flac.h:32
AV_DICT_DONT_OVERWRITE
#define AV_DICT_DONT_OVERWRITE
Don't overwrite existing entries.
Definition: dict.h:81
AVCodecParameters::ch_layout
AVChannelLayout ch_layout
Audio only.
Definition: codec_par.h:180
IAMFAudioElement::celement
const AVIAMFAudioElement * celement
Definition: iamf.h:90
c
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
Definition: undefined.txt:32
ff_mp4_read_descr
int ff_mp4_read_descr(void *logctx, AVIOContext *pb, int *tag)
Definition: isom.c:295
AVCodecParameters::sample_rate
int sample_rate
Audio only.
Definition: codec_par.h:184
ff_iamf_parse_obu_header
int ff_iamf_parse_obu_header(const uint8_t *buf, int buf_size, unsigned *obu_size, int *start_pos, enum IAMF_OBU_Type *type, unsigned *skip_samples, unsigned *discard_padding)
Definition: iamf_parse.c:1041
AVIAMFLayer::output_gain_flags
unsigned int output_gain_flags
Output gain channel flags as defined in section 3.6.2 of IAMF.
Definition: iamf.h:310
AVCodecID
AVCodecID
Identify the syntax and semantics of the bitstream.
Definition: codec_id.h:49
AVIAMFSubmixElement::audio_element_id
unsigned int audio_element_id
The id of the Audio Element this submix element references.
Definition: iamf.h:452
AVCodecParameters::extradata_size
int extradata_size
Size of the extradata content in bytes.
Definition: codec_par.h:73
AVIAMFSubmix
Submix layout as defined in section 3.7 of IAMF.
Definition: iamf.h:552
AV_CODEC_ID_AAC
@ AV_CODEC_ID_AAC
Definition: codec_id.h:442
IAMFAudioElement
Definition: iamf.h:89
AVIAMFReconGain::subblock_duration
unsigned int subblock_duration
Duration for the given subblock, in units of 1 / parameter_rate.
Definition: iamf.h:156
AVIOContext
Bytestream IO Context.
Definition: avio.h:160
IAMFCodecConfig::nb_samples
unsigned nb_samples
Definition: iamf.h:70
AV_CODEC_ID_PCM_S24LE
@ AV_CODEC_ID_PCM_S24LE
Definition: codec_id.h:340
AVIAMFDemixingInfo::subblock_duration
unsigned int subblock_duration
Duration for the given subblock, in units of 1 / parameter_rate.
Definition: iamf.h:136
ff_codec_get_id
enum AVCodecID ff_codec_get_id(const AVCodecTag *tags, unsigned int tag)
Definition: utils.c:133
IAMFCodecConfig
Definition: iamf.h:66
AVChannelLayout
An AVChannelLayout holds information about the channel layout of audio data.
Definition: channel_layout.h:311
IAMFCodecConfig::extradata_size
int extradata_size
Definition: iamf.h:73
avio_get_str
int avio_get_str(AVIOContext *pb, int maxlen, char *buf, int buflen)
Read a string from pb into buf.
Definition: aviobuf.c:866
size
int size
Definition: twinvq_data.h:10344
av_make_q
static AVRational av_make_q(int num, int den)
Create an AVRational.
Definition: rational.h:71
IAMFLayer::coupled_substream_count
unsigned int coupled_substream_count
Definition: iamf.h:79
AV_WL32A
#define AV_WL32A(p, v)
Definition: intreadwrite.h:571
MKBETAG
#define MKBETAG(a, b, c, d)
Definition: macros.h:56
AV_CODEC_ID_OPUS
@ AV_CODEC_ID_OPUS
Definition: codec_id.h:500
av_iamf_mix_presentation_alloc
AVIAMFMixPresentation * av_iamf_mix_presentation_alloc(void)
Allocates a AVIAMFMixPresentation, and initializes its fields with default values.
Definition: iamf.c:520
AVChannelLayout::u
union AVChannelLayout::@419 u
Details about which channels are present in this layout.
IAMFContext
Definition: iamf.h:128
header
static const uint8_t header[24]
Definition: sdr2.c:68
avio_r8
int avio_r8(AVIOContext *s)
Definition: aviobuf.c:603
IAMFAudioElement::substreams
IAMFSubStream * substreams
Definition: iamf.h:98
AVIAMFParamDefinition::constant_subblock_duration
unsigned int constant_subblock_duration
The duration of every subblock in the case where all subblocks, with the optional exception of the la...
Definition: iamf.h:238
av_iamf_submix_add_element
AVIAMFSubmixElement * av_iamf_submix_add_element(AVIAMFSubmix *submix)
Allocate a submix element and add it to a given AVIAMFSubmix.
AVIAMFAudioElement
Information on how to combine one or more audio streams, as defined in section 3.6 of IAMF.
Definition: iamf.h:356
ipcm_decoder_config
static int ipcm_decoder_config(IAMFCodecConfig *codec_config, AVIOContext *pb, int len)
Definition: iamf_parse.c:145
ffio_ensure_seekback
int ffio_ensure_seekback(AVIOContext *s, int64_t buf_size)
Ensures that the requested seekback buffer size will be available.
Definition: aviobuf.c:1023
offset
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf offset
Definition: writing_filters.txt:86
AVIAMFMixGain
Mix Gain Parameter Data as defined in section 3.8.1 of IAMF.
Definition: iamf.h:77
leb.h
AV_RB16A
#define AV_RB16A(p)
Definition: intreadwrite.h:561
avcodec_parameters_alloc
AVCodecParameters * avcodec_parameters_alloc(void)
Allocate a new AVCodecParameters and set its fields to default values (unknown/invalid/0).
Definition: codec_par.c:56
get_bits64
static uint64_t get_bits64(GetBitContext *s, int n)
Read 0-64 bits.
Definition: get_bits.h:453
log.h
AV_CODEC_ID_NONE
@ AV_CODEC_ID_NONE
Definition: codec_id.h:50
i
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:256
AVIAMFParamDefinition::parameter_id
unsigned int parameter_id
Identifier for the paremeter substream.
Definition: iamf.h:218
avio_internal.h
AVIAMFLayer::demixing_matrix
AVRational * demixing_matrix
Demixing matrix as defined in section 3.6.3 of IAMF.
Definition: iamf.h:340
av_malloc_array
#define av_malloc_array(a, b)
Definition: tableprint_vlc.h:31
AVIAMFMixPresentation
Information on how to render and mix one or more AVIAMFAudioElement to generate the final audio outpu...
Definition: iamf.h:609
aac_decoder_config
static int aac_decoder_config(IAMFCodecConfig *codec_config, AVIOContext *pb, int len, void *logctx)
Definition: iamf_parse.c:60
FFMIN
#define FFMIN(a, b)
Definition: macros.h:49
param_definition
static int param_definition(const IAMFContext *iamf, const IAMFParamDefinition *param_def, AVIOContext *dyn_bc, void *log_ctx)
Definition: iamf_writer.c:602
AV_CODEC_ID_PCM_S32BE
@ AV_CODEC_ID_PCM_S32BE
Definition: codec_id.h:337
av_mallocz
void * av_mallocz(size_t size)
Allocate a memory block with alignment suitable for all memory accesses (including vectors if availab...
Definition: mem.c:256
len
int len
Definition: vorbis_enc_data.h:426
IAMFMixPresentation::count_label
unsigned int count_label
Definition: iamf.h:117
av_calloc
void * av_calloc(size_t nmemb, size_t size)
Definition: mem.c:264
AVIAMFParamDefinition::nb_subblocks
unsigned int nb_subblocks
Number of subblocks in the array.
Definition: iamf.h:208
AV_IAMF_AUDIO_ELEMENT_TYPE_SCENE
@ AV_IAMF_AUDIO_ELEMENT_TYPE_SCENE
Definition: iamf.h:346
AV_WB32A
#define AV_WB32A(p, v)
Definition: intreadwrite.h:578
AV_WB8
#define AV_WB8(p, d)
Definition: intreadwrite.h:392
tag
uint32_t tag
Definition: movenc.c:1879
ret
ret
Definition: filter_design.txt:187
avio_seek
int64_t avio_seek(AVIOContext *s, int64_t offset, int whence)
fseek() equivalent for AVIOContext.
Definition: aviobuf.c:231
ff_iamf_free_audio_element
void ff_iamf_free_audio_element(IAMFAudioElement **paudio_element)
Definition: iamf.c:70
avio_rb16
unsigned int avio_rb16(AVIOContext *s)
Definition: aviobuf.c:746
AV_INPUT_BUFFER_PADDING_SIZE
#define AV_INPUT_BUFFER_PADDING_SIZE
Definition: defs.h:40
left
Tag MUST be and< 10hcoeff half pel interpolation filter coefficients, hcoeff[0] are the 2 middle coefficients[1] are the next outer ones and so on, resulting in a filter like:...eff[2], hcoeff[1], hcoeff[0], hcoeff[0], hcoeff[1], hcoeff[2] ... the sign of the coefficients is not explicitly stored but alternates after each coeff and coeff[0] is positive, so ...,+,-,+,-,+,+,-,+,-,+,... hcoeff[0] is not explicitly stored but found by subtracting the sum of all stored coefficients with signs from 32 hcoeff[0]=32 - hcoeff[1] - hcoeff[2] - ... a good choice for hcoeff and htaps is htaps=6 hcoeff={40,-10, 2} an alternative which requires more computations at both encoder and decoder side and may or may not be better is htaps=8 hcoeff={42,-14, 6,-2}ref_frames minimum of the number of available reference frames and max_ref_frames for example the first frame after a key frame always has ref_frames=1spatial_decomposition_type wavelet type 0 is a 9/7 symmetric compact integer wavelet 1 is a 5/3 symmetric compact integer wavelet others are reserved stored as delta from last, last is reset to 0 if always_reset||keyframeqlog quality(logarithmic quantizer scale) stored as delta from last, last is reset to 0 if always_reset||keyframemv_scale stored as delta from last, last is reset to 0 if always_reset||keyframe FIXME check that everything works fine if this changes between framesqbias dequantization bias stored as delta from last, last is reset to 0 if always_reset||keyframeblock_max_depth maximum depth of the block tree stored as delta from last, last is reset to 0 if always_reset||keyframequant_table quantization tableHighlevel bitstream structure:==============================--------------------------------------------|Header|--------------------------------------------|------------------------------------|||Block0||||split?||||yes no||||......... intra?||||:Block01 :yes no||||:Block02 :....... ..........||||:Block03 ::y DC ::ref index:||||:Block04 ::cb DC ::motion x :||||......... :cr DC ::motion y :||||....... ..........|||------------------------------------||------------------------------------|||Block1|||...|--------------------------------------------|------------ ------------ ------------|||Y subbands||Cb subbands||Cr subbands||||--- ---||--- ---||--- ---|||||LL0||HL0||||LL0||HL0||||LL0||HL0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||LH0||HH0||||LH0||HH0||||LH0||HH0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HL1||LH1||||HL1||LH1||||HL1||LH1|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HH1||HL2||||HH1||HL2||||HH1||HL2|||||...||...||...|||------------ ------------ ------------|--------------------------------------------Decoding process:=================------------|||Subbands|------------||||------------|Intra DC||||LL0 subband prediction ------------|\ Dequantization ------------------- \||Reference frames|\ IDWT|------- -------|Motion \|||Frame 0||Frame 1||Compensation . OBMC v -------|------- -------|--------------. \------> Frame n output Frame Frame<----------------------------------/|...|------------------- Range Coder:============Binary Range Coder:------------------- The implemented range coder is an adapted version based upon "Range encoding: an algorithm for removing redundancy from a digitised message." by G. N. N. Martin. The symbols encoded by the Snow range coder are bits(0|1). The associated probabilities are not fix but change depending on the symbol mix seen so far. bit seen|new state ---------+----------------------------------------------- 0|256 - state_transition_table[256 - old_state];1|state_transition_table[old_state];state_transition_table={ 0, 0, 0, 0, 0, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 194, 194, 195, 196, 197, 198, 199, 200, 201, 202, 202, 204, 205, 206, 207, 208, 209, 209, 210, 211, 212, 213, 215, 215, 216, 217, 218, 219, 220, 220, 222, 223, 224, 225, 226, 227, 227, 229, 229, 230, 231, 232, 234, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 248, 0, 0, 0, 0, 0, 0, 0};FIXME Range Coding of integers:------------------------- FIXME Neighboring Blocks:===================left and top are set to the respective blocks unless they are outside of the image in which case they are set to the Null block top-left is set to the top left block unless it is outside of the image in which case it is set to the left block if this block has no larger parent block or it is at the left side of its parent block and the top right block is not outside of the image then the top right block is used for top-right else the top-left block is used Null block y, cb, cr are 128 level, ref, mx and my are 0 Motion Vector Prediction:=========================1. the motion vectors of all the neighboring blocks are scaled to compensate for the difference of reference frames scaled_mv=(mv *(256 *(current_reference+1)/(mv.reference+1))+128)> the median of the scaled left
Definition: snow.txt:386
U
#define U(x)
Definition: vpx_arith.h:37
IAMF_OBU_Type
IAMF_OBU_Type
Definition: iamf.h:37
label_string
static int label_string(AVIOContext *pb, char **label)
Definition: iamf_parse.c:808
IAMFMixPresentation
Definition: iamf.h:107
AV_CHANNEL_ORDER_CUSTOM
@ AV_CHANNEL_ORDER_CUSTOM
The channel order does not correspond to any other predefined order and is stored as an explicit map.
Definition: channel_layout.h:129
IAMFMixPresentation::language_label
char ** language_label
Definition: iamf.h:118
AVIAMFSubmix::default_mix_gain
AVRational default_mix_gain
Default mix gain value to apply when there are no AVIAMFParamDefinition with output_mix_config's para...
Definition: iamf.h:599
AVIAMFSubmixLayout::album_anchored_loudness
AVRational album_anchored_loudness
The Album loudness information, as defined in ITU-1770-4.
Definition: iamf.h:543
AVIAMFSubmixLayout::true_peak
AVRational true_peak
The true peak of the audio signal, as defined in ITU-1770-4.
Definition: iamf.h:535
mode
mode
Definition: ebur128.h:83
AVIAMFSubmixLayout::sound_system
AVChannelLayout sound_system
Channel layout matching one of Sound Systems A to J of ITU-2051-3, plus 7.1.2ch and 3....
Definition: iamf.h:521
IAMFCodecConfig::codec_id
enum AVCodecID codec_id
Definition: iamf.h:68
IAMFCodecConfig::codec_config_id
unsigned codec_config_id
Definition: iamf.h:67
avio_read
int avio_read(AVIOContext *s, unsigned char *buf, int size)
Read size bytes from AVIOContext into buf.
Definition: aviobuf.c:612
sign_extend
static av_const int sign_extend(int val, unsigned bits)
Definition: mathops.h:131
AVIOContext::eof_reached
int eof_reached
true if was unable to read due to error or eof
Definition: avio.h:238
IAMF_ANCHOR_ELEMENT_DIALOGUE
@ IAMF_ANCHOR_ELEMENT_DIALOGUE
Definition: iamf.h:141
AV_IAMF_PARAMETER_DEFINITION_MIX_GAIN
@ AV_IAMF_PARAMETER_DEFINITION_MIX_GAIN
Subblocks are of struct type AVIAMFMixGain.
Definition: iamf.h:173
MAX_IAMF_OBU_HEADER_SIZE
#define MAX_IAMF_OBU_HEADER_SIZE
Definition: iamf.h:34
avio_skip
int64_t avio_skip(AVIOContext *s, int64_t offset)
Skip given number of bytes forward.
Definition: aviobuf.c:318
mix_presentation_obu
static int mix_presentation_obu(void *s, IAMFContext *c, AVIOContext *pb, int len)
Definition: iamf_parse.c:825
av_iamf_audio_element_add_layer
AVIAMFLayer * av_iamf_audio_element_add_layer(AVIAMFAudioElement *audio_element)
Allocate a layer and add it to a given AVIAMFAudioElement.
av_channel_layout_copy
int av_channel_layout_copy(AVChannelLayout *dst, const AVChannelLayout *src)
Make a copy of a channel layout.
Definition: channel_layout.c:444
av_strdup
char * av_strdup(const char *s)
Duplicate a string.
Definition: mem.c:272
AV_CODEC_ID_PCM_S32LE
@ AV_CODEC_ID_PCM_S32LE
Definition: codec_id.h:336
AVIAMFAudioElement::demixing_info
AVIAMFParamDefinition * demixing_info
Demixing information used to reconstruct a scalable channel audio representation.
Definition: iamf.h:376
AVIAMFDemixingInfo::dmixp_mode
unsigned int dmixp_mode
Pre-defined combination of demixing parameters.
Definition: iamf.h:140
mem.h
AVIAMFSubmix::output_mix_config
AVIAMFParamDefinition * output_mix_config
Information required for post-processing the mixed audio signal to generate the audio signal for play...
Definition: iamf.h:591
AVIAMFLayer::ambisonics_mode
enum AVIAMFAmbisonicsMode ambisonics_mode
Ambisonics mode as defined in section 3.6.3 of IAMF.
Definition: iamf.h:328
sample_format
enum AVSampleFormat sample_format
Definition: mediacodecdec_common.c:101
flush_put_bits
static void flush_put_bits(PutBitContext *s)
Pad the end of the output stream with zeros.
Definition: put_bits.h:143
AV_CHANNEL_LAYOUT_MONO
#define AV_CHANNEL_LAYOUT_MONO
Definition: channel_layout.h:386
AV_IAMF_SUBMIX_LAYOUT_TYPE_BINAURAL
@ AV_IAMF_SUBMIX_LAYOUT_TYPE_BINAURAL
The layout is binaural.
Definition: iamf.h:501
AVIAMFLayer::flags
unsigned int flags
A bitmask which may contain a combination of AV_IAMF_LAYER_FLAG_* flags.
Definition: iamf.h:302
av_free
#define av_free(p)
Definition: tableprint_vlc.h:33
AVIAMFLayer::output_gain
AVRational output_gain
Output gain as defined in section 3.6.2 of IAMF.
Definition: iamf.h:316
AVCodecParameters::codec_id
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:55
av_dict_set
int av_dict_set(AVDictionary **pm, const char *key, const char *value, int flags)
Set the given entry in *pm, overwriting an existing entry.
Definition: dict.c:88
AVIAMFSubmixElement::element_mix_config
AVIAMFParamDefinition * element_mix_config
Information required required for applying any processing to the referenced and rendered Audio Elemen...
Definition: iamf.h:461
AV_CHAN_AMBISONIC_BASE
@ AV_CHAN_AMBISONIC_BASE
Range of channels between AV_CHAN_AMBISONIC_BASE and AV_CHAN_AMBISONIC_END represent Ambisonic compon...
Definition: channel_layout.h:105
AVIAMFParamDefinition::parameter_rate
unsigned int parameter_rate
Sample rate for the paremeter substream.
Definition: iamf.h:222
ff_iamf_scalable_ch_layouts
const AVChannelLayout ff_iamf_scalable_ch_layouts[10]
Definition: iamf.c:27
AVIAMFSubmixElement
Submix element as defined in section 3.7 of IAMF.
Definition: iamf.h:446
AV_IAMF_AUDIO_ELEMENT_TYPE_CHANNEL
@ AV_IAMF_AUDIO_ELEMENT_TYPE_CHANNEL
Definition: iamf.h:345
IAMFAudioElement::codec_config_id
unsigned int codec_config_id
Definition: iamf.h:101
av_log
#define av_log(a,...)
Definition: tableprint_vlc.h:27
AVERROR_INVALIDDATA
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
Definition: error.h:61
ffio_read_size
int ffio_read_size(AVIOContext *s, unsigned char *buf, int size)
Read size bytes from AVIOContext into buf.
Definition: aviobuf.c:662
AVIAMFAudioElement::recon_gain_info
AVIAMFParamDefinition * recon_gain_info
Recon gain information used to reconstruct a scalable channel audio representation.
Definition: iamf.h:383
IAMFMixPresentation::mix
AVIAMFMixPresentation * mix
mix backs cmix iff the AVIAMFMixPresentation is owned by this structure.
Definition: iamf.h:113
scalable_channel_layout_config
static int scalable_channel_layout_config(void *s, AVIOContext *pb, IAMFAudioElement *audio_element, const IAMFCodecConfig *codec_config)
Definition: iamf_parse.c:350
flac.h
AV_RB24
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_WL16 uint64_t_TMPL AV_WB64 unsigned int_TMPL AV_WB32 unsigned int_TMPL AV_RB24
Definition: bytestream.h:97
IAMFSubStream::codecpar
AVCodecParameters * codecpar
Definition: iamf.h:86
put_bits.h
IAMFCodecConfig::audio_roll_distance
int audio_roll_distance
Definition: iamf.h:71
IAMFMixPresentation::mix_presentation_id
unsigned int mix_presentation_id
Definition: iamf.h:114
AV_CODEC_ID_PCM_S24BE
@ AV_CODEC_ID_PCM_S24BE
Definition: codec_id.h:341
AVChannelCustom::id
enum AVChannel id
Definition: channel_layout.h:276
skip
static void BS_FUNC() skip(BSCTX *bc, unsigned int n)
Skip n bits in the buffer.
Definition: bitstream_template.h:375
AV_WL16A
#define AV_WL16A(p, v)
Definition: intreadwrite.h:557
AV_IAMF_PARAMETER_DEFINITION_DEMIXING
@ AV_IAMF_PARAMETER_DEFINITION_DEMIXING
Subblocks are of struct type AVIAMFDemixingInfo.
Definition: iamf.h:177
MPEG4AudioConfig::sample_rate
int sample_rate
Definition: mpeg4audio.h:32
Generated on Fri Aug 22 2025 13:59:28 for FFmpeg by   doxygen 1.8.17