1/* 2 * Copyright (C) 2016 foo86 3 * 4 * This file is part of FFmpeg. 5 * 6 * FFmpeg is free software; you can redistribute it and/or 7 * modify it under the terms of the GNU Lesser General Public 8 * License as published by the Free Software Foundation; either 9 * version 2.1 of the License, or (at your option) any later version. 10 * 11 * FFmpeg is distributed in the hope that it will be useful, 12 * but WITHOUT ANY WARRANTY; without even the implied warranty of 13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 14 * Lesser General Public License for more details. 15 * 16 * You should have received a copy of the GNU Lesser General Public 17 * License along with FFmpeg; if not, write to the Free Software 18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA 19 */ 20 21#ifndef AVCODEC_DCA_CORE_H 22#define AVCODEC_DCA_CORE_H 23 24#include "libavutil/float_dsp.h" 25#include "libavutil/fixed_dsp.h" 26#include "libavutil/mem_internal.h" 27 28#include "avcodec.h" 29#include "get_bits.h" 30#include "dca.h" 31#include "dca_exss.h" 32#include "dcadsp.h" 33#include "dcadct.h" 34#include "dcamath.h" 35#include "dcahuff.h" 36#include "fft.h" 37#include "synth_filter.h" 38 39#define DCA_CHANNELS 7 40#define DCA_SUBBANDS 32 41#define DCA_SUBBANDS_X96 64 42#define DCA_SUBFRAMES 16 43#define DCA_SUBBAND_SAMPLES 8 44#define DCA_PCMBLOCK_SAMPLES 32 45#define DCA_LFE_HISTORY 8 46#define DCA_ABITS_MAX 26 47 48#define DCA_CORE_CHANNELS_MAX 6 49#define DCA_DMIX_CHANNELS_MAX 4 50#define DCA_XXCH_CHANNELS_MAX 2 51#define DCA_EXSS_CHANNELS_MAX 8 52#define DCA_EXSS_CHSETS_MAX 4 53 54#define DCA_FILTER_MODE_X96 0x01 55#define DCA_FILTER_MODE_FIXED 0x02 56 57enum DCACoreAudioMode { 58 DCA_AMODE_MONO, // Mode 0: A (mono) 59 DCA_AMODE_MONO_DUAL, // Mode 1: A + B (dual mono) 60 DCA_AMODE_STEREO, // Mode 2: L + R (stereo) 61 DCA_AMODE_STEREO_SUMDIFF, // Mode 3: (L+R) + (L-R) (sum-diff) 62 DCA_AMODE_STEREO_TOTAL, // Mode 4: LT + RT (left and right total) 63 DCA_AMODE_3F, // Mode 5: C + L + R 64 DCA_AMODE_2F1R, // Mode 6: L + R + S 65 DCA_AMODE_3F1R, // Mode 7: C + L + R + S 66 DCA_AMODE_2F2R, // Mode 8: L + R + SL + SR 67 DCA_AMODE_3F2R, // Mode 9: C + L + R + SL + SR 68 69 DCA_AMODE_COUNT 70}; 71 72enum DCACoreExtAudioType { 73 DCA_EXT_AUDIO_XCH = 0, 74 DCA_EXT_AUDIO_X96 = 2, 75 DCA_EXT_AUDIO_XXCH = 6 76}; 77 78enum DCACoreLFEFlag { 79 DCA_LFE_FLAG_NONE, 80 DCA_LFE_FLAG_128, 81 DCA_LFE_FLAG_64, 82 DCA_LFE_FLAG_INVALID 83}; 84 85typedef struct DCADSPData { 86 union { 87 struct { 88 DECLARE_ALIGNED(32, float, hist1)[1024]; 89 DECLARE_ALIGNED(32, float, hist2)[64]; 90 } flt; 91 struct { 92 DECLARE_ALIGNED(32, int32_t, hist1)[1024]; 93 DECLARE_ALIGNED(32, int32_t, hist2)[64]; 94 } fix; 95 } u; 96 int offset; 97} DCADSPData; 98 99typedef struct DCACoreDecoder { 100 AVCodecContext *avctx; 101 GetBitContext gb; 102 GetBitContext gb_in; 103 104 // Bit stream header 105 int crc_present; ///< CRC present flag 106 int npcmblocks; ///< Number of PCM sample blocks 107 int frame_size; ///< Primary frame byte size 108 int audio_mode; ///< Audio channel arrangement 109 int sample_rate; ///< Core audio sampling frequency 110 int bit_rate; ///< Transmission bit rate 111 int drc_present; ///< Embedded dynamic range flag 112 int ts_present; ///< Embedded time stamp flag 113 int aux_present; ///< Auxiliary data flag 114 int ext_audio_type; ///< Extension audio descriptor flag 115 int ext_audio_present; ///< Extended coding flag 116 int sync_ssf; ///< Audio sync word insertion flag 117 int lfe_present; ///< Low frequency effects flag 118 int predictor_history; ///< Predictor history flag switch 119 int filter_perfect; ///< Multirate interpolator switch 120 int source_pcm_res; ///< Source PCM resolution 121 int es_format; ///< Extended surround (ES) mastering flag 122 int sumdiff_front; ///< Front sum/difference flag 123 int sumdiff_surround; ///< Surround sum/difference flag 124 125 // Primary audio coding header 126 int nsubframes; ///< Number of subframes 127 int nchannels; ///< Number of primary audio channels (incl. extension channels) 128 int ch_mask; ///< Speaker layout mask (incl. LFE and extension channels) 129 int8_t nsubbands[DCA_CHANNELS]; ///< Subband activity count 130 int8_t subband_vq_start[DCA_CHANNELS]; ///< High frequency VQ start subband 131 int8_t joint_intensity_index[DCA_CHANNELS]; ///< Joint intensity coding index 132 int8_t transition_mode_sel[DCA_CHANNELS]; ///< Transient mode code book 133 int8_t scale_factor_sel[DCA_CHANNELS]; ///< Scale factor code book 134 int8_t bit_allocation_sel[DCA_CHANNELS]; ///< Bit allocation quantizer select 135 int8_t quant_index_sel[DCA_CHANNELS][DCA_CODE_BOOKS]; ///< Quantization index codebook select 136 int32_t scale_factor_adj[DCA_CHANNELS][DCA_CODE_BOOKS]; ///< Scale factor adjustment 137 138 // Primary audio coding side information 139 int8_t nsubsubframes[DCA_SUBFRAMES]; ///< Subsubframe count for each subframe 140 int8_t prediction_mode[DCA_CHANNELS][DCA_SUBBANDS_X96]; ///< Prediction mode 141 int16_t prediction_vq_index[DCA_CHANNELS][DCA_SUBBANDS_X96]; ///< Prediction coefficients VQ address 142 int8_t bit_allocation[DCA_CHANNELS][DCA_SUBBANDS_X96]; ///< Bit allocation index 143 int8_t transition_mode[DCA_SUBFRAMES][DCA_CHANNELS][DCA_SUBBANDS]; ///< Transition mode 144 int32_t scale_factors[DCA_CHANNELS][DCA_SUBBANDS][2]; ///< Scale factors (2x for transients and X96) 145 int8_t joint_scale_sel[DCA_CHANNELS]; ///< Joint subband codebook select 146 int32_t joint_scale_factors[DCA_CHANNELS][DCA_SUBBANDS_X96]; ///< Scale factors for joint subband coding 147 148 // Auxiliary data 149 int prim_dmix_embedded; ///< Auxiliary dynamic downmix flag 150 int prim_dmix_type; ///< Auxiliary primary channel downmix type 151 int prim_dmix_coeff[DCA_DMIX_CHANNELS_MAX * DCA_CORE_CHANNELS_MAX]; ///< Dynamic downmix code coefficients 152 153 // Core extensions 154 int ext_audio_mask; ///< Bit mask of fully decoded core extensions 155 156 // XCH extension data 157 int xch_pos; ///< Bit position of XCH frame in core substream 158 159 // XXCH extension data 160 int xxch_crc_present; ///< CRC presence flag for XXCH channel set header 161 int xxch_mask_nbits; ///< Number of bits for loudspeaker mask 162 int xxch_core_mask; ///< Core loudspeaker activity mask 163 int xxch_spkr_mask; ///< Loudspeaker layout mask 164 int xxch_dmix_embedded; ///< Downmix already performed by encoder 165 int xxch_dmix_scale_inv; ///< Downmix scale factor 166 int xxch_dmix_mask[DCA_XXCH_CHANNELS_MAX]; ///< Downmix channel mapping mask 167 int xxch_dmix_coeff[DCA_XXCH_CHANNELS_MAX * DCA_CORE_CHANNELS_MAX]; ///< Downmix coefficients 168 int xxch_pos; ///< Bit position of XXCH frame in core substream 169 170 // X96 extension data 171 int x96_rev_no; ///< X96 revision number 172 int x96_crc_present; ///< CRC presence flag for X96 channel set header 173 int x96_nchannels; ///< Number of primary channels in X96 extension 174 int x96_high_res; ///< X96 high resolution flag 175 int x96_subband_start; ///< First encoded subband in X96 extension 176 int x96_rand; ///< Random seed for generating samples for unallocated X96 subbands 177 int x96_pos; ///< Bit position of X96 frame in core substream 178 179 // Sample buffers 180 unsigned int x96_subband_size; 181 int32_t *x96_subband_buffer; ///< X96 subband sample buffer base 182 int32_t *x96_subband_samples[DCA_CHANNELS][DCA_SUBBANDS_X96]; ///< X96 subband samples 183 184 unsigned int subband_size; 185 int32_t *subband_buffer; ///< Subband sample buffer base 186 int32_t *subband_samples[DCA_CHANNELS][DCA_SUBBANDS]; ///< Subband samples 187 int32_t *lfe_samples; ///< Decimated LFE samples 188 189 // DSP contexts 190 DCADSPData dcadsp_data[DCA_CHANNELS]; ///< FIR history buffers 191 DCADSPContext *dcadsp; 192 DCADCTContext dcadct; 193 FFTContext imdct[2]; 194 SynthFilterContext synth; 195 AVFloatDSPContext *float_dsp; 196 AVFixedDSPContext *fixed_dsp; 197 198 // PCM output data 199 unsigned int output_size; 200 void *output_buffer; ///< PCM output buffer base 201 int32_t *output_samples[DCA_SPEAKER_COUNT]; ///< PCM output for fixed point mode 202 int32_t output_history_lfe_fixed; ///< LFE PCM history for X96 filter 203 float output_history_lfe_float; ///< LFE PCM history for X96 filter 204 205 int ch_remap[DCA_SPEAKER_COUNT]; ///< Channel to speaker map 206 int request_mask; ///< Requested channel layout (for stereo downmix) 207 208 int npcmsamples; ///< Number of PCM samples per channel 209 int output_rate; ///< Output sample rate (1x or 2x header rate) 210 211 int filter_mode; ///< Previous filtering mode for detecting changes 212} DCACoreDecoder; 213 214static inline int ff_dca_core_map_spkr(DCACoreDecoder *core, int spkr) 215{ 216 if (core->ch_mask & (1U << spkr)) 217 return spkr; 218 if (spkr == DCA_SPEAKER_Lss && (core->ch_mask & DCA_SPEAKER_MASK_Ls)) 219 return DCA_SPEAKER_Ls; 220 if (spkr == DCA_SPEAKER_Rss && (core->ch_mask & DCA_SPEAKER_MASK_Rs)) 221 return DCA_SPEAKER_Rs; 222 return -1; 223} 224 225static inline void ff_dca_core_dequantize(int32_t *output, const int32_t *input, 226 int32_t step_size, int32_t scale, int residual, int len) 227{ 228 // Account for quantizer step size 229 int64_t step_scale = (int64_t)step_size * scale; 230 int n, shift = 0; 231 232 // Limit scale factor resolution to 22 bits 233 if (step_scale > (1 << 23)) { 234 shift = av_log2(step_scale >> 23) + 1; 235 step_scale >>= shift; 236 } 237 238 // Scale the samples 239 if (residual) { 240 for (n = 0; n < len; n++) 241 output[n] += clip23(norm__(input[n] * step_scale, 22 - shift)); 242 } else { 243 for (n = 0; n < len; n++) 244 output[n] = clip23(norm__(input[n] * step_scale, 22 - shift)); 245 } 246} 247 248int ff_dca_core_parse(DCACoreDecoder *s, const uint8_t *data, int size); 249int ff_dca_core_parse_exss(DCACoreDecoder *s, const uint8_t *data, DCAExssAsset *asset); 250int ff_dca_core_filter_fixed(DCACoreDecoder *s, int x96_synth); 251int ff_dca_core_filter_frame(DCACoreDecoder *s, AVFrame *frame); 252av_cold void ff_dca_core_flush(DCACoreDecoder *s); 253av_cold int ff_dca_core_init(DCACoreDecoder *s); 254av_cold void ff_dca_core_close(DCACoreDecoder *s); 255 256#endif 257