swresample_internal.h 12.9 KB
Newer Older
Michael Niedermayer's avatar
Michael Niedermayer committed
1
/*
2
 * Copyright (C) 2011-2013 Michael Niedermayer (michaelni@gmx.at)
Michael Niedermayer's avatar
Michael Niedermayer committed
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
 *
 * This file is part of libswresample
 *
 * libswresample 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.
 *
 * libswresample 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 libswresample; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#ifndef SWR_INTERNAL_H
#define SWR_INTERNAL_H

#include "swresample.h"
25
#include "libavutil/channel_layout.h"
26
#include "config.h"
27

28
#define SWR_CH_MAX 64
29

30
#define SQRT3_2      1.22474487139158904909  /* sqrt(3/2) */
Michael Niedermayer's avatar
Michael Niedermayer committed
31

32 33
#define NS_TAPS 20

34 35 36 37 38
#if ARCH_X86_64
typedef int64_t integer;
#else
typedef int integer;
#endif
39

40 41 42 43
typedef void (mix_1_1_func_type)(void *out, const void *in, void *coeffp, integer index, integer len);
typedef void (mix_2_1_func_type)(void *out, const void *in1, const void *in2, void *coeffp, integer index1, integer index2, integer len);

typedef void (mix_any_func_type)(uint8_t **out, const uint8_t **in1, void *coeffp, integer len);
44

Michael Niedermayer's avatar
Michael Niedermayer committed
45
typedef struct AudioData{
46 47 48 49 50 51
    uint8_t *ch[SWR_CH_MAX];    ///< samples buffer per channel
    uint8_t *data;              ///< samples buffer
    int ch_count;               ///< number of channels
    int bps;                    ///< bytes per sample
    int count;                  ///< number of samples
    int planar;                 ///< 1 if planar audio, 0 otherwise
52
    enum AVSampleFormat fmt;    ///< sample format
Michael Niedermayer's avatar
Michael Niedermayer committed
53 54
} AudioData;

55
struct DitherContext {
56
    int method;
57
    int noise_pos;
58
    float scale;
59
    float noise_scale;                              ///< Noise scale
60 61 62 63 64 65 66
    int ns_taps;                                    ///< Noise shaping dither taps
    float ns_scale;                                 ///< Noise shaping dither scale
    float ns_scale_1;                               ///< Noise shaping dither scale^-1
    int ns_pos;                                     ///< Noise shaping dither position
    float ns_coeffs[NS_TAPS];                       ///< Noise shaping filter coefficients
    float ns_errors[SWR_CH_MAX][2*NS_TAPS];
    AudioData noise;                                ///< noise used for dithering
Andreas Cadhalpun's avatar
Andreas Cadhalpun committed
67
    AudioData temp;                                 ///< temporary storage when writing into the input buffer isn't possible
68
    int output_sample_bits;                         ///< the number of used output bits, needed to scale dither correctly
69 70
};

71
typedef struct ResampleContext * (* resample_init_func)(struct ResampleContext *c, int out_rate, int in_rate, int filter_size, int phase_shift, int linear,
72
                                    double cutoff, enum AVSampleFormat format, enum SwrFilterType filter_type, double kaiser_beta, double precision, int cheby);
73 74 75 76 77 78
typedef void    (* resample_free_func)(struct ResampleContext **c);
typedef int     (* multiple_resample_func)(struct ResampleContext *c, AudioData *dst, int dst_size, AudioData *src, int src_size, int *consumed);
typedef int     (* resample_flush_func)(struct SwrContext *c);
typedef int     (* set_compensation_func)(struct ResampleContext *c, int sample_delta, int compensation_distance);
typedef int64_t (* get_delay_func)(struct SwrContext *s, int64_t base);
typedef int     (* invert_initial_buffer_func)(struct ResampleContext *c, AudioData *dst, const AudioData *src, int src_size, int *dst_idx, int *dst_count);
79
typedef int64_t (* get_out_samples_func)(struct SwrContext *s, int in_samples);
80 81 82 83 84 85 86 87 88

struct Resampler {
  resample_init_func            init;
  resample_free_func            free;
  multiple_resample_func        multiple_resample;
  resample_flush_func           flush;
  set_compensation_func         set_compensation;
  get_delay_func                get_delay;
  invert_initial_buffer_func    invert_initial_buffer;
89
  get_out_samples_func          get_out_samples;
90 91 92
};

extern struct Resampler const swri_resampler;
93
extern struct Resampler const swri_soxr_resampler;
94

95
struct SwrContext {
96 97 98 99
    const AVClass *av_class;                        ///< AVClass used for AVOption and av_log()
    int log_level_offset;                           ///< logging level offset
    void *log_ctx;                                  ///< parent logging context
    enum AVSampleFormat  in_sample_fmt;             ///< input sample format
100
    enum AVSampleFormat int_sample_fmt;             ///< internal sample format (AV_SAMPLE_FMT_FLTP or AV_SAMPLE_FMT_S16P)
101 102 103 104 105 106
    enum AVSampleFormat out_sample_fmt;             ///< output sample format
    int64_t  in_ch_layout;                          ///< input channel layout
    int64_t out_ch_layout;                          ///< output channel layout
    int      in_sample_rate;                        ///< input sample rate
    int     out_sample_rate;                        ///< output sample rate
    int flags;                                      ///< miscellaneous flags such as SWR_FLAG_RESAMPLE
107
    float slev;                                     ///< surround mixing level
108
    float clev;                                     ///< center mixing level
Justin Ruggles's avatar
Justin Ruggles committed
109
    float lfe_mix_level;                            ///< LFE mixing level
110
    float rematrix_volume;                          ///< rematrixing volume coefficient
111
    float rematrix_maxval;                          ///< maximum value for rematrixing output
112
    int matrix_encoding;                            /**< matrixed stereo encoding */
113 114
    const int *channel_map;                         ///< channel index (or -1 if muted channel) map
    int used_ch_count;                              ///< number of used input channels (mapped channel count if channel_map, otherwise in.ch_count)
115
    int engine;
116

117 118 119
    int user_in_ch_count;                           ///< User set input channel count
    int user_out_ch_count;                          ///< User set output channel count
    int user_used_ch_count;                         ///< User set used channel count
120 121
    int64_t user_in_ch_layout;                      ///< User set input channel layout
    int64_t user_out_ch_layout;                     ///< User set output channel layout
122
    enum AVSampleFormat user_int_sample_fmt;        ///< User set internal sample format
123

124
    struct DitherContext dither;
125

126 127 128
    int filter_size;                                /**< length of each FIR filter in the resampling filterbank relative to the cutoff frequency */
    int phase_shift;                                /**< log2 of the number of entries in the resampling polyphase filterbank */
    int linear_interp;                              /**< if 1 then the resampling FIR filter will be linearly interpolated */
Rob Sykes's avatar
Rob Sykes committed
129
    double cutoff;                                  /**< resampling cutoff frequency (swr: 6dB point; soxr: 0dB point). 1.0 corresponds to half the output sample rate */
130
    int filter_type;                                /**< swr resampling filter type */
131
    double kaiser_beta;                                /**< swr beta value for Kaiser window (only applicable if filter_type == AV_FILTER_TYPE_KAISER) */
Rob Sykes's avatar
Rob Sykes committed
132 133
    double precision;                               /**< soxr resampling precision (in bits) */
    int cheby;                                      /**< soxr: if 1 then passband rolloff will be none (Chebyshev) & irrational ratio approximation precision will be higher */
Michael Niedermayer's avatar
Michael Niedermayer committed
134

Rob Sykes's avatar
Rob Sykes committed
135 136 137 138 139
    float min_compensation;                         ///< swr minimum below which no compensation will happen
    float min_hard_compensation;                    ///< swr minimum below which no silence inject / sample drop will happen
    float soft_compensation_duration;               ///< swr duration over which soft compensation is applied
    float max_soft_compensation;                    ///< swr maximum soft compensation in seconds over soft_compensation_duration
    float async;                                    ///< swr simple 1 parameter async, similar to ffmpegs -async
140
    int64_t firstpts_in_samples;                    ///< swr first pts in samples
141

142 143
    int resample_first;                             ///< 1 if resampling must come first, 0 if rematrixing
    int rematrix;                                   ///< flag to indicate if rematrixing is needed (basically if input and output layouts mismatch)
144
    int rematrix_custom;                            ///< flag to indicate that a custom matrix has been defined
Michael Niedermayer's avatar
Michael Niedermayer committed
145

146 147 148 149 150 151
    AudioData in;                                   ///< input audio data
    AudioData postin;                               ///< post-input audio data: used for rematrix/resample
    AudioData midbuf;                               ///< intermediate audio data (postin/preout)
    AudioData preout;                               ///< pre-output audio data: used for rematrix/resample
    AudioData out;                                  ///< converted output audio data
    AudioData in_buffer;                            ///< cached audio data (convert and resample purpose)
152
    AudioData silence;                              ///< temporary with silence
153
    AudioData drop_temp;                            ///< temporary used to discard output
154 155 156
    int in_buffer_index;                            ///< cached buffer position
    int in_buffer_count;                            ///< cached buffer length
    int resample_in_constraint;                     ///< 1 if the input end was reach before the output end, 0 otherwise
157
    int flushed;                                    ///< 1 if data is to be flushed and no further input is expected
158
    int64_t outpts;                                 ///< output PTS
159
    int64_t firstpts;                               ///< first PTS
160
    int drop_output;                                ///< number of output samples to drop
161
    double delayed_samples_fixup;                   ///< soxr 0.1.1: needed to fixup delayed_samples after flush has been called.
Michael Niedermayer's avatar
Michael Niedermayer committed
162

163 164 165 166
    struct AudioConvert *in_convert;                ///< input conversion context
    struct AudioConvert *out_convert;               ///< output conversion context
    struct AudioConvert *full_convert;              ///< full conversion context (single conversion for input and output)
    struct ResampleContext *resample;               ///< resampling context
167
    struct Resampler const *resampler;              ///< resampler virtual function table
Michael Niedermayer's avatar
Michael Niedermayer committed
168

169
    float matrix[SWR_CH_MAX][SWR_CH_MAX];           ///< floating point rematrixing coefficients
170 171
    uint8_t *native_matrix;
    uint8_t *native_one;
172
    uint8_t *native_simd_one;
173
    uint8_t *native_simd_matrix;
174 175
    int32_t matrix32[SWR_CH_MAX][SWR_CH_MAX];       ///< 17.15 fixed point rematrixing coefficients
    uint8_t matrix_ch[SWR_CH_MAX][SWR_CH_MAX+1];    ///< Lists of input channels per output channel that have non zero rematrixing coefficients
176
    mix_1_1_func_type *mix_1_1_f;
177 178
    mix_1_1_func_type *mix_1_1_simd;

179
    mix_2_1_func_type *mix_2_1_f;
180
    mix_2_1_func_type *mix_2_1_simd;
Michael Niedermayer's avatar
Michael Niedermayer committed
181

182 183
    mix_any_func_type *mix_any_f;

184
    /* TODO: callbacks for ASM optimizations */
185
};
Michael Niedermayer's avatar
Michael Niedermayer committed
186

187
av_warn_unused_result
188
int swri_realloc_audio(AudioData *a, int count);
Michael Niedermayer's avatar
Michael Niedermayer committed
189

190 191 192 193
void swri_noise_shaping_int16 (SwrContext *s, AudioData *dsts, const AudioData *srcs, const AudioData *noises, int count);
void swri_noise_shaping_int32 (SwrContext *s, AudioData *dsts, const AudioData *srcs, const AudioData *noises, int count);
void swri_noise_shaping_float (SwrContext *s, AudioData *dsts, const AudioData *srcs, const AudioData *noises, int count);
void swri_noise_shaping_double(SwrContext *s, AudioData *dsts, const AudioData *srcs, const AudioData *noises, int count);
194

195
av_warn_unused_result
196
int swri_rematrix_init(SwrContext *s);
197
void swri_rematrix_free(SwrContext *s);
198
int swri_rematrix(SwrContext *s, AudioData *out, AudioData *in, int len, int mustcopy);
199
int swri_rematrix_init_x86(struct SwrContext *s);
200

201
av_warn_unused_result
202
int swri_get_dither(SwrContext *s, void *dst, int len, unsigned seed, enum AVSampleFormat noise_fmt);
203
av_warn_unused_result
204
int swri_dither_init(SwrContext *s, enum AVSampleFormat out_fmt, enum AVSampleFormat in_fmt);
205

206 207 208 209
void swri_audio_convert_init_aarch64(struct AudioConvert *ac,
                                 enum AVSampleFormat out_fmt,
                                 enum AVSampleFormat in_fmt,
                                 int channels);
210 211 212 213
void swri_audio_convert_init_arm(struct AudioConvert *ac,
                                 enum AVSampleFormat out_fmt,
                                 enum AVSampleFormat in_fmt,
                                 int channels);
214 215 216 217
void swri_audio_convert_init_x86(struct AudioConvert *ac,
                                 enum AVSampleFormat out_fmt,
                                 enum AVSampleFormat in_fmt,
                                 int channels);
218

Michael Niedermayer's avatar
Michael Niedermayer committed
219
#endif