static av_cold void opus_decode_flush(AVCodecContext *ctx)
{
OpusContext *c = ctx->priv_data;
int i;
for (i = 0; i < c->nb_streams; i++) {
OpusStreamContext *s = &c->streams[i];
memset(&s->packet, 0, sizeof(s->packet));
s->delayed_samples = 0;
if (s->celt_delay)
av_audio_fifo_drain(s->celt_delay, av_audio_fifo_size(s->celt_delay));
#if CONFIG_SWRESAMPLE
swr_close(s->swr);
#elif CONFIG_AVRESAMPLE
avresample_close(s->avr);
#endif
av_audio_fifo_drain(c->sync_buffers[i], av_audio_fifo_size(c->sync_buffers[i]));
ff_silk_flush(s->silk);
ff_celt_flush(s->celt);
}
}
void AudioSource::init(void)
{
position = 0;
samples_frac = 0;
queue_index = 0;
#if defined(LIBTAS_ENABLE_AVDUMPING) || defined(LIBTAS_ENABLE_SOUNDPLAYBACK)
if (swr_is_initialized(swr))
swr_close(swr);
#endif
}
static int control(struct af_instance *af, int cmd, void *arg)
{
struct af_resample *s = af->priv;
switch (cmd) {
case AF_CONTROL_REINIT: {
struct mp_audio *in = arg;
struct mp_audio *out = af->data;
struct mp_audio orig_in = *in;
if (((out->rate == in->rate) || (out->rate == 0)) &&
(out->format == in->format) &&
(mp_chmap_equals(&out->channels, &in->channels) || out->nch == 0) &&
s->allow_detach && s->playback_speed == 1.0)
return AF_DETACH;
if (out->rate == 0)
out->rate = in->rate;
if (mp_chmap_is_empty(&out->channels))
mp_audio_set_channels(out, &in->channels);
if (af_to_avformat(in->format) == AV_SAMPLE_FMT_NONE)
mp_audio_set_format(in, AF_FORMAT_FLOAT);
if (check_output_conversion(out->format) == AV_SAMPLE_FMT_NONE)
mp_audio_set_format(out, in->format);
int r = ((in->format == orig_in.format) &&
mp_chmap_equals(&in->channels, &orig_in.channels))
? AF_OK : AF_FALSE;
if (r == AF_OK)
r = configure_lavrr(af, in, out, true);
return r;
}
case AF_CONTROL_SET_PLAYBACK_SPEED_RESAMPLE: {
s->playback_speed = *(double *)arg;
return AF_OK;
}
case AF_CONTROL_RESET:
if (s->avrctx) {
#if HAVE_LIBSWRESAMPLE
swr_close(s->avrctx);
if (swr_init(s->avrctx) < 0) {
close_lavrr(af);
return AF_ERROR;
}
#else
while (avresample_read(s->avrctx, NULL, 1000) > 0) {}
#endif
}
return AF_OK;
}
return AF_UNKNOWN;
}
static av_cold void opus_decode_flush(AVCodecContext *ctx)
{
OpusContext *c = ctx->priv_data;
int i;
for (i = 0; i < c->nb_streams; i++) {
OpusStreamContext *s = &c->streams[i];
memset(&s->packet, 0, sizeof(s->packet));
s->delayed_samples = 0;
if (s->celt_delay)
av_audio_fifo_drain(s->celt_delay, av_audio_fifo_size(s->celt_delay));
swr_close(s->swr);
ff_silk_flush(s->silk);
ff_celt_flush(s->celt);
}
}
av_cold int swr_init(struct SwrContext *s){
int ret;
char l1[1024], l2[1024];
clear_context(s);
if(s-> in_sample_fmt >= AV_SAMPLE_FMT_NB){
av_log(s, AV_LOG_ERROR, "Requested input sample format %d is invalid\n", s->in_sample_fmt);
return AVERROR(EINVAL);
}
if(s->out_sample_fmt >= AV_SAMPLE_FMT_NB){
av_log(s, AV_LOG_ERROR, "Requested output sample format %d is invalid\n", s->out_sample_fmt);
return AVERROR(EINVAL);
}
s->out.ch_count = s-> user_out_ch_count;
s-> in.ch_count = s-> user_in_ch_count;
s->used_ch_count = s->user_used_ch_count;
s-> in_ch_layout = s-> user_in_ch_layout;
s->out_ch_layout = s->user_out_ch_layout;
s->int_sample_fmt= s->user_int_sample_fmt;
if(av_get_channel_layout_nb_channels(s-> in_ch_layout) > SWR_CH_MAX) {
av_log(s, AV_LOG_WARNING, "Input channel layout 0x%"PRIx64" is invalid or unsupported.\n", s-> in_ch_layout);
s->in_ch_layout = 0;
}
if(av_get_channel_layout_nb_channels(s->out_ch_layout) > SWR_CH_MAX) {
av_log(s, AV_LOG_WARNING, "Output channel layout 0x%"PRIx64" is invalid or unsupported.\n", s->out_ch_layout);
s->out_ch_layout = 0;
}
switch(s->engine){
#if CONFIG_LIBSOXR
extern struct Resampler const soxr_resampler;
case SWR_ENGINE_SOXR: s->resampler = &soxr_resampler; break;
#endif
case SWR_ENGINE_SWR : s->resampler = &swri_resampler; break;
default:
av_log(s, AV_LOG_ERROR, "Requested resampling engine is unavailable\n");
return AVERROR(EINVAL);
}
if(!s->used_ch_count)
s->used_ch_count= s->in.ch_count;
if(s->used_ch_count && s-> in_ch_layout && s->used_ch_count != av_get_channel_layout_nb_channels(s-> in_ch_layout)){
av_log(s, AV_LOG_WARNING, "Input channel layout has a different number of channels than the number of used channels, ignoring layout\n");
s-> in_ch_layout= 0;
}
if(!s-> in_ch_layout)
s-> in_ch_layout= av_get_default_channel_layout(s->used_ch_count);
if(!s->out_ch_layout)
s->out_ch_layout= av_get_default_channel_layout(s->out.ch_count);
s->rematrix= s->out_ch_layout !=s->in_ch_layout || s->rematrix_volume!=1.0 ||
s->rematrix_custom;
if(s->int_sample_fmt == AV_SAMPLE_FMT_NONE){
if(av_get_planar_sample_fmt(s->in_sample_fmt) <= AV_SAMPLE_FMT_S16P){
s->int_sample_fmt= AV_SAMPLE_FMT_S16P;
}else if( av_get_planar_sample_fmt(s-> in_sample_fmt) == AV_SAMPLE_FMT_S32P
&& av_get_planar_sample_fmt(s->out_sample_fmt) == AV_SAMPLE_FMT_S32P
&& !s->rematrix
&& s->engine != SWR_ENGINE_SOXR){
s->int_sample_fmt= AV_SAMPLE_FMT_S32P;
}else if(av_get_planar_sample_fmt(s->in_sample_fmt) <= AV_SAMPLE_FMT_FLTP){
s->int_sample_fmt= AV_SAMPLE_FMT_FLTP;
}else{
av_log(s, AV_LOG_DEBUG, "Using double precision mode\n");
s->int_sample_fmt= AV_SAMPLE_FMT_DBLP;
}
}
if( s->int_sample_fmt != AV_SAMPLE_FMT_S16P
&&s->int_sample_fmt != AV_SAMPLE_FMT_S32P
&&s->int_sample_fmt != AV_SAMPLE_FMT_FLTP
&&s->int_sample_fmt != AV_SAMPLE_FMT_DBLP){
av_log(s, AV_LOG_ERROR, "Requested sample format %s is not supported internally, S16/S32/FLT/DBL is supported\n", av_get_sample_fmt_name(s->int_sample_fmt));
return AVERROR(EINVAL);
}
set_audiodata_fmt(&s-> in, s-> in_sample_fmt);
set_audiodata_fmt(&s->out, s->out_sample_fmt);
if (s->firstpts_in_samples != AV_NOPTS_VALUE) {
if (!s->async && s->min_compensation >= FLT_MAX/2)
s->async = 1;
s->firstpts =
s->outpts = s->firstpts_in_samples * s->out_sample_rate;
} else
s->firstpts = AV_NOPTS_VALUE;
if (s->async) {
if (s->min_compensation >= FLT_MAX/2)
s->min_compensation = 0.001;
if (s->async > 1.0001) {
s->max_soft_compensation = s->async / (double) s->in_sample_rate;
}
}
if (s->out_sample_rate!=s->in_sample_rate || (s->flags & SWR_FLAG_RESAMPLE)){
s->resample = s->resampler->init(s->resample, s->out_sample_rate, s->in_sample_rate, s->filter_size, s->phase_shift, s->linear_interp, s->cutoff, s->int_sample_fmt, s->filter_type, s->kaiser_beta, s->precision, s->cheby);
if (!s->resample) {
av_log(s, AV_LOG_ERROR, "Failed to initilaize resampler\n");
return AVERROR(ENOMEM);
}
}else
s->resampler->free(&s->resample);
if( s->int_sample_fmt != AV_SAMPLE_FMT_S16P
&& s->int_sample_fmt != AV_SAMPLE_FMT_S32P
&& s->int_sample_fmt != AV_SAMPLE_FMT_FLTP
&& s->int_sample_fmt != AV_SAMPLE_FMT_DBLP
&& s->resample){
av_log(s, AV_LOG_ERROR, "Resampling only supported with internal s16/s32/flt/dbl\n");
ret = AVERROR(EINVAL);
goto fail;
}
#define RSC 1 //FIXME finetune
if(!s-> in.ch_count)
s-> in.ch_count= av_get_channel_layout_nb_channels(s-> in_ch_layout);
if(!s->used_ch_count)
s->used_ch_count= s->in.ch_count;
if(!s->out.ch_count)
s->out.ch_count= av_get_channel_layout_nb_channels(s->out_ch_layout);
if(!s-> in.ch_count){
av_assert0(!s->in_ch_layout);
av_log(s, AV_LOG_ERROR, "Input channel count and layout are unset\n");
ret = AVERROR(EINVAL);
goto fail;
}
av_get_channel_layout_string(l1, sizeof(l1), s-> in.ch_count, s-> in_ch_layout);
av_get_channel_layout_string(l2, sizeof(l2), s->out.ch_count, s->out_ch_layout);
if (s->out_ch_layout && s->out.ch_count != av_get_channel_layout_nb_channels(s->out_ch_layout)) {
av_log(s, AV_LOG_ERROR, "Output channel layout %s mismatches specified channel count %d\n", l2, s->out.ch_count);
ret = AVERROR(EINVAL);
goto fail;
}
if (s->in_ch_layout && s->used_ch_count != av_get_channel_layout_nb_channels(s->in_ch_layout)) {
av_log(s, AV_LOG_ERROR, "Input channel layout %s mismatches specified channel count %d\n", l1, s->used_ch_count);
ret = AVERROR(EINVAL);
goto fail;
}
if ((!s->out_ch_layout || !s->in_ch_layout) && s->used_ch_count != s->out.ch_count && !s->rematrix_custom) {
av_log(s, AV_LOG_ERROR, "Rematrix is needed between %s and %s "
"but there is not enough information to do it\n", l1, l2);
ret = AVERROR(EINVAL);
goto fail;
}
av_assert0(s->used_ch_count);
av_assert0(s->out.ch_count);
s->resample_first= RSC*s->out.ch_count/s->in.ch_count - RSC < s->out_sample_rate/(float)s-> in_sample_rate - 1.0;
s->in_buffer= s->in;
s->silence = s->in;
s->drop_temp= s->out;
if(!s->resample && !s->rematrix && !s->channel_map && !s->dither.method){
s->full_convert = swri_audio_convert_alloc(s->out_sample_fmt,
s-> in_sample_fmt, s-> in.ch_count, NULL, 0);
return 0;
}
s->in_convert = swri_audio_convert_alloc(s->int_sample_fmt,
s-> in_sample_fmt, s->used_ch_count, s->channel_map, 0);
s->out_convert= swri_audio_convert_alloc(s->out_sample_fmt,
s->int_sample_fmt, s->out.ch_count, NULL, 0);
if (!s->in_convert || !s->out_convert) {
ret = AVERROR(ENOMEM);
goto fail;
}
s->postin= s->in;
s->preout= s->out;
s->midbuf= s->in;
if(s->channel_map){
s->postin.ch_count=
s->midbuf.ch_count= s->used_ch_count;
if(s->resample)
s->in_buffer.ch_count= s->used_ch_count;
}
if(!s->resample_first){
s->midbuf.ch_count= s->out.ch_count;
if(s->resample)
s->in_buffer.ch_count = s->out.ch_count;
}
set_audiodata_fmt(&s->postin, s->int_sample_fmt);
set_audiodata_fmt(&s->midbuf, s->int_sample_fmt);
set_audiodata_fmt(&s->preout, s->int_sample_fmt);
if(s->resample){
set_audiodata_fmt(&s->in_buffer, s->int_sample_fmt);
}
if ((ret = swri_dither_init(s, s->out_sample_fmt, s->int_sample_fmt)) < 0)
goto fail;
if(s->rematrix || s->dither.method) {
ret = swri_rematrix_init(s);
if (ret < 0)
goto fail;
}
return 0;
fail:
swr_close(s);
return ret;
}
static int opus_decode_subpacket(OpusStreamContext *s,
const uint8_t *buf, int buf_size,
int nb_samples)
{
int output_samples = 0;
int flush_needed = 0;
int i, j, ret;
/* check if we need to flush the resampler */
if (swr_is_initialized(s->swr)) {
if (buf) {
int64_t cur_samplerate;
av_opt_get_int(s->swr, "in_sample_rate", 0, &cur_samplerate);
flush_needed = (s->packet.mode == OPUS_MODE_CELT) || (cur_samplerate != s->silk_samplerate);
} else {
flush_needed = !!s->delayed_samples;
}
}
if (!buf && !flush_needed)
return 0;
/* use dummy output buffers if the channel is not mapped to anything */
if (!s->out[0] ||
(s->output_channels == 2 && !s->out[1])) {
av_fast_malloc(&s->out_dummy, &s->out_dummy_allocated_size, s->out_size);
if (!s->out_dummy)
return AVERROR(ENOMEM);
if (!s->out[0])
s->out[0] = s->out_dummy;
if (!s->out[1])
s->out[1] = s->out_dummy;
}
/* flush the resampler if necessary */
if (flush_needed) {
ret = opus_flush_resample(s, s->delayed_samples);
if (ret < 0) {
av_log(s->avctx, AV_LOG_ERROR, "Error flushing the resampler.\n");
return ret;
}
swr_close(s->swr);
output_samples += s->delayed_samples;
s->delayed_samples = 0;
if (!buf)
goto finish;
}
/* decode all the frames in the packet */
for (i = 0; i < s->packet.frame_count; i++) {
int size = s->packet.frame_size[i];
int samples = opus_decode_frame(s, buf + s->packet.frame_offset[i], size);
if (samples < 0) {
av_log(s->avctx, AV_LOG_ERROR, "Error decoding an Opus frame.\n");
if (s->avctx->err_recognition & AV_EF_EXPLODE)
return samples;
for (j = 0; j < s->output_channels; j++)
memset(s->out[j], 0, s->packet.frame_duration * sizeof(float));
samples = s->packet.frame_duration;
}
output_samples += samples;
for (j = 0; j < s->output_channels; j++)
s->out[j] += samples;
s->out_size -= samples * sizeof(float);
}
finish:
s->out[0] = s->out[1] = NULL;
s->out_size = 0;
return output_samples;
}
static int opus_decode_frame(OpusStreamContext *s, const uint8_t *data, int size)
{
int samples = s->packet.frame_duration;
int redundancy = 0;
int redundancy_size, redundancy_pos;
int ret, i, consumed;
int delayed_samples = s->delayed_samples;
ret = ff_opus_rc_dec_init(&s->rc, data, size);
if (ret < 0)
return ret;
/* decode the silk frame */
if (s->packet.mode == OPUS_MODE_SILK || s->packet.mode == OPUS_MODE_HYBRID) {
#if CONFIG_SWRESAMPLE
if (!swr_is_initialized(s->swr)) {
#elif CONFIG_AVRESAMPLE
if (!avresample_is_open(s->avr)) {
#endif
ret = opus_init_resample(s);
if (ret < 0)
return ret;
}
samples = ff_silk_decode_superframe(s->silk, &s->rc, s->silk_output,
FFMIN(s->packet.bandwidth, OPUS_BANDWIDTH_WIDEBAND),
s->packet.stereo + 1,
silk_frame_duration_ms[s->packet.config]);
if (samples < 0) {
av_log(s->avctx, AV_LOG_ERROR, "Error decoding a SILK frame.\n");
return samples;
}
#if CONFIG_SWRESAMPLE
samples = swr_convert(s->swr,
(uint8_t**)s->out, s->packet.frame_duration,
(const uint8_t**)s->silk_output, samples);
#elif CONFIG_AVRESAMPLE
samples = avresample_convert(s->avr, (uint8_t**)s->out, s->out_size,
s->packet.frame_duration,
(uint8_t**)s->silk_output,
sizeof(s->silk_buf[0]),
samples);
#endif
if (samples < 0) {
av_log(s->avctx, AV_LOG_ERROR, "Error resampling SILK data.\n");
return samples;
}
av_assert2((samples & 7) == 0);
s->delayed_samples += s->packet.frame_duration - samples;
} else
ff_silk_flush(s->silk);
// decode redundancy information
consumed = opus_rc_tell(&s->rc);
if (s->packet.mode == OPUS_MODE_HYBRID && consumed + 37 <= size * 8)
redundancy = ff_opus_rc_dec_log(&s->rc, 12);
else if (s->packet.mode == OPUS_MODE_SILK && consumed + 17 <= size * 8)
redundancy = 1;
if (redundancy) {
redundancy_pos = ff_opus_rc_dec_log(&s->rc, 1);
if (s->packet.mode == OPUS_MODE_HYBRID)
redundancy_size = ff_opus_rc_dec_uint(&s->rc, 256) + 2;
else
redundancy_size = size - (consumed + 7) / 8;
size -= redundancy_size;
if (size < 0) {
av_log(s->avctx, AV_LOG_ERROR, "Invalid redundancy frame size.\n");
return AVERROR_INVALIDDATA;
}
if (redundancy_pos) {
ret = opus_decode_redundancy(s, data + size, redundancy_size);
if (ret < 0)
return ret;
ff_celt_flush(s->celt);
}
}
/* decode the CELT frame */
if (s->packet.mode == OPUS_MODE_CELT || s->packet.mode == OPUS_MODE_HYBRID) {
float *out_tmp[2] = { s->out[0], s->out[1] };
float **dst = (s->packet.mode == OPUS_MODE_CELT) ?
out_tmp : s->celt_output;
int celt_output_samples = samples;
int delay_samples = av_audio_fifo_size(s->celt_delay);
if (delay_samples) {
if (s->packet.mode == OPUS_MODE_HYBRID) {
av_audio_fifo_read(s->celt_delay, (void**)s->celt_output, delay_samples);
for (i = 0; i < s->output_channels; i++) {
s->fdsp->vector_fmac_scalar(out_tmp[i], s->celt_output[i], 1.0,
delay_samples);
out_tmp[i] += delay_samples;
}
celt_output_samples -= delay_samples;
} else {
av_log(s->avctx, AV_LOG_WARNING,
"Spurious CELT delay samples present.\n");
av_audio_fifo_drain(s->celt_delay, delay_samples);
if (s->avctx->err_recognition & AV_EF_EXPLODE)
return AVERROR_BUG;
}
}
ff_opus_rc_dec_raw_init(&s->rc, data + size, size);
ret = ff_celt_decode_frame(s->celt, &s->rc, dst,
s->packet.stereo + 1,
s->packet.frame_duration,
(s->packet.mode == OPUS_MODE_HYBRID) ? 17 : 0,
ff_celt_band_end[s->packet.bandwidth]);
if (ret < 0)
return ret;
if (s->packet.mode == OPUS_MODE_HYBRID) {
int celt_delay = s->packet.frame_duration - celt_output_samples;
void *delaybuf[2] = { s->celt_output[0] + celt_output_samples,
s->celt_output[1] + celt_output_samples };
for (i = 0; i < s->output_channels; i++) {
s->fdsp->vector_fmac_scalar(out_tmp[i],
s->celt_output[i], 1.0,
celt_output_samples);
}
ret = av_audio_fifo_write(s->celt_delay, delaybuf, celt_delay);
if (ret < 0)
return ret;
}
} else
ff_celt_flush(s->celt);
if (s->redundancy_idx) {
for (i = 0; i < s->output_channels; i++)
opus_fade(s->out[i], s->out[i],
s->redundancy_output[i] + 120 + s->redundancy_idx,
ff_celt_window2 + s->redundancy_idx, 120 - s->redundancy_idx);
s->redundancy_idx = 0;
}
if (redundancy) {
if (!redundancy_pos) {
ff_celt_flush(s->celt);
ret = opus_decode_redundancy(s, data + size, redundancy_size);
if (ret < 0)
return ret;
for (i = 0; i < s->output_channels; i++) {
opus_fade(s->out[i] + samples - 120 + delayed_samples,
s->out[i] + samples - 120 + delayed_samples,
s->redundancy_output[i] + 120,
ff_celt_window2, 120 - delayed_samples);
if (delayed_samples)
s->redundancy_idx = 120 - delayed_samples;
}
} else {
for (i = 0; i < s->output_channels; i++) {
memcpy(s->out[i] + delayed_samples, s->redundancy_output[i], 120 * sizeof(float));
opus_fade(s->out[i] + 120 + delayed_samples,
s->redundancy_output[i] + 120,
s->out[i] + 120 + delayed_samples,
ff_celt_window2, 120);
}
}
}
return samples;
}
static int opus_decode_subpacket(OpusStreamContext *s,
const uint8_t *buf, int buf_size,
float **out, int out_size,
int nb_samples)
{
int output_samples = 0;
int flush_needed = 0;
int i, j, ret;
s->out[0] = out[0];
s->out[1] = out[1];
s->out_size = out_size;
/* check if we need to flush the resampler */
#if CONFIG_SWRESAMPLE
if (swr_is_initialized(s->swr)) {
if (buf) {
int64_t cur_samplerate;
av_opt_get_int(s->swr, "in_sample_rate", 0, &cur_samplerate);
flush_needed = (s->packet.mode == OPUS_MODE_CELT) || (cur_samplerate != s->silk_samplerate);
} else {
flush_needed = !!s->delayed_samples;
}
}
#elif CONFIG_AVRESAMPLE
if (avresample_is_open(s->avr)) {
if (buf) {
int64_t cur_samplerate;
av_opt_get_int(s->avr, "in_sample_rate", 0, &cur_samplerate);
flush_needed = (s->packet.mode == OPUS_MODE_CELT) || (cur_samplerate != s->silk_samplerate);
} else {
flush_needed = !!s->delayed_samples;
}
}
#endif
if (!buf && !flush_needed)
return 0;
/* use dummy output buffers if the channel is not mapped to anything */
if (!s->out[0] ||
(s->output_channels == 2 && !s->out[1])) {
av_fast_malloc(&s->out_dummy, &s->out_dummy_allocated_size, s->out_size);
if (!s->out_dummy)
return AVERROR(ENOMEM);
if (!s->out[0])
s->out[0] = s->out_dummy;
if (!s->out[1])
s->out[1] = s->out_dummy;
}
/* flush the resampler if necessary */
if (flush_needed) {
ret = opus_flush_resample(s, s->delayed_samples);
if (ret < 0) {
av_log(s->avctx, AV_LOG_ERROR, "Error flushing the resampler.\n");
return ret;
}
#if CONFIG_SWRESAMPLE
swr_close(s->swr);
#elif CONFIG_AVRESAMPLE
avresample_close(s->avr);
#endif
output_samples += s->delayed_samples;
s->delayed_samples = 0;
if (!buf)
goto finish;
}
/* decode all the frames in the packet */
for (i = 0; i < s->packet.frame_count; i++) {
int size = s->packet.frame_size[i];
int samples = opus_decode_frame(s, buf + s->packet.frame_offset[i], size);
if (samples < 0) {
av_log(s->avctx, AV_LOG_ERROR, "Error decoding an Opus frame.\n");
if (s->avctx->err_recognition & AV_EF_EXPLODE)
return samples;
for (j = 0; j < s->output_channels; j++)
memset(s->out[j], 0, s->packet.frame_duration * sizeof(float));
samples = s->packet.frame_duration;
}
output_samples += samples;
for (j = 0; j < s->output_channels; j++)
s->out[j] += samples;
s->out_size -= samples * sizeof(float);
}
finish:
s->out[0] = s->out[1] = NULL;
s->out_size = 0;
return output_samples;
}