X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=avconv.c;h=719df6ddc4cfd061418c409aac2687ba7adfd4ce;hb=4ebd422c04849a21a6bb2128a715979e8067e6a6;hp=d0895129ac3e273a4fe2333ec3846f079cec2b23;hpb=64dca32cdf925b7f6b1308479924b53a1681597b;p=ffmpeg diff --git a/avconv.c b/avconv.c index d0895129ac3..719df6ddc4c 100644 --- a/avconv.c +++ b/avconv.c @@ -47,11 +47,10 @@ #include "libavutil/imgutils.h" #include "libavformat/os_support.h" -#if CONFIG_AVFILTER # include "libavfilter/avfilter.h" # include "libavfilter/avfiltergraph.h" +# include "libavfilter/buffersrc.h" # include "libavfilter/vsrc_buffer.h" -#endif #if HAVE_SYS_RESOURCE_H #include @@ -75,6 +74,11 @@ #include "libavutil/avassert.h" +#define VSYNC_AUTO -1 +#define VSYNC_PASSTHROUGH 0 +#define VSYNC_CFR 1 +#define VSYNC_VFR 2 + const char program_name[] = "avconv"; const int program_birth_year = 2000; @@ -110,10 +114,10 @@ static int do_hex_dump = 0; static int do_pkt_dump = 0; static int do_pass = 0; static char *pass_logfilename_prefix = NULL; -static int video_sync_method= -1; -static int audio_sync_method= 0; -static float audio_drift_threshold= 0.1; -static int copy_ts= 0; +static int video_sync_method = VSYNC_AUTO; +static int audio_sync_method = 0; +static float audio_drift_threshold = 0.1; +static int copy_ts = 0; static int copy_tb = 1; static int opt_shortest = 0; static char *vstats_filename; @@ -135,8 +139,9 @@ static float dts_delta_threshold = 10; static int print_stats = 1; static uint8_t *audio_buf; -static uint8_t *audio_out; -static unsigned int allocated_audio_out_size, allocated_audio_buf_size; +static unsigned int allocated_audio_buf_size; +static uint8_t *async_buf; +static unsigned int allocated_async_buf_size; #define DEFAULT_PASS_LOGFILENAME_PREFIX "av2pass" @@ -163,9 +168,11 @@ typedef struct InputStream { AVFrame *filtered_frame; int64_t start; /* time when read started */ - int64_t next_pts; /* synthetic pts for cases where pkt.pts - is not defined */ - int64_t pts; /* current pts */ + /* predicted dts of the next packet read for this stream or (when there are + * several frames in a packet) of the next frame in current packet */ + int64_t next_dts; + /* dts of the last packet read for this stream */ + int64_t last_dts; PtsCorrectionContext pts_ctx; double ts_scale; int is_start; /* is 1 at the start and after a discontinuity */ @@ -196,17 +203,19 @@ typedef struct OutputStream { int frame_number; /* input pts and corresponding output pts for A/V sync */ - //double sync_ipts; /* dts from the AVPacket of the demuxer in second units */ + // double sync_ipts; /* dts from the AVPacket of the demuxer in second units */ struct InputStream *sync_ist; /* input stream to sync against */ - int64_t sync_opts; /* output frame counter, could be changed to some true timestamp */ //FIXME look at frame_number + int64_t sync_opts; /* output frame counter, could be changed to some true timestamp */ // FIXME look at frame_number + /* pts of the first frame encoded for this stream, used for limiting + * recording time */ + int64_t first_pts; AVBitStreamFilterContext *bitstream_filters; AVCodec *enc; int64_t max_frames; + AVFrame *output_frame; /* video only */ int video_resample; - AVFrame pict_tmp; /* temporary image for resampling */ - struct SwsContext *img_resample_ctx; /* for image resampling */ int resample_height; int resample_width; int resample_pix_fmt; @@ -232,13 +241,10 @@ typedef struct OutputStream { AVFifoBuffer *fifo; /* for compression: one audio fifo per codec */ FILE *logfile; -#if CONFIG_AVFILTER AVFilterContext *output_video_filter; AVFilterContext *input_video_filter; - AVFilterBufferRef *picref; char *avfilter; AVFilterGraph *graph; -#endif int64_t sws_flags; AVDictionary *opts; @@ -258,9 +264,9 @@ typedef struct OutputFile { uint64_t limit_filesize; } OutputFile; -static InputStream *input_streams = NULL; +static InputStream *input_streams = NULL; static int nb_input_streams = 0; -static InputFile *input_files = NULL; +static InputFile *input_files = NULL; static int nb_input_files = 0; static OutputStream *output_streams = NULL; @@ -355,10 +361,8 @@ typedef struct OptionsContext { int nb_presets; SpecifierOpt *copy_initial_nonkeyframes; int nb_copy_initial_nonkeyframes; -#if CONFIG_AVFILTER SpecifierOpt *filters; int nb_filters; -#endif } OptionsContext; #define MATCH_PER_STREAM_OPT(name, type, outvar, fmtctx, st)\ @@ -411,11 +415,10 @@ static void reset_options(OptionsContext *o) init_opts(); } -static int alloc_buffer(InputStream *ist, FrameBuffer **pbuf) +static int alloc_buffer(InputStream *ist, AVCodecContext *s, FrameBuffer **pbuf) { - AVCodecContext *s = ist->st->codec; FrameBuffer *buf = av_mallocz(sizeof(*buf)); - int ret; + int i, ret; const int pixel_size = av_pix_fmt_descriptors[s->pix_fmt].comp[0].step_minus1+1; int h_chroma_shift, v_chroma_shift; int edge = 32; // XXX should be avcodec_get_edge_width(), but that fails on svq1 @@ -438,12 +441,12 @@ static int alloc_buffer(InputStream *ist, FrameBuffer **pbuf) /* XXX this shouldn't be needed, but some tests break without this line * those decoders are buggy and need to be fixed. * the following tests fail: - * bethsoft-vid, cdgraphics, ansi, aasc, fraps-v1, qtrle-1bit + * cdgraphics, ansi, aasc, fraps-v1, qtrle-1bit */ memset(buf->base[0], 128, ret); avcodec_get_chroma_sub_sample(s->pix_fmt, &h_chroma_shift, &v_chroma_shift); - for (int i = 0; i < FF_ARRAY_ELEMS(buf->data); i++) { + for (i = 0; i < FF_ARRAY_ELEMS(buf->data); i++) { const int h_shift = i==0 ? 0 : h_chroma_shift; const int v_shift = i==0 ? 0 : v_chroma_shift; if (s->flags & CODEC_FLAG_EMU_EDGE) @@ -489,7 +492,7 @@ static int codec_get_buffer(AVCodecContext *s, AVFrame *frame) FrameBuffer *buf; int ret, i; - if (!ist->buffer_pool && (ret = alloc_buffer(ist, &ist->buffer_pool)) < 0) + if (!ist->buffer_pool && (ret = alloc_buffer(ist, s, &ist->buffer_pool)) < 0) return ret; buf = ist->buffer_pool; @@ -498,7 +501,7 @@ static int codec_get_buffer(AVCodecContext *s, AVFrame *frame) if (buf->w != s->width || buf->h != s->height || buf->pix_fmt != s->pix_fmt) { av_freep(&buf->base[0]); av_free(buf); - if ((ret = alloc_buffer(ist, &buf)) < 0) + if ((ret = alloc_buffer(ist, s, &buf)) < 0) return ret; } buf->refcount++; @@ -507,6 +510,10 @@ static int codec_get_buffer(AVCodecContext *s, AVFrame *frame) frame->type = FF_BUFFER_TYPE_USER; frame->extended_data = frame->data; frame->pkt_pts = s->pkt ? s->pkt->pts : AV_NOPTS_VALUE; + frame->width = buf->w; + frame->height = buf->h; + frame->format = buf->pix_fmt; + frame->sample_aspect_ratio = s->sample_aspect_ratio; for (i = 0; i < FF_ARRAY_ELEMS(buf->data); i++) { frame->base[i] = buf->base[i]; // XXX h264.c uses base though it shouldn't @@ -529,7 +536,12 @@ static void codec_release_buffer(AVCodecContext *s, AVFrame *frame) unref_buffer(ist, buf); } -#if CONFIG_AVFILTER +static void filter_release_buffer(AVFilterBuffer *fb) +{ + FrameBuffer *buf = fb->priv; + av_free(fb); + unref_buffer(buf->ist, buf); +} static int configure_video_filters(InputStream *ist, OutputStream *ost) { @@ -537,16 +549,16 @@ static int configure_video_filters(InputStream *ist, OutputStream *ost) /** filter graph containing all filters including input & output */ AVCodecContext *codec = ost->st->codec; AVCodecContext *icodec = ist->st->codec; - FFSinkContext ffsink_ctx = { .pix_fmt = codec->pix_fmt }; + SinkContext sink_ctx = { .pix_fmt = codec->pix_fmt }; AVRational sample_aspect_ratio; char args[255]; int ret; ost->graph = avfilter_graph_alloc(); - if (ist->st->sample_aspect_ratio.num){ + if (ist->st->sample_aspect_ratio.num) { sample_aspect_ratio = ist->st->sample_aspect_ratio; - }else + } else sample_aspect_ratio = ist->st->codec->sample_aspect_ratio; snprintf(args, 255, "%d:%d:%d:%d:%d:%d:%d", ist->st->codec->width, @@ -557,13 +569,13 @@ static int configure_video_filters(InputStream *ist, OutputStream *ost) "src", args, NULL, ost->graph); if (ret < 0) return ret; - ret = avfilter_graph_create_filter(&ost->output_video_filter, &ffsink, - "out", NULL, &ffsink_ctx, ost->graph); + ret = avfilter_graph_create_filter(&ost->output_video_filter, &sink, + "out", NULL, &sink_ctx, ost->graph); if (ret < 0) return ret; last_filter = ost->input_video_filter; - if (codec->width != icodec->width || codec->height != icodec->height) { + if (codec->width != icodec->width || codec->height != icodec->height) { snprintf(args, 255, "%d:%d:flags=0x%X", codec->width, codec->height, @@ -580,8 +592,8 @@ static int configure_video_filters(InputStream *ist, OutputStream *ost) ost->graph->scale_sws_opts = av_strdup(args); if (ost->avfilter) { - AVFilterInOut *outputs = av_malloc(sizeof(AVFilterInOut)); - AVFilterInOut *inputs = av_malloc(sizeof(AVFilterInOut)); + AVFilterInOut *outputs = avfilter_inout_alloc(); + AVFilterInOut *inputs = avfilter_inout_alloc(); outputs->name = av_strdup("in"); outputs->filter_ctx = last_filter; @@ -595,7 +607,6 @@ static int configure_video_filters(InputStream *ist, OutputStream *ost) if ((ret = avfilter_graph_parse(ost->graph, ost->avfilter, inputs, outputs, NULL)) < 0) return ret; - av_freep(&ost->avfilter); } else { if ((ret = avfilter_link(last_filter, 0, ost->output_video_filter, 0)) < 0) return ret; @@ -608,12 +619,11 @@ static int configure_video_filters(InputStream *ist, OutputStream *ost) codec->height = ost->output_video_filter->inputs[0]->h; codec->sample_aspect_ratio = ost->st->sample_aspect_ratio = ost->frame_aspect_ratio ? // overridden by the -aspect cli option - av_d2q(ost->frame_aspect_ratio*codec->height/codec->width, 255) : + av_d2q(ost->frame_aspect_ratio * codec->height/codec->width, 255) : ost->output_video_filter->inputs[0]->sample_aspect_ratio; return 0; } -#endif /* CONFIG_AVFILTER */ static void term_exit(void) { @@ -633,7 +643,7 @@ sigterm_handler(int sig) static void term_init(void) { - signal(SIGINT , sigterm_handler); /* Interrupt (ANSI). */ + signal(SIGINT , sigterm_handler); /* Interrupt (ANSI). */ signal(SIGTERM, sigterm_handler); /* Termination (ANSI). */ #ifdef SIGXCPU signal(SIGXCPU, sigterm_handler); @@ -652,14 +662,32 @@ void exit_program(int ret) int i; /* close files */ - for(i=0;ioformat->flags & AVFMT_NOFILE) && s->pb) avio_close(s->pb); avformat_free_context(s); av_dict_free(&output_files[i].opts); } - for(i=0;inext; + av_bitstream_filter_close(bsfc); + bsfc = next; + } + output_streams[i].bitstream_filters = NULL; + + if (output_streams[i].output_frame) { + AVFrame *frame = output_streams[i].output_frame; + if (frame->extended_data != frame->data) + av_freep(&frame->extended_data); + av_freep(&frame); + } + + av_freep(&output_streams[i].avfilter); + } + for (i = 0; i < nb_input_files; i++) { avformat_close_input(&input_files[i].ctx); } for (i = 0; i < nb_input_streams; i++) { @@ -680,12 +708,11 @@ void exit_program(int ret) uninit_opts(); av_free(audio_buf); - av_free(audio_out); - allocated_audio_buf_size= allocated_audio_out_size= 0; + allocated_audio_buf_size = 0; + av_free(async_buf); + allocated_async_buf_size = 0; -#if CONFIG_AVFILTER avfilter_uninit(); -#endif avformat_network_deinit(); if (received_sigterm) { @@ -725,10 +752,10 @@ static void assert_codec_experimental(AVCodecContext *c, int encoder) static void choose_sample_fmt(AVStream *st, AVCodec *codec) { - if(codec && codec->sample_fmts){ - const enum AVSampleFormat *p= codec->sample_fmts; - for(; *p!=-1; p++){ - if(*p == st->codec->sample_fmt) + if (codec && codec->sample_fmts) { + const enum AVSampleFormat *p = codec->sample_fmts; + for (; *p != -1; p++) { + if (*p == st->codec->sample_fmt) break; } if (*p == -1) { @@ -784,72 +811,90 @@ static void update_sample_fmt(AVCodecContext *dec, AVCodec *dec_codec, static void choose_sample_rate(AVStream *st, AVCodec *codec) { - if(codec && codec->supported_samplerates){ - const int *p= codec->supported_samplerates; - int best=0; - int best_dist=INT_MAX; - for(; *p; p++){ - int dist= abs(st->codec->sample_rate - *p); - if(dist < best_dist){ - best_dist= dist; - best= *p; + if (codec && codec->supported_samplerates) { + const int *p = codec->supported_samplerates; + int best = 0; + int best_dist = INT_MAX; + for (; *p; p++) { + int dist = abs(st->codec->sample_rate - *p); + if (dist < best_dist) { + best_dist = dist; + best = *p; } } - if(best_dist){ + if (best_dist) { av_log(st->codec, AV_LOG_WARNING, "Requested sampling rate unsupported using closest supported (%d)\n", best); } - st->codec->sample_rate= best; + st->codec->sample_rate = best; } } static void choose_pixel_fmt(AVStream *st, AVCodec *codec) { - if(codec && codec->pix_fmts){ - const enum PixelFormat *p= codec->pix_fmts; - if(st->codec->strict_std_compliance <= FF_COMPLIANCE_UNOFFICIAL){ - if(st->codec->codec_id==CODEC_ID_MJPEG){ - p= (const enum PixelFormat[]){PIX_FMT_YUVJ420P, PIX_FMT_YUVJ422P, PIX_FMT_YUV420P, PIX_FMT_YUV422P, PIX_FMT_NONE}; - }else if(st->codec->codec_id==CODEC_ID_LJPEG){ - p= (const enum PixelFormat[]){PIX_FMT_YUVJ420P, PIX_FMT_YUVJ422P, PIX_FMT_YUVJ444P, PIX_FMT_YUV420P, PIX_FMT_YUV422P, PIX_FMT_YUV444P, PIX_FMT_BGRA, PIX_FMT_NONE}; + if (codec && codec->pix_fmts) { + const enum PixelFormat *p = codec->pix_fmts; + if (st->codec->strict_std_compliance <= FF_COMPLIANCE_UNOFFICIAL) { + if (st->codec->codec_id == CODEC_ID_MJPEG) { + p = (const enum PixelFormat[]) { PIX_FMT_YUVJ420P, PIX_FMT_YUVJ422P, PIX_FMT_YUV420P, PIX_FMT_YUV422P, PIX_FMT_NONE }; + } else if (st->codec->codec_id == CODEC_ID_LJPEG) { + p = (const enum PixelFormat[]) { PIX_FMT_YUVJ420P, PIX_FMT_YUVJ422P, PIX_FMT_YUVJ444P, PIX_FMT_YUV420P, + PIX_FMT_YUV422P, PIX_FMT_YUV444P, PIX_FMT_BGRA, PIX_FMT_NONE }; } } for (; *p != PIX_FMT_NONE; p++) { - if(*p == st->codec->pix_fmt) + if (*p == st->codec->pix_fmt) break; } if (*p == PIX_FMT_NONE) { - if(st->codec->pix_fmt != PIX_FMT_NONE) + if (st->codec->pix_fmt != PIX_FMT_NONE) av_log(NULL, AV_LOG_WARNING, - "Incompatible pixel format '%s' for codec '%s', auto-selecting format '%s'\n", - av_pix_fmt_descriptors[st->codec->pix_fmt].name, - codec->name, - av_pix_fmt_descriptors[codec->pix_fmts[0]].name); + "Incompatible pixel format '%s' for codec '%s', auto-selecting format '%s'\n", + av_pix_fmt_descriptors[st->codec->pix_fmt].name, + codec->name, + av_pix_fmt_descriptors[codec->pix_fmts[0]].name); st->codec->pix_fmt = codec->pix_fmts[0]; } } } static double -get_sync_ipts(const OutputStream *ost) +get_sync_ipts(const OutputStream *ost, int64_t pts) { - const InputStream *ist = ost->sync_ist; OutputFile *of = &output_files[ost->file_index]; - return (double)(ist->pts - of->start_time)/AV_TIME_BASE; + return (double)(pts - of->start_time) / AV_TIME_BASE; } -static void write_frame(AVFormatContext *s, AVPacket *pkt, AVCodecContext *avctx, AVBitStreamFilterContext *bsfc){ +static void write_frame(AVFormatContext *s, AVPacket *pkt, OutputStream *ost) +{ + AVBitStreamFilterContext *bsfc = ost->bitstream_filters; + AVCodecContext *avctx = ost->st->codec; int ret; - while(bsfc){ - AVPacket new_pkt= *pkt; - int a= av_bitstream_filter_filter(bsfc, avctx, NULL, - &new_pkt.data, &new_pkt.size, - pkt->data, pkt->size, - pkt->flags & AV_PKT_FLAG_KEY); - if(a>0){ + /* + * Audio encoders may split the packets -- #frames in != #packets out. + * But there is no reordering, so we can limit the number of output packets + * by simply dropping them here. + * Counting encoded video frames needs to be done separately because of + * reordering, see do_video_out() + */ + if (!(avctx->codec_type == AVMEDIA_TYPE_VIDEO && avctx->codec)) { + if (ost->frame_number >= ost->max_frames) { av_free_packet(pkt); - new_pkt.destruct= av_destruct_packet; - } else if(a<0){ + return; + } + ost->frame_number++; + } + + while (bsfc) { + AVPacket new_pkt = *pkt; + int a = av_bitstream_filter_filter(bsfc, avctx, NULL, + &new_pkt.data, &new_pkt.size, + pkt->data, pkt->size, + pkt->flags & AV_PKT_FLAG_KEY); + if (a > 0) { + av_free_packet(pkt); + new_pkt.destruct = av_destruct_packet; + } else if (a < 0) { av_log(NULL, AV_LOG_ERROR, "%s failed for stream %d, codec %s", bsfc->filter->name, pkt->stream_index, avctx->codec ? avctx->codec->name : "copy"); @@ -857,18 +902,32 @@ static void write_frame(AVFormatContext *s, AVPacket *pkt, AVCodecContext *avctx if (exit_on_error) exit_program(1); } - *pkt= new_pkt; + *pkt = new_pkt; - bsfc= bsfc->next; + bsfc = bsfc->next; } - ret= av_interleaved_write_frame(s, pkt); - if(ret < 0){ + pkt->stream_index = ost->index; + ret = av_interleaved_write_frame(s, pkt); + if (ret < 0) { print_error("av_interleaved_write_frame()", ret); exit_program(1); } } +static int check_recording_time(OutputStream *ost) +{ + OutputFile *of = &output_files[ost->file_index]; + + if (of->recording_time != INT64_MAX && + av_compare_ts(ost->sync_opts - ost->first_pts, ost->st->codec->time_base, of->recording_time, + AV_TIME_BASE_Q) >= 0) { + ost->is_past_recording_time = 1; + return 0; + } + return 1; +} + static void generate_silence(uint8_t* buf, enum AVSampleFormat sample_fmt, size_t size) { int fill_char = 0x00; @@ -877,43 +936,110 @@ static void generate_silence(uint8_t* buf, enum AVSampleFormat sample_fmt, size_ memset(buf, fill_char, size); } +static int encode_audio_frame(AVFormatContext *s, OutputStream *ost, + const uint8_t *buf, int buf_size) +{ + AVCodecContext *enc = ost->st->codec; + AVFrame *frame = NULL; + AVPacket pkt; + int ret, got_packet; + + av_init_packet(&pkt); + pkt.data = NULL; + pkt.size = 0; + + if (buf) { + if (!ost->output_frame) { + ost->output_frame = avcodec_alloc_frame(); + if (!ost->output_frame) { + av_log(NULL, AV_LOG_FATAL, "out-of-memory in encode_audio_frame()\n"); + exit_program(1); + } + } + frame = ost->output_frame; + if (frame->extended_data != frame->data) + av_freep(&frame->extended_data); + avcodec_get_frame_defaults(frame); + + frame->nb_samples = buf_size / + (enc->channels * av_get_bytes_per_sample(enc->sample_fmt)); + if ((ret = avcodec_fill_audio_frame(frame, enc->channels, enc->sample_fmt, + buf, buf_size, 1)) < 0) { + av_log(NULL, AV_LOG_FATAL, "Audio encoding failed\n"); + exit_program(1); + } + + if (!check_recording_time(ost)) + return 0; + + frame->pts = ost->sync_opts; + ost->sync_opts += frame->nb_samples; + } + + got_packet = 0; + if (avcodec_encode_audio2(enc, &pkt, frame, &got_packet) < 0) { + av_log(NULL, AV_LOG_FATAL, "Audio encoding failed\n"); + exit_program(1); + } + + if (got_packet) { + if (pkt.pts != AV_NOPTS_VALUE) + pkt.pts = av_rescale_q(pkt.pts, enc->time_base, ost->st->time_base); + if (pkt.dts != AV_NOPTS_VALUE) + pkt.dts = av_rescale_q(pkt.dts, enc->time_base, ost->st->time_base); + if (pkt.duration > 0) + pkt.duration = av_rescale_q(pkt.duration, enc->time_base, ost->st->time_base); + + write_frame(s, &pkt, ost); + + audio_size += pkt.size; + } + + return pkt.size; +} + +static int alloc_audio_output_buf(AVCodecContext *dec, AVCodecContext *enc, + int nb_samples) +{ + int64_t audio_buf_samples; + int audio_buf_size; + + /* calculate required number of samples to allocate */ + audio_buf_samples = ((int64_t)nb_samples * enc->sample_rate + dec->sample_rate) / + dec->sample_rate; + audio_buf_samples = 4 * audio_buf_samples + 16; // safety factors for resampling + audio_buf_samples = FFMAX(audio_buf_samples, enc->frame_size); + if (audio_buf_samples > INT_MAX) + return AVERROR(EINVAL); + + audio_buf_size = av_samples_get_buffer_size(NULL, enc->channels, + audio_buf_samples, + enc->sample_fmt, 0); + if (audio_buf_size < 0) + return audio_buf_size; + + av_fast_malloc(&audio_buf, &allocated_audio_buf_size, audio_buf_size); + if (!audio_buf) + return AVERROR(ENOMEM); + + return 0; +} + static void do_audio_out(AVFormatContext *s, OutputStream *ost, InputStream *ist, AVFrame *decoded_frame) { uint8_t *buftmp; - int64_t audio_out_size, audio_buf_size; - int size_out, frame_bytes, ret, resample_changed; - AVCodecContext *enc= ost->st->codec; - AVCodecContext *dec= ist->st->codec; + int size_out, frame_bytes, resample_changed; + AVCodecContext *enc = ost->st->codec; + AVCodecContext *dec = ist->st->codec; int osize = av_get_bytes_per_sample(enc->sample_fmt); int isize = av_get_bytes_per_sample(dec->sample_fmt); - const int coded_bps = av_get_bits_per_sample(enc->codec->id); uint8_t *buf = decoded_frame->data[0]; int size = decoded_frame->nb_samples * dec->channels * isize; - int64_t allocated_for_size = size; - -need_realloc: - audio_buf_size= (allocated_for_size + isize*dec->channels - 1) / (isize*dec->channels); - audio_buf_size= (audio_buf_size*enc->sample_rate + dec->sample_rate) / dec->sample_rate; - audio_buf_size= audio_buf_size*2 + 10000; //safety factors for the deprecated resampling API - audio_buf_size= FFMAX(audio_buf_size, enc->frame_size); - audio_buf_size*= osize*enc->channels; - - audio_out_size= FFMAX(audio_buf_size, enc->frame_size * osize * enc->channels); - if(coded_bps > 8*osize) - audio_out_size= audio_out_size * coded_bps / (8*osize); - audio_out_size += FF_MIN_BUFFER_SIZE; - - if(audio_out_size > INT_MAX || audio_buf_size > INT_MAX){ - av_log(NULL, AV_LOG_FATAL, "Buffer sizes too large\n"); - exit_program(1); - } - av_fast_malloc(&audio_buf, &allocated_audio_buf_size, audio_buf_size); - av_fast_malloc(&audio_out, &allocated_audio_out_size, audio_out_size); - if (!audio_buf || !audio_out){ - av_log(NULL, AV_LOG_FATAL, "Out of memory in do_audio_out\n"); + if (alloc_audio_output_buf(dec, enc, decoded_frame->nb_samples) < 0) { + av_log(NULL, AV_LOG_FATAL, "Error allocating audio buffer\n"); exit_program(1); } @@ -960,8 +1086,8 @@ need_realloc: } #define MAKE_SFMT_PAIR(a,b) ((a)+AV_SAMPLE_FMT_NB*(b)) - if (!ost->audio_resample && dec->sample_fmt!=enc->sample_fmt && - MAKE_SFMT_PAIR(enc->sample_fmt,dec->sample_fmt)!=ost->reformat_pair) { + if (!ost->audio_resample && dec->sample_fmt != enc->sample_fmt && + MAKE_SFMT_PAIR(enc->sample_fmt,dec->sample_fmt) != ost->reformat_pair) { if (ost->reformat_ctx) av_audio_convert_free(ost->reformat_ctx); ost->reformat_ctx = av_audio_convert_alloc(enc->sample_fmt, 1, @@ -972,45 +1098,49 @@ need_realloc: av_get_sample_fmt_name(enc->sample_fmt)); exit_program(1); } - ost->reformat_pair=MAKE_SFMT_PAIR(enc->sample_fmt,dec->sample_fmt); + ost->reformat_pair = MAKE_SFMT_PAIR(enc->sample_fmt,dec->sample_fmt); } - if(audio_sync_method){ - double delta = get_sync_ipts(ost) * enc->sample_rate - ost->sync_opts - - av_fifo_size(ost->fifo)/(enc->channels * osize); + if (audio_sync_method > 0) { + double delta = get_sync_ipts(ost, ist->last_dts) * enc->sample_rate - ost->sync_opts - + av_fifo_size(ost->fifo) / (enc->channels * osize); int idelta = delta * dec->sample_rate / enc->sample_rate; int byte_delta = idelta * isize * dec->channels; - //FIXME resample delay - if(fabs(delta) > 50){ - if(ist->is_start || fabs(delta) > audio_drift_threshold*enc->sample_rate){ - if(byte_delta < 0){ - byte_delta= FFMAX(byte_delta, -size); + // FIXME resample delay + if (fabs(delta) > 50) { + if (ist->is_start || fabs(delta) > audio_drift_threshold*enc->sample_rate) { + if (byte_delta < 0) { + byte_delta = FFMAX(byte_delta, -size); size += byte_delta; buf -= byte_delta; av_log(NULL, AV_LOG_VERBOSE, "discarding %d audio samples\n", -byte_delta / (isize * dec->channels)); - if(!size) + if (!size) return; - ist->is_start=0; - }else{ - static uint8_t *input_tmp= NULL; - input_tmp= av_realloc(input_tmp, byte_delta + size); - - if(byte_delta > allocated_for_size - size){ - allocated_for_size= byte_delta + (int64_t)size; - goto need_realloc; + ist->is_start = 0; + } else { + av_fast_malloc(&async_buf, &allocated_async_buf_size, + byte_delta + size); + if (!async_buf) { + av_log(NULL, AV_LOG_FATAL, "Out of memory in do_audio_out\n"); + exit_program(1); + } + + if (alloc_audio_output_buf(dec, enc, decoded_frame->nb_samples + idelta) < 0) { + av_log(NULL, AV_LOG_FATAL, "Error allocating audio buffer\n"); + exit_program(1); } - ist->is_start=0; + ist->is_start = 0; - generate_silence(input_tmp, dec->sample_fmt, byte_delta); - memcpy(input_tmp + byte_delta, buf, size); - buf= input_tmp; + generate_silence(async_buf, dec->sample_fmt, byte_delta); + memcpy(async_buf + byte_delta, buf, size); + buf = async_buf; size += byte_delta; av_log(NULL, AV_LOG_VERBOSE, "adding %d audio samples of silence\n", idelta); } - }else if(audio_sync_method>1){ - int comp= av_clip(delta, -audio_sync_method, audio_sync_method); + } else if (audio_sync_method > 1) { + int comp = av_clip(delta, -audio_sync_method, audio_sync_method); av_assert0(ost->audio_resample); av_log(NULL, AV_LOG_VERBOSE, "compensating audio timestamp drift:%f compensation:%d in:%d\n", delta, comp, enc->sample_rate); @@ -1018,9 +1148,9 @@ need_realloc: av_resample_compensate(*(struct AVResampleContext**)ost->resample, comp, enc->sample_rate); } } - }else - ost->sync_opts= lrintf(get_sync_ipts(ost) * enc->sample_rate) - - av_fifo_size(ost->fifo)/(enc->channels * osize); //FIXME wrong + } else if (audio_sync_method == 0) + ost->sync_opts = lrintf(get_sync_ipts(ost, ist->last_dts) * enc->sample_rate) - + av_fifo_size(ost->fifo) / (enc->channels * osize); // FIXME wrong if (ost->audio_resample) { buftmp = audio_buf; @@ -1033,24 +1163,24 @@ need_realloc: size_out = size; } - if (!ost->audio_resample && dec->sample_fmt!=enc->sample_fmt) { - const void *ibuf[6]= {buftmp}; - void *obuf[6]= {audio_buf}; - int istride[6]= {isize}; - int ostride[6]= {osize}; - int len= size_out/istride[0]; - if (av_audio_convert(ost->reformat_ctx, obuf, ostride, ibuf, istride, len)<0) { + if (!ost->audio_resample && dec->sample_fmt != enc->sample_fmt) { + const void *ibuf[6] = { buftmp }; + void *obuf[6] = { audio_buf }; + int istride[6] = { isize }; + int ostride[6] = { osize }; + int len = size_out / istride[0]; + if (av_audio_convert(ost->reformat_ctx, obuf, ostride, ibuf, istride, len) < 0) { printf("av_audio_convert() failed\n"); if (exit_on_error) exit_program(1); return; } buftmp = audio_buf; - size_out = len*osize; + size_out = len * osize; } /* now encode as many frames as possible */ - if (enc->frame_size > 1) { + if (!(enc->codec->capabilities & CODEC_CAP_VARIABLE_FRAME_SIZE)) { /* output resampled raw samples */ if (av_fifo_realloc2(ost->fifo, av_fifo_size(ost->fifo) + size_out) < 0) { av_log(NULL, AV_LOG_FATAL, "av_fifo_realloc2() failed\n"); @@ -1061,62 +1191,11 @@ need_realloc: frame_bytes = enc->frame_size * osize * enc->channels; while (av_fifo_size(ost->fifo) >= frame_bytes) { - AVPacket pkt; - av_init_packet(&pkt); - av_fifo_generic_read(ost->fifo, audio_buf, frame_bytes, NULL); - - //FIXME pass ost->sync_opts as AVFrame.pts in avcodec_encode_audio() - - ret = avcodec_encode_audio(enc, audio_out, audio_out_size, - (short *)audio_buf); - if (ret < 0) { - av_log(NULL, AV_LOG_FATAL, "Audio encoding failed\n"); - exit_program(1); - } - audio_size += ret; - pkt.stream_index= ost->index; - pkt.data= audio_out; - pkt.size= ret; - if(enc->coded_frame && enc->coded_frame->pts != AV_NOPTS_VALUE) - pkt.pts= av_rescale_q(enc->coded_frame->pts, enc->time_base, ost->st->time_base); - pkt.flags |= AV_PKT_FLAG_KEY; - write_frame(s, &pkt, enc, ost->bitstream_filters); - - ost->sync_opts += enc->frame_size; + encode_audio_frame(s, ost, audio_buf, frame_bytes); } } else { - AVPacket pkt; - av_init_packet(&pkt); - - ost->sync_opts += size_out / (osize * enc->channels); - - /* output a pcm frame */ - /* determine the size of the coded buffer */ - size_out /= osize; - if (coded_bps) - size_out = size_out*coded_bps/8; - - if(size_out > audio_out_size){ - av_log(NULL, AV_LOG_FATAL, "Internal error, buffer size too small\n"); - exit_program(1); - } - - //FIXME pass ost->sync_opts as AVFrame.pts in avcodec_encode_audio() - ret = avcodec_encode_audio(enc, audio_out, size_out, - (short *)buftmp); - if (ret < 0) { - av_log(NULL, AV_LOG_FATAL, "Audio encoding failed\n"); - exit_program(1); - } - audio_size += ret; - pkt.stream_index= ost->index; - pkt.data= audio_out; - pkt.size= ret; - if(enc->coded_frame && enc->coded_frame->pts != AV_NOPTS_VALUE) - pkt.pts= av_rescale_q(enc->coded_frame->pts, enc->time_base, ost->st->time_base); - pkt.flags |= AV_PKT_FLAG_KEY; - write_frame(s, &pkt, enc, ost->bitstream_filters); + encode_audio_frame(s, ost, buftmp, size_out); } } @@ -1135,14 +1214,14 @@ static void pre_process_video_frame(InputStream *ist, AVPicture *picture, void * /* create temporary picture */ size = avpicture_get_size(dec->pix_fmt, dec->width, dec->height); - buf = av_malloc(size); + buf = av_malloc(size); if (!buf) return; picture2 = &picture_tmp; avpicture_fill(picture2, buf, dec->pix_fmt, dec->width, dec->height); - if(avpicture_deinterlace(picture2, picture, + if (avpicture_deinterlace(picture2, picture, dec->pix_fmt, dec->width, dec->height) < 0) { /* if error, do not deinterlace */ av_log(NULL, AV_LOG_WARNING, "Deinterlacing failed\n"); @@ -1192,11 +1271,15 @@ static void do_subtitle_out(AVFormatContext *s, else nb = 1; - for(i = 0; i < nb; i++) { + for (i = 0; i < nb; i++) { + ost->sync_opts = av_rescale_q(pts, ist->st->time_base, enc->time_base); + if (!check_recording_time(ost)) + return; + sub->pts = av_rescale_q(pts, ist->st->time_base, AV_TIME_BASE_Q); // start_display_time is required to be 0 - sub->pts += av_rescale_q(sub->start_display_time, (AVRational){1, 1000}, AV_TIME_BASE_Q); - sub->end_display_time -= sub->start_display_time; + sub->pts += av_rescale_q(sub->start_display_time, (AVRational){ 1, 1000 }, AV_TIME_BASE_Q); + sub->end_display_time -= sub->start_display_time; sub->start_display_time = 0; subtitle_out_size = avcodec_encode_subtitle(enc, subtitle_out, subtitle_out_max_size, sub); @@ -1206,10 +1289,9 @@ static void do_subtitle_out(AVFormatContext *s, } av_init_packet(&pkt); - pkt.stream_index = ost->index; pkt.data = subtitle_out; pkt.size = subtitle_out_size; - pkt.pts = av_rescale_q(sub->pts, AV_TIME_BASE_Q, ost->st->time_base); + pkt.pts = av_rescale_q(sub->pts, AV_TIME_BASE_Q, ost->st->time_base); if (enc->codec_id == CODEC_ID_DVB_SUBTITLE) { /* XXX: the pts correction is handled here. Maybe handling it in the codec would be better */ @@ -1218,89 +1300,23 @@ static void do_subtitle_out(AVFormatContext *s, else pkt.pts += 90 * sub->end_display_time; } - write_frame(s, &pkt, ost->st->codec, ost->bitstream_filters); + write_frame(s, &pkt, ost); } } -static int bit_buffer_size= 1024*256; -static uint8_t *bit_buffer= NULL; - -static void do_video_resample(OutputStream *ost, - InputStream *ist, - AVFrame *in_picture, - AVFrame **out_picture) -{ - int resample_changed = 0; - AVCodecContext *dec = ist->st->codec; - *out_picture = in_picture; - - resample_changed = ost->resample_width != dec->width || - ost->resample_height != dec->height || - ost->resample_pix_fmt != dec->pix_fmt; - - if (resample_changed) { - av_log(NULL, AV_LOG_INFO, - "Input stream #%d:%d frame changed from size:%dx%d fmt:%s to size:%dx%d fmt:%s\n", - ist->file_index, ist->st->index, - ost->resample_width, ost->resample_height, av_get_pix_fmt_name(ost->resample_pix_fmt), - dec->width , dec->height , av_get_pix_fmt_name(dec->pix_fmt)); - if(!ost->video_resample) - ost->video_resample = 1; - } - -#if !CONFIG_AVFILTER - if (ost->video_resample) { - *out_picture = &ost->pict_tmp; - if (resample_changed) { - /* initialize a new scaler context */ - sws_freeContext(ost->img_resample_ctx); - ost->img_resample_ctx = sws_getContext( - ist->st->codec->width, - ist->st->codec->height, - ist->st->codec->pix_fmt, - ost->st->codec->width, - ost->st->codec->height, - ost->st->codec->pix_fmt, - ost->sws_flags, NULL, NULL, NULL); - if (ost->img_resample_ctx == NULL) { - av_log(NULL, AV_LOG_FATAL, "Cannot get resampling context\n"); - exit_program(1); - } - } - sws_scale(ost->img_resample_ctx, in_picture->data, in_picture->linesize, - 0, ost->resample_height, (*out_picture)->data, (*out_picture)->linesize); - } -#else - if (resample_changed) { - avfilter_graph_free(&ost->graph); - if (configure_video_filters(ist, ost)) { - av_log(NULL, AV_LOG_FATAL, "Error reinitializing filters!\n"); - exit_program(1); - } - } -#endif - if (resample_changed) { - ost->resample_width = dec->width; - ost->resample_height = dec->height; - ost->resample_pix_fmt = dec->pix_fmt; - } -} - - static void do_video_out(AVFormatContext *s, OutputStream *ost, - InputStream *ist, AVFrame *in_picture, int *frame_size, float quality) { int nb_frames, i, ret, format_video_sync; - AVFrame *final_picture; AVCodecContext *enc; - double sync_ipts; + double sync_ipts, delta; enc = ost->st->codec; - sync_ipts = get_sync_ipts(ost) / av_q2d(enc->time_base); + sync_ipts = get_sync_ipts(ost, in_picture->pts) / av_q2d(enc->time_base); + delta = sync_ipts - ost->sync_opts; /* by default, we output a single frame */ nb_frames = 1; @@ -1308,44 +1324,53 @@ static void do_video_out(AVFormatContext *s, *frame_size = 0; format_video_sync = video_sync_method; - if (format_video_sync < 0) - format_video_sync = (s->oformat->flags & AVFMT_NOTIMESTAMPS) ? 0 : - (s->oformat->flags & AVFMT_VARIABLE_FPS) ? 2 : 1; - - if (format_video_sync) { - double vdelta = sync_ipts - ost->sync_opts; - //FIXME set to 0.5 after we fix some dts/pts bugs like in avidec.c - if (vdelta < -1.1) + if (format_video_sync == VSYNC_AUTO) + format_video_sync = (s->oformat->flags & AVFMT_NOTIMESTAMPS) ? VSYNC_PASSTHROUGH : + (s->oformat->flags & AVFMT_VARIABLE_FPS) ? VSYNC_VFR : VSYNC_CFR; + + switch (format_video_sync) { + case VSYNC_CFR: + // FIXME set to 0.5 after we fix some dts/pts bugs like in avidec.c + if (delta < -1.1) nb_frames = 0; - else if (format_video_sync == 2) { - if(vdelta<=-0.6){ - nb_frames=0; - }else if(vdelta>0.6) - ost->sync_opts= lrintf(sync_ipts); - }else if (vdelta > 1.1) - nb_frames = lrintf(vdelta); -//fprintf(stderr, "vdelta:%f, ost->sync_opts:%"PRId64", ost->sync_ipts:%f nb_frames:%d\n", vdelta, ost->sync_opts, get_sync_ipts(ost), nb_frames); - if (nb_frames == 0){ - ++nb_frames_drop; - av_log(NULL, AV_LOG_VERBOSE, "*** drop!\n"); - }else if (nb_frames > 1) { - nb_frames_dup += nb_frames - 1; - av_log(NULL, AV_LOG_VERBOSE, "*** %d dup!\n", nb_frames-1); - } - }else - ost->sync_opts= lrintf(sync_ipts); + else if (delta > 1.1) + nb_frames = lrintf(delta); + break; + case VSYNC_VFR: + if (delta <= -0.6) + nb_frames = 0; + else if (delta > 0.6) + ost->sync_opts = lrintf(sync_ipts); + break; + case VSYNC_PASSTHROUGH: + ost->sync_opts = lrintf(sync_ipts); + break; + default: + av_assert0(0); + } nb_frames = FFMIN(nb_frames, ost->max_frames - ost->frame_number); - if (nb_frames <= 0) + if (nb_frames == 0) { + nb_frames_drop++; + av_log(NULL, AV_LOG_VERBOSE, "*** drop!\n"); return; + } else if (nb_frames > 1) { + nb_frames_dup += nb_frames - 1; + av_log(NULL, AV_LOG_VERBOSE, "*** %d dup!\n", nb_frames - 1); + } - do_video_resample(ost, ist, in_picture, &final_picture); + if (!ost->frame_number) + ost->first_pts = ost->sync_opts; /* duplicates frame if needed */ - for(i=0;iindex; + pkt.data = NULL; + pkt.size = 0; + + if (!check_recording_time(ost)) + return; if (s->oformat->flags & AVFMT_RAWPICTURE && enc->codec->id == CODEC_ID_RAWVIDEO) { @@ -1354,16 +1379,17 @@ static void do_video_out(AVFormatContext *s, method. */ enc->coded_frame->interlaced_frame = in_picture->interlaced_frame; enc->coded_frame->top_field_first = in_picture->top_field_first; - pkt.data= (uint8_t *)final_picture; - pkt.size= sizeof(AVPicture); - pkt.pts= av_rescale_q(ost->sync_opts, enc->time_base, ost->st->time_base); + pkt.data = (uint8_t *)in_picture; + pkt.size = sizeof(AVPicture); + pkt.pts = av_rescale_q(ost->sync_opts, enc->time_base, ost->st->time_base); pkt.flags |= AV_PKT_FLAG_KEY; - write_frame(s, &pkt, ost->st->codec, ost->bitstream_filters); + write_frame(s, &pkt, ost); } else { + int got_packet; AVFrame big_picture; - big_picture= *final_picture; + big_picture = *in_picture; /* better than nothing: use input picture interlaced settings */ big_picture.interlaced_frame = in_picture->interlaced_frame; @@ -1379,39 +1405,28 @@ static void do_video_out(AVFormatContext *s, big_picture.quality = quality; if (!enc->me_threshold) big_picture.pict_type = 0; -// big_picture.pts = AV_NOPTS_VALUE; - big_picture.pts= ost->sync_opts; -// big_picture.pts= av_rescale(ost->sync_opts, AV_TIME_BASE*(int64_t)enc->time_base.num, enc->time_base.den); -//av_log(NULL, AV_LOG_DEBUG, "%"PRId64" -> encoder\n", ost->sync_opts); + big_picture.pts = ost->sync_opts; if (ost->forced_kf_index < ost->forced_kf_count && big_picture.pts >= ost->forced_kf_pts[ost->forced_kf_index]) { big_picture.pict_type = AV_PICTURE_TYPE_I; ost->forced_kf_index++; } - ret = avcodec_encode_video(enc, - bit_buffer, bit_buffer_size, - &big_picture); + ret = avcodec_encode_video2(enc, &pkt, &big_picture, &got_packet); if (ret < 0) { av_log(NULL, AV_LOG_FATAL, "Video encoding failed\n"); exit_program(1); } - if(ret>0){ - pkt.data= bit_buffer; - pkt.size= ret; - if(enc->coded_frame->pts != AV_NOPTS_VALUE) - pkt.pts= av_rescale_q(enc->coded_frame->pts, enc->time_base, ost->st->time_base); -/*av_log(NULL, AV_LOG_DEBUG, "encoder -> %"PRId64"/%"PRId64"\n", - pkt.pts != AV_NOPTS_VALUE ? av_rescale(pkt.pts, enc->time_base.den, AV_TIME_BASE*(int64_t)enc->time_base.num) : -1, - pkt.dts != AV_NOPTS_VALUE ? av_rescale(pkt.dts, enc->time_base.den, AV_TIME_BASE*(int64_t)enc->time_base.num) : -1);*/ - - if(enc->coded_frame->key_frame) - pkt.flags |= AV_PKT_FLAG_KEY; - write_frame(s, &pkt, ost->st->codec, ost->bitstream_filters); - *frame_size = ret; - video_size += ret; - //fprintf(stderr,"\nFrame: %3d size: %5d type: %d", - // enc->frame_number-1, ret, enc->pict_type); + if (got_packet) { + if (pkt.pts != AV_NOPTS_VALUE) + pkt.pts = av_rescale_q(pkt.pts, enc->time_base, ost->st->time_base); + if (pkt.dts != AV_NOPTS_VALUE) + pkt.dts = av_rescale_q(pkt.dts, enc->time_base, ost->st->time_base); + + write_frame(s, &pkt, ost); + *frame_size = pkt.size; + video_size += pkt.size; + /* if two pass, output log */ if (ost->logfile && enc->stats_out) { fprintf(ost->logfile, "%s", enc->stats_out); @@ -1419,12 +1434,18 @@ static void do_video_out(AVFormatContext *s, } } ost->sync_opts++; + /* + * For video, number of frames in == number of packets out. + * But there may be reordering, so we can't throw away frames on encoder + * flush, we need to limit them here, before they go into encoder. + */ ost->frame_number++; } } -static double psnr(double d){ - return -10.0*log(d)/log(10.0); +static double psnr(double d) +{ + return -10.0 * log(d) / log(10.0); } static void do_video_stats(AVFormatContext *os, OutputStream *ost, @@ -1446,9 +1467,9 @@ static void do_video_stats(AVFormatContext *os, OutputStream *ost, enc = ost->st->codec; if (enc->codec_type == AVMEDIA_TYPE_VIDEO) { frame_number = ost->frame_number; - fprintf(vstats_file, "frame= %5d q= %2.1f ", frame_number, enc->coded_frame->quality/(float)FF_QP2LAMBDA); + fprintf(vstats_file, "frame= %5d q= %2.1f ", frame_number, enc->coded_frame->quality / (float)FF_QP2LAMBDA); if (enc->flags&CODEC_FLAG_PSNR) - fprintf(vstats_file, "PSNR= %6.2f ", psnr(enc->coded_frame->error[0]/(enc->width*enc->height*255.0*255.0))); + fprintf(vstats_file, "PSNR= %6.2f ", psnr(enc->coded_frame->error[0] / (enc->width * enc->height * 255.0 * 255.0))); fprintf(vstats_file,"f_size= %6d ", frame_size); /* compute pts value */ @@ -1456,10 +1477,10 @@ static void do_video_stats(AVFormatContext *os, OutputStream *ost, if (ti1 < 0.01) ti1 = 0.01; - bitrate = (frame_size * 8) / av_q2d(enc->time_base) / 1000.0; + bitrate = (frame_size * 8) / av_q2d(enc->time_base) / 1000.0; avg_bitrate = (double)(video_size * 8) / ti1 / 1000.0; fprintf(vstats_file, "s_size= %8.0fkB time= %0.3f br= %7.1fkbits/s avg_br= %7.1fkbits/s ", - (double)video_size / 1024, ti1, bitrate, avg_bitrate); + (double)video_size / 1024, ti1, bitrate, avg_bitrate); fprintf(vstats_file, "type= %c\n", av_get_picture_type_char(enc->coded_frame->pict_type)); } } @@ -1498,57 +1519,58 @@ static void print_report(OutputFile *output_files, oc = output_files[0].ctx; total_size = avio_size(oc->pb); - if(total_size<0) // FIXME improve avio_size() so it works with non seekable output too - total_size= avio_tell(oc->pb); + if (total_size < 0) // FIXME improve avio_size() so it works with non seekable output too + total_size = avio_tell(oc->pb); buf[0] = '\0'; ti1 = 1e10; vid = 0; - for(i=0;ist->codec; if (!ost->stream_copy && enc->coded_frame) - q = enc->coded_frame->quality/(float)FF_QP2LAMBDA; + q = enc->coded_frame->quality / (float)FF_QP2LAMBDA; if (vid && enc->codec_type == AVMEDIA_TYPE_VIDEO) { snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "q=%2.1f ", q); } if (!vid && enc->codec_type == AVMEDIA_TYPE_VIDEO) { - float t = (av_gettime()-timer_start) / 1000000.0; + float t = (av_gettime() - timer_start) / 1000000.0; frame_number = ost->frame_number; snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "frame=%5d fps=%3d q=%3.1f ", - frame_number, (t>1)?(int)(frame_number/t+0.5) : 0, q); - if(is_last_report) + frame_number, (t > 1) ? (int)(frame_number / t + 0.5) : 0, q); + if (is_last_report) snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "L"); - if(qp_hist){ + if (qp_hist) { int j; int qp = lrintf(q); - if(qp>=0 && qp= 0 && qp < FF_ARRAY_ELEMS(qp_histogram)) qp_histogram[qp]++; - for(j=0; j<32; j++) - snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "%X", (int)lrintf(log(qp_histogram[j]+1)/log(2))); + for (j = 0; j < 32; j++) + snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "%X", (int)lrintf(log(qp_histogram[j] + 1) / log(2))); } - if (enc->flags&CODEC_FLAG_PSNR){ + if (enc->flags&CODEC_FLAG_PSNR) { int j; - double error, error_sum=0; - double scale, scale_sum=0; - char type[3]= {'Y','U','V'}; + double error, error_sum = 0; + double scale, scale_sum = 0; + char type[3] = { 'Y','U','V' }; snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "PSNR="); - for(j=0; j<3; j++){ - if(is_last_report){ - error= enc->error[j]; - scale= enc->width*enc->height*255.0*255.0*frame_number; - }else{ - error= enc->coded_frame->error[j]; - scale= enc->width*enc->height*255.0*255.0; + for (j = 0; j < 3; j++) { + if (is_last_report) { + error = enc->error[j]; + scale = enc->width * enc->height * 255.0 * 255.0 * frame_number; + } else { + error = enc->coded_frame->error[j]; + scale = enc->width * enc->height * 255.0 * 255.0; } - if(j) scale/=4; + if (j) + scale /= 4; error_sum += error; scale_sum += scale; - snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "%c:%2.2f ", type[j], psnr(error/scale)); + snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "%c:%2.2f ", type[j], psnr(error / scale)); } - snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "*:%2.2f ", psnr(error_sum/scale_sum)); + snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "*:%2.2f ", psnr(error_sum / scale_sum)); } vid = 1; } @@ -1578,10 +1600,10 @@ static void print_report(OutputFile *output_files, int64_t raw= audio_size + video_size + extra_size; av_log(NULL, AV_LOG_INFO, "\n"); av_log(NULL, AV_LOG_INFO, "video:%1.0fkB audio:%1.0fkB global headers:%1.0fkB muxing overhead %f%%\n", - video_size/1024.0, - audio_size/1024.0, - extra_size/1024.0, - 100.0*(total_size - raw)/raw + video_size / 1024.0, + audio_size / 1024.0, + extra_size / 1024.0, + 100.0 * (total_size - raw) / raw ); } } @@ -1594,79 +1616,75 @@ static void flush_encoders(OutputStream *ost_table, int nb_ostreams) OutputStream *ost = &ost_table[i]; AVCodecContext *enc = ost->st->codec; AVFormatContext *os = output_files[ost->file_index].ctx; + int stop_encoding = 0; if (!ost->encoding_needed) continue; - if (ost->st->codec->codec_type == AVMEDIA_TYPE_AUDIO && enc->frame_size <=1) + if (ost->st->codec->codec_type == AVMEDIA_TYPE_AUDIO && enc->frame_size <= 1) continue; if (ost->st->codec->codec_type == AVMEDIA_TYPE_VIDEO && (os->oformat->flags & AVFMT_RAWPICTURE) && enc->codec->id == CODEC_ID_RAWVIDEO) continue; - for(;;) { + for (;;) { AVPacket pkt; - int fifo_bytes; + int fifo_bytes, got_packet; av_init_packet(&pkt); - pkt.stream_index= ost->index; + pkt.data = NULL; + pkt.size = 0; switch (ost->st->codec->codec_type) { case AVMEDIA_TYPE_AUDIO: fifo_bytes = av_fifo_size(ost->fifo); - ret = 0; - /* encode any samples remaining in fifo */ if (fifo_bytes > 0) { - int osize = av_get_bytes_per_sample(enc->sample_fmt); - int fs_tmp = enc->frame_size; + /* encode any samples remaining in fifo */ + int frame_bytes = fifo_bytes; av_fifo_generic_read(ost->fifo, audio_buf, fifo_bytes, NULL); - if (enc->codec->capabilities & CODEC_CAP_SMALL_LAST_FRAME) { - enc->frame_size = fifo_bytes / (osize * enc->channels); - } else { /* pad */ - int frame_bytes = enc->frame_size*osize*enc->channels; + + /* pad last frame with silence if needed */ + if (!(enc->codec->capabilities & CODEC_CAP_SMALL_LAST_FRAME)) { + frame_bytes = enc->frame_size * enc->channels * + av_get_bytes_per_sample(enc->sample_fmt); if (allocated_audio_buf_size < frame_bytes) exit_program(1); generate_silence(audio_buf+fifo_bytes, enc->sample_fmt, frame_bytes - fifo_bytes); } - - ret = avcodec_encode_audio(enc, bit_buffer, bit_buffer_size, (short *)audio_buf); - pkt.duration = av_rescale((int64_t)enc->frame_size*ost->st->time_base.den, - ost->st->time_base.num, enc->sample_rate); - enc->frame_size = fs_tmp; - } - if (ret <= 0) { - ret = avcodec_encode_audio(enc, bit_buffer, bit_buffer_size, NULL); - } - if (ret < 0) { - av_log(NULL, AV_LOG_FATAL, "Audio encoding failed\n"); - exit_program(1); + encode_audio_frame(os, ost, audio_buf, frame_bytes); + } else { + /* flush encoder with NULL frames until it is done + returning packets */ + if (encode_audio_frame(os, ost, NULL, 0) == 0) { + stop_encoding = 1; + break; + } } - audio_size += ret; - pkt.flags |= AV_PKT_FLAG_KEY; break; case AVMEDIA_TYPE_VIDEO: - ret = avcodec_encode_video(enc, bit_buffer, bit_buffer_size, NULL); + ret = avcodec_encode_video2(enc, &pkt, NULL, &got_packet); if (ret < 0) { av_log(NULL, AV_LOG_FATAL, "Video encoding failed\n"); exit_program(1); } video_size += ret; - if(enc->coded_frame && enc->coded_frame->key_frame) - pkt.flags |= AV_PKT_FLAG_KEY; if (ost->logfile && enc->stats_out) { fprintf(ost->logfile, "%s", enc->stats_out); } + if (!got_packet) { + stop_encoding = 1; + break; + } + if (pkt.pts != AV_NOPTS_VALUE) + pkt.pts = av_rescale_q(pkt.pts, enc->time_base, ost->st->time_base); + if (pkt.dts != AV_NOPTS_VALUE) + pkt.dts = av_rescale_q(pkt.dts, enc->time_base, ost->st->time_base); + write_frame(os, &pkt, ost); break; default: - ret=-1; + stop_encoding = 1; } - - if (ret <= 0) + if (stop_encoding) break; - pkt.data = bit_buffer; - pkt.size = ret; - if (enc->coded_frame && enc->coded_frame->pts != AV_NOPTS_VALUE) - pkt.pts= av_rescale_q(enc->coded_frame->pts, enc->time_base, ost->st->time_base); - write_frame(os, &pkt, ost->st->codec, ost->bitstream_filters); } } } @@ -1682,15 +1700,8 @@ static int check_output_constraints(InputStream *ist, OutputStream *ost) if (ost->source_index != ist_index) return 0; - if (of->start_time && ist->pts < of->start_time) - return 0; - - if (of->recording_time != INT64_MAX && - av_compare_ts(ist->pts, AV_TIME_BASE_Q, of->recording_time + of->start_time, - (AVRational){1, 1000000}) >= 0) { - ost->is_past_recording_time = 1; + if (of->start_time && ist->last_dts < of->start_time) return 0; - } return 1; } @@ -1707,6 +1718,12 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p !ost->copy_initial_nonkeyframes) return; + if (of->recording_time != INT64_MAX && + ist->last_dts >= of->recording_time + of->start_time) { + ost->is_past_recording_time = 1; + return; + } + /* force the input stream PTS */ if (ost->st->codec->codec_type == AVMEDIA_TYPE_AUDIO) audio_size += pkt->size; @@ -1715,14 +1732,13 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p ost->sync_opts++; } - opkt.stream_index = ost->index; if (pkt->pts != AV_NOPTS_VALUE) opkt.pts = av_rescale_q(pkt->pts, ist->st->time_base, ost->st->time_base) - ost_tb_start_time; else opkt.pts = AV_NOPTS_VALUE; if (pkt->dts == AV_NOPTS_VALUE) - opkt.dts = av_rescale_q(ist->pts, AV_TIME_BASE_Q, ost->st->time_base); + opkt.dts = av_rescale_q(ist->last_dts, AV_TIME_BASE_Q, ost->st->time_base); else opkt.dts = av_rescale_q(pkt->dts, ist->st->time_base, ost->st->time_base); opkt.dts -= ost_tb_start_time; @@ -1730,10 +1746,11 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p opkt.duration = av_rescale_q(pkt->duration, ist->st->time_base, ost->st->time_base); opkt.flags = pkt->flags; - //FIXME remove the following 2 lines they shall be replaced by the bitstream filters - if( ost->st->codec->codec_id != CODEC_ID_H264 + // FIXME remove the following 2 lines they shall be replaced by the bitstream filters + if ( ost->st->codec->codec_id != CODEC_ID_H264 && ost->st->codec->codec_id != CODEC_ID_MPEG1VIDEO && ost->st->codec->codec_id != CODEC_ID_MPEG2VIDEO + && ost->st->codec->codec_id != CODEC_ID_VC1 ) { if (av_parser_change(ist->st->parser, ost->st->codec, &opkt.data, &opkt.size, pkt->data, pkt->size, pkt->flags & AV_PKT_FLAG_KEY)) opkt.destruct = av_destruct_packet; @@ -1742,16 +1759,15 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p opkt.size = pkt->size; } - write_frame(of->ctx, &opkt, ost->st->codec, ost->bitstream_filters); + write_frame(of->ctx, &opkt, ost); ost->st->codec->frame_number++; - ost->frame_number++; av_free_packet(&opkt); } static void rate_emu_sleep(InputStream *ist) { if (input_files[ist->file_index].rate_emu) { - int64_t pts = av_rescale(ist->pts, 1000000, AV_TIME_BASE); + int64_t pts = av_rescale(ist->last_dts, 1000000, AV_TIME_BASE); int64_t now = av_gettime() - ist->start; if (pts > now) usleep(pts - now); @@ -1784,11 +1800,11 @@ static int transcode_audio(InputStream *ist, AVPacket *pkt, int *got_output) /* if the decoder provides a pts, use it instead of the last packet pts. the decoder could be delaying output by a packet or more. */ if (decoded_frame->pts != AV_NOPTS_VALUE) - ist->next_pts = decoded_frame->pts; + ist->next_dts = decoded_frame->pts; - /* increment next_pts to use for the case where the input stream does not + /* increment next_dts to use for the case where the input stream does not have timestamps or there are multiple frames in the packet */ - ist->next_pts += ((int64_t)AV_TIME_BASE * decoded_frame->nb_samples) / + ist->next_dts += ((int64_t)AV_TIME_BASE * decoded_frame->nb_samples) / avctx->sample_rate; // preprocess audio (volume) @@ -1868,9 +1884,7 @@ static int transcode_video(InputStream *ist, AVPacket *pkt, int *got_output, int void *buffer_to_free = NULL; int i, ret = 0; float quality; -#if CONFIG_AVFILTER int frame_available = 1; -#endif if (!ist->decoded_frame && !(ist->decoded_frame = avcodec_alloc_frame())) return AVERROR(ENOMEM); @@ -1878,7 +1892,7 @@ static int transcode_video(InputStream *ist, AVPacket *pkt, int *got_output, int avcodec_get_frame_defaults(ist->decoded_frame); decoded_frame = ist->decoded_frame; pkt->pts = *pkt_pts; - pkt->dts = ist->pts; + pkt->dts = ist->last_dts; *pkt_pts = AV_NOPTS_VALUE; ret = avcodec_decode_video2(ist->st->codec, @@ -1891,69 +1905,91 @@ static int transcode_video(InputStream *ist, AVPacket *pkt, int *got_output, int /* no picture yet */ return ret; } - ist->next_pts = ist->pts = guess_correct_pts(&ist->pts_ctx, decoded_frame->pkt_pts, - decoded_frame->pkt_dts); - if (pkt->duration) - ist->next_pts += av_rescale_q(pkt->duration, ist->st->time_base, AV_TIME_BASE_Q); - else if (ist->st->codec->time_base.num != 0) { - int ticks = ist->st->parser ? ist->st->parser->repeat_pict + 1 : - ist->st->codec->ticks_per_frame; - ist->next_pts += ((int64_t)AV_TIME_BASE * - ist->st->codec->time_base.num * ticks) / - ist->st->codec->time_base.den; - } + decoded_frame->pts = guess_correct_pts(&ist->pts_ctx, decoded_frame->pkt_pts, + decoded_frame->pkt_dts); pkt->size = 0; pre_process_video_frame(ist, (AVPicture *)decoded_frame, &buffer_to_free); rate_emu_sleep(ist); + if (ist->st->sample_aspect_ratio.num) + decoded_frame->sample_aspect_ratio = ist->st->sample_aspect_ratio; + for (i = 0; i < nb_output_streams; i++) { OutputStream *ost = &output_streams[i]; - int frame_size; + int frame_size, resample_changed; if (!check_output_constraints(ist, ost) || !ost->encoding_needed) continue; -#if CONFIG_AVFILTER - if (ost->input_video_filter) { - AVRational sar; - if (ist->st->sample_aspect_ratio.num) - sar = ist->st->sample_aspect_ratio; - else - sar = ist->st->codec->sample_aspect_ratio; - av_vsrc_buffer_add_frame(ost->input_video_filter, decoded_frame, ist->pts, sar); - if (!ist->filtered_frame && !(ist->filtered_frame = avcodec_alloc_frame())) { - av_free(buffer_to_free); - return AVERROR(ENOMEM); - } else - avcodec_get_frame_defaults(ist->filtered_frame); - filtered_frame = ist->filtered_frame; - frame_available = avfilter_poll_frame(ost->output_video_filter->inputs[0]); + resample_changed = ost->resample_width != decoded_frame->width || + ost->resample_height != decoded_frame->height || + ost->resample_pix_fmt != decoded_frame->format; + if (resample_changed) { + av_log(NULL, AV_LOG_INFO, + "Input stream #%d:%d frame changed from size:%dx%d fmt:%s to size:%dx%d fmt:%s\n", + ist->file_index, ist->st->index, + ost->resample_width, ost->resample_height, av_get_pix_fmt_name(ost->resample_pix_fmt), + decoded_frame->width, decoded_frame->height, av_get_pix_fmt_name(decoded_frame->format)); + + avfilter_graph_free(&ost->graph); + if (configure_video_filters(ist, ost)) { + av_log(NULL, AV_LOG_FATAL, "Error reinitializing filters!\n"); + exit_program(1); + } + + ost->resample_width = decoded_frame->width; + ost->resample_height = decoded_frame->height; + ost->resample_pix_fmt = decoded_frame->format; } + + if (ist->st->codec->codec->capabilities & CODEC_CAP_DR1) { + FrameBuffer *buf = decoded_frame->opaque; + AVFilterBufferRef *fb = avfilter_get_video_buffer_ref_from_arrays( + decoded_frame->data, decoded_frame->linesize, + AV_PERM_READ | AV_PERM_PRESERVE, + ist->st->codec->width, ist->st->codec->height, + ist->st->codec->pix_fmt); + + avfilter_copy_frame_props(fb, decoded_frame); + fb->buf->priv = buf; + fb->buf->free = filter_release_buffer; + + buf->refcount++; + av_buffersrc_buffer(ost->input_video_filter, fb); + } else + av_vsrc_buffer_add_frame(ost->input_video_filter, decoded_frame, + decoded_frame->pts, decoded_frame->sample_aspect_ratio); + + if (!ist->filtered_frame && !(ist->filtered_frame = avcodec_alloc_frame())) { + ret = AVERROR(ENOMEM); + goto fail; + } else + avcodec_get_frame_defaults(ist->filtered_frame); + filtered_frame = ist->filtered_frame; + + frame_available = avfilter_poll_frame(ost->output_video_filter->inputs[0]); while (frame_available) { + AVFilterBufferRef *picref; AVRational ist_pts_tb; - if (ost->output_video_filter) - get_filtered_video_frame(ost->output_video_filter, filtered_frame, &ost->picref, &ist_pts_tb); - if (ost->picref) - ist->pts = av_rescale_q(ost->picref->pts, ist_pts_tb, AV_TIME_BASE_Q); - if (ost->picref->video && !ost->frame_aspect_ratio) - ost->st->codec->sample_aspect_ratio = ost->picref->video->pixel_aspect; -#else - filtered_frame = decoded_frame; -#endif - - do_video_out(output_files[ost->file_index].ctx, ost, ist, filtered_frame, &frame_size, + if ((ret = get_filtered_video_frame(ost->output_video_filter, + filtered_frame, &picref, + &ist_pts_tb)) < 0) + goto fail; + filtered_frame->pts = av_rescale_q(picref->pts, ist_pts_tb, AV_TIME_BASE_Q); + if (!ost->frame_aspect_ratio) + ost->st->codec->sample_aspect_ratio = picref->video->pixel_aspect; + + do_video_out(output_files[ost->file_index].ctx, ost, filtered_frame, &frame_size, same_quant ? quality : ost->st->codec->global_quality); if (vstats_filename && frame_size) do_video_stats(output_files[ost->file_index].ctx, ost, frame_size); -#if CONFIG_AVFILTER frame_available = ost->output_video_filter && avfilter_poll_frame(ost->output_video_filter->inputs[0]); - if (ost->picref) - avfilter_unref_buffer(ost->picref); + avfilter_unref_buffer(picref); } -#endif } +fail: av_free(buffer_to_free); return ret; } @@ -1993,8 +2029,8 @@ static int output_packet(InputStream *ist, int64_t pkt_pts = AV_NOPTS_VALUE; AVPacket avpkt; - if (ist->next_pts == AV_NOPTS_VALUE) - ist->next_pts = ist->pts; + if (ist->next_dts == AV_NOPTS_VALUE) + ist->next_dts = ist->last_dts; if (pkt == NULL) { /* EOF handling */ @@ -2006,17 +2042,17 @@ static int output_packet(InputStream *ist, avpkt = *pkt; } - if(pkt->dts != AV_NOPTS_VALUE) - ist->next_pts = ist->pts = av_rescale_q(pkt->dts, ist->st->time_base, AV_TIME_BASE_Q); - if(pkt->pts != AV_NOPTS_VALUE) + if (pkt->dts != AV_NOPTS_VALUE) + ist->next_dts = ist->last_dts = av_rescale_q(pkt->dts, ist->st->time_base, AV_TIME_BASE_Q); + if (pkt->pts != AV_NOPTS_VALUE) pkt_pts = av_rescale_q(pkt->pts, ist->st->time_base, AV_TIME_BASE_Q); - //while we have more to decode or while the decoder did output something on EOF + // while we have more to decode or while the decoder did output something on EOF while (ist->decoding_needed && (avpkt.size > 0 || (!pkt && got_output))) { int ret = 0; handle_eof: - ist->pts = ist->next_pts; + ist->last_dts = ist->next_dts; if (avpkt.size && avpkt.size != pkt->size) { av_log(NULL, ist->showed_multi_packet_warning ? AV_LOG_VERBOSE : AV_LOG_WARNING, @@ -2024,12 +2060,23 @@ static int output_packet(InputStream *ist, ist->showed_multi_packet_warning = 1; } - switch(ist->st->codec->codec_type) { + switch (ist->st->codec->codec_type) { case AVMEDIA_TYPE_AUDIO: ret = transcode_audio (ist, &avpkt, &got_output); break; case AVMEDIA_TYPE_VIDEO: ret = transcode_video (ist, &avpkt, &got_output, &pkt_pts); + if (avpkt.duration) + ist->next_dts += av_rescale_q(avpkt.duration, ist->st->time_base, AV_TIME_BASE_Q); + else if (ist->st->r_frame_rate.num) + ist->next_dts += av_rescale_q(1, (AVRational){ist->st->r_frame_rate.den, + ist->st->r_frame_rate.num}, + AV_TIME_BASE_Q); + else if (ist->st->codec->time_base.num != 0) { + int ticks = ist->st->parser ? ist->st->parser->repeat_pict + 1 : + ist->st->codec->ticks_per_frame; + ist->next_dts += av_rescale_q(ticks, ist->st->codec->time_base, AV_TIME_BASE_Q); + } break; case AVMEDIA_TYPE_SUBTITLE: ret = transcode_subtitles(ist, &avpkt, &got_output); @@ -2053,16 +2100,16 @@ static int output_packet(InputStream *ist, /* handle stream copy */ if (!ist->decoding_needed) { rate_emu_sleep(ist); - ist->pts = ist->next_pts; + ist->last_dts = ist->next_dts; switch (ist->st->codec->codec_type) { case AVMEDIA_TYPE_AUDIO: - ist->next_pts += ((int64_t)AV_TIME_BASE * ist->st->codec->frame_size) / + ist->next_dts += ((int64_t)AV_TIME_BASE * ist->st->codec->frame_size) / ist->st->codec->sample_rate; break; case AVMEDIA_TYPE_VIDEO: if (ist->st->codec->time_base.num != 0) { - int ticks = ist->st->parser ? ist->st->parser->repeat_pict+1 : ist->st->codec->ticks_per_frame; - ist->next_pts += ((int64_t)AV_TIME_BASE * + int ticks = ist->st->parser ? ist->st->parser->repeat_pict + 1 : ist->st->codec->ticks_per_frame; + ist->next_dts += ((int64_t)AV_TIME_BASE * ist->st->codec->time_base.num * ticks) / ist->st->codec->time_base.den; } @@ -2085,7 +2132,7 @@ static void print_sdp(OutputFile *output_files, int n) { char sdp[2048]; int i; - AVFormatContext **avc = av_malloc(sizeof(*avc)*n); + AVFormatContext **avc = av_malloc(sizeof(*avc) * n); if (!avc) exit_program(1); @@ -2098,6 +2145,51 @@ static void print_sdp(OutputFile *output_files, int n) av_freep(&avc); } +static void get_default_channel_layouts(OutputStream *ost, InputStream *ist) +{ + char layout_name[256]; + AVCodecContext *enc = ost->st->codec; + AVCodecContext *dec = ist->st->codec; + + if (!dec->channel_layout) { + if (enc->channel_layout && dec->channels == enc->channels) { + dec->channel_layout = enc->channel_layout; + } else { + dec->channel_layout = av_get_default_channel_layout(dec->channels); + + if (!dec->channel_layout) { + av_log(NULL, AV_LOG_FATAL, "Unable to find default channel " + "layout for Input Stream #%d.%d\n", ist->file_index, + ist->st->index); + exit_program(1); + } + } + av_get_channel_layout_string(layout_name, sizeof(layout_name), + dec->channels, dec->channel_layout); + av_log(NULL, AV_LOG_WARNING, "Guessed Channel Layout for Input Stream " + "#%d.%d : %s\n", ist->file_index, ist->st->index, layout_name); + } + if (!enc->channel_layout) { + if (dec->channels == enc->channels) { + enc->channel_layout = dec->channel_layout; + return; + } else { + enc->channel_layout = av_get_default_channel_layout(enc->channels); + } + if (!enc->channel_layout) { + av_log(NULL, AV_LOG_FATAL, "Unable to find default channel layout " + "for Output Stream #%d.%d\n", ost->file_index, + ost->st->index); + exit_program(1); + } + av_get_channel_layout_string(layout_name, sizeof(layout_name), + enc->channels, enc->channel_layout); + av_log(NULL, AV_LOG_WARNING, "Guessed Channel Layout for Output Stream " + "#%d.%d : %s\n", ost->file_index, ost->st->index, layout_name); + } +} + + static int init_input_stream(int ist_index, OutputStream *output_streams, int nb_output_streams, char *error, int error_len) { @@ -2127,6 +2219,8 @@ static int init_input_stream(int ist_index, OutputStream *output_streams, int nb ist->st->codec->opaque = ist; } + if (!av_dict_get(ist->opts, "threads", NULL, 0)) + av_dict_set(&ist->opts, "threads", "auto", 0); if (avcodec_open2(ist->st->codec, codec, &ist->opts) < 0) { snprintf(error, error_len, "Error while opening decoder for input stream #%d:%d", ist->file_index, ist->st->index); @@ -2134,10 +2228,21 @@ static int init_input_stream(int ist_index, OutputStream *output_streams, int nb } assert_codec_experimental(ist->st->codec, 0); assert_avoptions(ist->opts); + + if (ist->st->codec->codec_type == AVMEDIA_TYPE_AUDIO) { + for (i = 0; i < nb_output_streams; i++) { + OutputStream *ost = &output_streams[i]; + if (ost->source_index == ist_index) { + if (!ist->st->codec->channel_layout || !ost->st->codec->channel_layout) + get_default_channel_layouts(ost, ist); + break; + } + } + } } - ist->pts = ist->st->avg_frame_rate.num ? - ist->st->codec->has_b_frames*AV_TIME_BASE / av_q2d(ist->st->avg_frame_rate) : 0; - ist->next_pts = AV_NOPTS_VALUE; + ist->last_dts = ist->st->avg_frame_rate.num ? - ist->st->codec->has_b_frames * AV_TIME_BASE / av_q2d(ist->st->avg_frame_rate) : 0; + ist->next_dts = AV_NOPTS_VALUE; init_pts_correction(&ist->pts_ctx); ist->is_start = 1; @@ -2227,9 +2332,9 @@ static int transcode_init(OutputFile *output_files, } else codec->time_base = ist->st->time_base; - switch(codec->codec_type) { + switch (codec->codec_type) { case AVMEDIA_TYPE_AUDIO: - if(audio_volume != 256) { + if (audio_volume != 256) { av_log(NULL, AV_LOG_FATAL, "-acodec copy and -vol are incompatible (frames are not decoded)\n"); exit_program(1); } @@ -2270,7 +2375,7 @@ static int transcode_init(OutputFile *output_files, ist->decoding_needed = 1; ost->encoding_needed = 1; - switch(codec->codec_type) { + switch (codec->codec_type) { case AVMEDIA_TYPE_AUDIO: ost->fifo = av_fifo_alloc(1024); if (!ost->fifo) { @@ -2281,7 +2386,7 @@ static int transcode_init(OutputFile *output_files, if (!codec->sample_rate) codec->sample_rate = icodec->sample_rate; choose_sample_rate(ost->st, ost->enc); - codec->time_base = (AVRational){1, codec->sample_rate}; + codec->time_base = (AVRational){ 1, codec->sample_rate }; if (codec->sample_fmt == AV_SAMPLE_FMT_NONE) codec->sample_fmt = icodec->sample_fmt; @@ -2318,49 +2423,45 @@ static int transcode_init(OutputFile *output_files, codec->height != icodec->height || codec->pix_fmt != icodec->pix_fmt; if (ost->video_resample) { -#if !CONFIG_AVFILTER - avcodec_get_frame_defaults(&ost->pict_tmp); - if(avpicture_alloc((AVPicture*)&ost->pict_tmp, codec->pix_fmt, - codec->width, codec->height)) { - av_log(NULL, AV_LOG_FATAL, "Cannot allocate temp picture, check pix fmt\n"); - exit_program(1); - } - ost->img_resample_ctx = sws_getContext( - icodec->width, - icodec->height, - icodec->pix_fmt, - codec->width, - codec->height, - codec->pix_fmt, - ost->sws_flags, NULL, NULL, NULL); - if (ost->img_resample_ctx == NULL) { - av_log(NULL, AV_LOG_FATAL, "Cannot get resampling context\n"); - exit_program(1); - } -#endif - codec->bits_per_raw_sample= 0; + codec->bits_per_raw_sample = 0; } ost->resample_height = icodec->height; ost->resample_width = icodec->width; ost->resample_pix_fmt = icodec->pix_fmt; - if (!ost->frame_rate.num) - ost->frame_rate = ist->st->r_frame_rate.num ? ist->st->r_frame_rate : (AVRational){25,1}; - if (ost->enc && ost->enc->supported_framerates && !ost->force_fps) { - int idx = av_find_nearest_q_idx(ost->frame_rate, ost->enc->supported_framerates); - ost->frame_rate = ost->enc->supported_framerates[idx]; + /* + * We want CFR output if and only if one of those is true: + * 1) user specified output framerate with -r + * 2) user specified -vsync cfr + * 3) output format is CFR and the user didn't force vsync to + * something else than CFR + * + * in such a case, set ost->frame_rate + */ + if (!ost->frame_rate.num && + (video_sync_method == VSYNC_CFR || + (video_sync_method == VSYNC_AUTO && + !(oc->oformat->flags & (AVFMT_NOTIMESTAMPS | AVFMT_VARIABLE_FPS))))) { + ost->frame_rate = ist->st->r_frame_rate.num ? ist->st->r_frame_rate : (AVRational){25, 1}; + if (ost->enc && ost->enc->supported_framerates && !ost->force_fps) { + int idx = av_find_nearest_q_idx(ost->frame_rate, ost->enc->supported_framerates); + ost->frame_rate = ost->enc->supported_framerates[idx]; + } } - codec->time_base = (AVRational){ost->frame_rate.den, ost->frame_rate.num}; + if (ost->frame_rate.num) { + codec->time_base = (AVRational){ost->frame_rate.den, ost->frame_rate.num}; + video_sync_method = VSYNC_CFR; + } else + codec->time_base = ist->st->time_base; -#if CONFIG_AVFILTER if (configure_video_filters(ist, ost)) { av_log(NULL, AV_LOG_FATAL, "Error opening filters!\n"); exit(1); } -#endif break; case AVMEDIA_TYPE_SUBTITLE: + codec->time_base = (AVRational){1, 1000}; break; default: abort(); @@ -2374,38 +2475,30 @@ static int transcode_init(OutputFile *output_files, snprintf(logfilename, sizeof(logfilename), "%s-%d.log", pass_logfilename_prefix ? pass_logfilename_prefix : DEFAULT_PASS_LOGFILENAME_PREFIX, i); - if (codec->flags & CODEC_FLAG_PASS1) { - f = fopen(logfilename, "wb"); - if (!f) { - av_log(NULL, AV_LOG_FATAL, "Cannot write log file '%s' for pass-1 encoding: %s\n", - logfilename, strerror(errno)); - exit_program(1); - } - ost->logfile = f; + if (!strcmp(ost->enc->name, "libx264")) { + av_dict_set(&ost->opts, "stats", logfilename, AV_DICT_DONT_OVERWRITE); } else { - char *logbuffer; - size_t logbuffer_size; - if (cmdutils_read_file(logfilename, &logbuffer, &logbuffer_size) < 0) { - av_log(NULL, AV_LOG_FATAL, "Error reading log file '%s' for pass-2 encoding\n", - logfilename); - exit_program(1); + if (codec->flags & CODEC_FLAG_PASS1) { + f = fopen(logfilename, "wb"); + if (!f) { + av_log(NULL, AV_LOG_FATAL, "Cannot write log file '%s' for pass-1 encoding: %s\n", + logfilename, strerror(errno)); + exit_program(1); + } + ost->logfile = f; + } else { + char *logbuffer; + size_t logbuffer_size; + if (cmdutils_read_file(logfilename, &logbuffer, &logbuffer_size) < 0) { + av_log(NULL, AV_LOG_FATAL, "Error reading log file '%s' for pass-2 encoding\n", + logfilename); + exit_program(1); + } + codec->stats_in = logbuffer; } - codec->stats_in = logbuffer; } } } - if(codec->codec_type == AVMEDIA_TYPE_VIDEO){ - int size = codec->width * codec->height; - bit_buffer_size = FFMAX(bit_buffer_size, 6*size + 200); - } - } - - if (!bit_buffer) - bit_buffer = av_malloc(bit_buffer_size); - if (!bit_buffer) { - av_log(NULL, AV_LOG_ERROR, "Cannot allocate %d bytes output buffer\n", - bit_buffer_size); - return AVERROR(ENOMEM); } /* open each encoder */ @@ -2429,6 +2522,8 @@ static int transcode_init(OutputFile *output_files, memcpy(ost->st->codec->subtitle_header, dec->subtitle_header, dec->subtitle_header_size); ost->st->codec->subtitle_header_size = dec->subtitle_header_size; } + if (!av_dict_get(ost->opts, "threads", NULL, 0)) + av_dict_set(&ost->opts, "threads", "auto", 0); if (avcodec_open2(ost->st->codec, codec, &ost->opts) < 0) { snprintf(error, sizeof(error), "Error while opening encoder for output stream #%d:%d - maybe incorrect parameters such as bit_rate, rate, width or height", ost->file_index, ost->index); @@ -2544,7 +2639,7 @@ static int transcode(OutputFile *output_files, OutputStream *ost; InputStream *ist; uint8_t *no_packet; - int no_packet_count=0; + int no_packet_count = 0; int64_t timer_start; if (!(no_packet = av_mallocz(nb_input_files))) @@ -2559,14 +2654,14 @@ static int transcode(OutputFile *output_files, timer_start = av_gettime(); - for(; received_sigterm == 0;) { + for (; received_sigterm == 0;) { int file_index, ist_index; AVPacket pkt; int64_t ipts_min; double opts_min; ipts_min = INT64_MAX; - opts_min= 1e100; + opts_min = 1e100; /* select the stream that we must read now by looking at the smallest output pts */ @@ -2583,15 +2678,16 @@ static int transcode(OutputFile *output_files, (os->pb && avio_tell(os->pb) >= of->limit_filesize)) continue; opts = ost->st->pts.val * av_q2d(ost->st->time_base); - ipts = ist->pts; - if (!input_files[ist->file_index].eof_reached){ - if(ipts < ipts_min) { + ipts = ist->last_dts; + if (!input_files[ist->file_index].eof_reached) { + if (ipts < ipts_min) { ipts_min = ipts; - if(input_sync ) file_index = ist->file_index; + if (input_sync) + file_index = ist->file_index; } - if(opts < opts_min) { + if (opts < opts_min) { opts_min = opts; - if(!input_sync) file_index = ist->file_index; + if (!input_sync) file_index = ist->file_index; } } if (ost->frame_number >= ost->max_frames) { @@ -2603,8 +2699,8 @@ static int transcode(OutputFile *output_files, } /* if none, if is finished */ if (file_index < 0) { - if(no_packet_count){ - no_packet_count=0; + if (no_packet_count) { + no_packet_count = 0; memset(no_packet, 0, nb_input_files); usleep(10000); continue; @@ -2613,10 +2709,10 @@ static int transcode(OutputFile *output_files, } /* read a frame from it and output it in the fifo */ - is = input_files[file_index].ctx; - ret= av_read_frame(is, &pkt); - if(ret == AVERROR(EAGAIN)){ - no_packet[file_index]=1; + is = input_files[file_index].ctx; + ret = av_read_frame(is, &pkt); + if (ret == AVERROR(EAGAIN)) { + no_packet[file_index] = 1; no_packet_count++; continue; } @@ -2628,7 +2724,7 @@ static int transcode(OutputFile *output_files, continue; } - no_packet_count=0; + no_packet_count = 0; memset(no_packet, 0, nb_input_files); if (do_pkt_dump) { @@ -2649,27 +2745,31 @@ static int transcode(OutputFile *output_files, if (pkt.pts != AV_NOPTS_VALUE) pkt.pts += av_rescale_q(input_files[ist->file_index].ts_offset, AV_TIME_BASE_Q, ist->st->time_base); - if(pkt.pts != AV_NOPTS_VALUE) + if (pkt.pts != AV_NOPTS_VALUE) pkt.pts *= ist->ts_scale; - if(pkt.dts != AV_NOPTS_VALUE) + if (pkt.dts != AV_NOPTS_VALUE) pkt.dts *= ist->ts_scale; -// fprintf(stderr, "next:%"PRId64" dts:%"PRId64" off:%"PRId64" %d\n", ist->next_pts, pkt.dts, input_files[ist->file_index].ts_offset, ist->st->codec->codec_type); - if (pkt.dts != AV_NOPTS_VALUE && ist->next_pts != AV_NOPTS_VALUE + //fprintf(stderr, "next:%"PRId64" dts:%"PRId64" off:%"PRId64" %d\n", + // ist->next_dts, + // pkt.dts, input_files[ist->file_index].ts_offset, + // ist->st->codec->codec_type); + if (pkt.dts != AV_NOPTS_VALUE && ist->next_dts != AV_NOPTS_VALUE && (is->iformat->flags & AVFMT_TS_DISCONT)) { - int64_t pkt_dts= av_rescale_q(pkt.dts, ist->st->time_base, AV_TIME_BASE_Q); - int64_t delta= pkt_dts - ist->next_pts; - if((FFABS(delta) > 1LL*dts_delta_threshold*AV_TIME_BASE || pkt_dts+1pts)&& !copy_ts){ + int64_t pkt_dts = av_rescale_q(pkt.dts, ist->st->time_base, AV_TIME_BASE_Q); + int64_t delta = pkt_dts - ist->next_dts; + if ((FFABS(delta) > 1LL * dts_delta_threshold * AV_TIME_BASE || pkt_dts + 1 < ist->last_dts) && !copy_ts) { input_files[ist->file_index].ts_offset -= delta; - av_log(NULL, AV_LOG_DEBUG, "timestamp discontinuity %"PRId64", new offset= %"PRId64"\n", + av_log(NULL, AV_LOG_DEBUG, + "timestamp discontinuity %"PRId64", new offset= %"PRId64"\n", delta, input_files[ist->file_index].ts_offset); pkt.dts-= av_rescale_q(delta, AV_TIME_BASE_Q, ist->st->time_base); - if(pkt.pts != AV_NOPTS_VALUE) + if (pkt.pts != AV_NOPTS_VALUE) pkt.pts-= av_rescale_q(delta, AV_TIME_BASE_Q, ist->st->time_base); } } - //fprintf(stderr,"read #%d.%d size=%d\n", ist->file_index, ist->st->index, pkt.size); + // fprintf(stderr,"read #%d.%d size=%d\n", ist->file_index, ist->st->index, pkt.size); if (output_packet(ist, output_streams, nb_output_streams, &pkt) < 0) { av_log(NULL, AV_LOG_ERROR, "Error while decoding stream #%d:%d\n", @@ -2699,7 +2799,7 @@ static int transcode(OutputFile *output_files, term_exit(); /* write the trailer if needed and close file */ - for(i=0;ist->codec->stats_in); avcodec_close(ost->st->codec); } -#if CONFIG_AVFILTER avfilter_graph_free(&ost->graph); -#endif } /* close each decoder */ @@ -2731,7 +2829,6 @@ static int transcode(OutputFile *output_files, ret = 0; fail: - av_freep(&bit_buffer); av_freep(&no_packet); if (output_streams) { @@ -2747,10 +2844,7 @@ static int transcode(OutputFile *output_files, av_fifo_free(ost->fifo); /* works even if fifo is not initialized but set to zero */ av_freep(&ost->st->codec->subtitle_header); - av_free(ost->pict_tmp.data[0]); av_free(ost->forced_kf_pts); - if (ost->video_resample) - sws_freeContext(ost->img_resample_ctx); if (ost->resample) audio_resample_close(ost->resample); if (ost->reformat_ctx) @@ -2773,7 +2867,7 @@ static double parse_frame_aspect_ratio(const char *arg) if (p) { x = strtol(arg, &end, 10); if (end == p) - y = strtol(end+1, &end, 10); + y = strtol(end + 1, &end, 10); if (x > 0 && y > 0) ar = (double)x / (double)y; } else @@ -3011,11 +3105,11 @@ static AVCodec *find_codec_or_die(const char *name, enum AVMediaType type, int e codec = encoder ? avcodec_find_encoder_by_name(name) : avcodec_find_decoder_by_name(name); - if(!codec) { + if (!codec) { av_log(NULL, AV_LOG_FATAL, "Unknown %s '%s'\n", codec_string, name); exit_program(1); } - if(codec->type != type) { + if (codec->type != type) { av_log(NULL, AV_LOG_FATAL, "Invalid %s type '%s'\n", codec_string, name); exit_program(1); } @@ -3041,7 +3135,7 @@ static AVCodec *choose_decoder(OptionsContext *o, AVFormatContext *s, AVStream * */ static void add_input_streams(OptionsContext *o, AVFormatContext *ic) { - int i, rfps, rfps_base; + int i; for (i = 0; i < ic->nb_streams; i++) { AVStream *st = ic->streams[i]; @@ -3053,6 +3147,7 @@ static void add_input_streams(OptionsContext *o, AVFormatContext *ic) ist->st = st; ist->file_index = nb_input_files; ist->discard = 1; + st->discard = AVDISCARD_ALL; ist->opts = filter_codec_opts(codec_opts, ist->st->codec->codec_id, ic, st); ist->ts_scale = 1.0; @@ -3061,37 +3156,17 @@ static void add_input_streams(OptionsContext *o, AVFormatContext *ic) ist->dec = choose_decoder(o, ic, st); switch (dec->codec_type) { - case AVMEDIA_TYPE_AUDIO: - if (o->audio_disable) - st->discard= AVDISCARD_ALL; - break; case AVMEDIA_TYPE_VIDEO: - rfps = ic->streams[i]->r_frame_rate.num; - rfps_base = ic->streams[i]->r_frame_rate.den; if (dec->lowres) { dec->flags |= CODEC_FLAG_EMU_EDGE; dec->height >>= dec->lowres; dec->width >>= dec->lowres; } - if (dec->time_base.den != rfps*dec->ticks_per_frame || dec->time_base.num != rfps_base) { - - av_log(NULL, AV_LOG_INFO,"\nSeems stream %d codec frame rate differs from container frame rate: %2.2f (%d/%d) -> %2.2f (%d/%d)\n", - i, (float)dec->time_base.den / dec->time_base.num, dec->time_base.den, dec->time_base.num, - (float)rfps / rfps_base, rfps, rfps_base); - } - - if (o->video_disable) - st->discard= AVDISCARD_ALL; - else if(video_discard) - st->discard= video_discard; break; + case AVMEDIA_TYPE_AUDIO: case AVMEDIA_TYPE_DATA: - break; case AVMEDIA_TYPE_SUBTITLE: - if (o->subtitle_disable) - st->discard = AVDISCARD_ALL; - break; case AVMEDIA_TYPE_ATTACHMENT: case AVMEDIA_TYPE_UNKNOWN: break; @@ -3286,7 +3361,7 @@ static void parse_forced_key_frames(char *kf, OutputStream *ost, if (*p == ',') n++; ost->forced_kf_count = n; - ost->forced_kf_pts = av_malloc(sizeof(*ost->forced_kf_pts) * n); + ost->forced_kf_pts = av_malloc(sizeof(*ost->forced_kf_pts) * n); if (!ost->forced_kf_pts) { av_log(NULL, AV_LOG_FATAL, "Could not allocate forced key frames array.\n"); exit_program(1); @@ -3383,8 +3458,8 @@ static OutputStream *new_output_stream(OptionsContext *o, AVFormatContext *oc, e nb_output_streams + 1); ost = &output_streams[nb_output_streams - 1]; ost->file_index = nb_output_files; - ost->index = idx; - ost->st = st; + ost->index = idx; + ost->st = st; st->codec->codec_type = type; choose_encoder(o, oc, ost); if (ost->enc) { @@ -3464,12 +3539,12 @@ static void parse_matrix_coeffs(uint16_t *dest, const char *str) { int i; const char *p = str; - for(i = 0;; i++) { + for (i = 0;; i++) { dest[i] = atoi(p); - if(i == 63) + if (i == 63) break; p = strchr(p, ','); - if(!p) { + if (!p) { av_log(NULL, AV_LOG_FATAL, "Syntax error in matrix \"%s\" at coeff %d\n", str, i); exit_program(1); } @@ -3535,33 +3610,33 @@ static OutputStream *new_video_stream(OptionsContext *o, AVFormatContext *oc) } MATCH_PER_STREAM_OPT(rc_overrides, str, p, oc, st); - for(i=0; p; i++){ + for (i = 0; p; i++) { int start, end, q; - int e=sscanf(p, "%d,%d,%d", &start, &end, &q); - if(e!=3){ + int e = sscanf(p, "%d,%d,%d", &start, &end, &q); + if (e != 3) { av_log(NULL, AV_LOG_FATAL, "error parsing rc_override\n"); exit_program(1); } - video_enc->rc_override= + video_enc->rc_override = av_realloc(video_enc->rc_override, - sizeof(RcOverride)*(i+1)); - video_enc->rc_override[i].start_frame= start; - video_enc->rc_override[i].end_frame = end; - if(q>0){ - video_enc->rc_override[i].qscale= q; - video_enc->rc_override[i].quality_factor= 1.0; + sizeof(RcOverride) * (i + 1)); + video_enc->rc_override[i].start_frame = start; + video_enc->rc_override[i].end_frame = end; + if (q > 0) { + video_enc->rc_override[i].qscale = q; + video_enc->rc_override[i].quality_factor = 1.0; } - else{ - video_enc->rc_override[i].qscale= 0; - video_enc->rc_override[i].quality_factor= -q/100.0; + else { + video_enc->rc_override[i].qscale = 0; + video_enc->rc_override[i].quality_factor = -q/100.0; } - p= strchr(p, '/'); - if(p) p++; + p = strchr(p, '/'); + if (p) p++; } - video_enc->rc_override_count=i; + video_enc->rc_override_count = i; if (!video_enc->rc_initial_buffer_occupancy) - video_enc->rc_initial_buffer_occupancy = video_enc->rc_buffer_size*3/4; - video_enc->intra_dc_precision= intra_dc_precision - 8; + video_enc->rc_initial_buffer_occupancy = video_enc->rc_buffer_size * 3 / 4; + video_enc->intra_dc_precision = intra_dc_precision - 8; /* two pass mode */ if (do_pass) { @@ -3581,13 +3656,11 @@ static OutputStream *new_video_stream(OptionsContext *o, AVFormatContext *oc) ost->top_field_first = -1; MATCH_PER_STREAM_OPT(top_field_first, i, ost->top_field_first, oc, st); - MATCH_PER_STREAM_OPT(copy_initial_nonkeyframes, i, ost->copy_initial_nonkeyframes, oc ,st); - -#if CONFIG_AVFILTER MATCH_PER_STREAM_OPT(filters, str, filters, oc, st); if (filters) ost->avfilter = av_strdup(filters); -#endif + } else { + MATCH_PER_STREAM_OPT(copy_initial_nonkeyframes, i, ost->copy_initial_nonkeyframes, oc ,st); } return ost; @@ -3689,7 +3762,7 @@ static int copy_chapters(InputFile *ifile, OutputFile *ofile, int copy_metadata) for (i = 0; i < is->nb_chapters; i++) { AVChapter *in_ch = is->chapters[i], *out_ch; int64_t ts_off = av_rescale_q(ofile->start_time - ifile->ts_offset, - AV_TIME_BASE_Q, in_ch->time_base); + AV_TIME_BASE_Q, in_ch->time_base); int64_t rt = (ofile->recording_time == INT64_MAX) ? INT64_MAX : av_rescale_q(ofile->recording_time, AV_TIME_BASE_Q, in_ch->time_base); @@ -3712,7 +3785,7 @@ static int copy_chapters(InputFile *ifile, OutputFile *ofile, int copy_metadata) av_dict_copy(&out_ch->metadata, in_ch->metadata, 0); os->nb_chapters++; - os->chapters = av_realloc(os->chapters, sizeof(AVChapter)*os->nb_chapters); + os->chapters = av_realloc(os->chapters, sizeof(AVChapter) * os->nb_chapters); if (!os->chapters) return AVERROR(ENOMEM); os->chapters[os->nb_chapters - 1] = out_ch; @@ -3765,6 +3838,7 @@ static void opt_output_file(void *optctx, const char *filename) ost->source_index = index;\ ost->sync_ist = &input_streams[index];\ input_streams[index].discard = 0;\ + input_streams[index].st->discard = AVDISCARD_NONE;\ } /* video: highest resolution */ @@ -3825,9 +3899,10 @@ static void opt_output_file(void *optctx, const char *filename) } ost->source_index = input_files[map->file_index].ist_index + map->stream_index; - ost->sync_ist = &input_streams[input_files[map->sync_file_index].ist_index + + ost->sync_ist = &input_streams[input_files[map->sync_file_index].ist_index + map->sync_stream_index]; ist->discard = 0; + ist->st->discard = AVDISCARD_NONE; } } @@ -3871,6 +3946,8 @@ static void opt_output_file(void *optctx, const char *filename) output_files[nb_output_files - 1].ctx = oc; output_files[nb_output_files - 1].ost_index = nb_output_streams - oc->nb_streams; output_files[nb_output_files - 1].recording_time = o->recording_time; + if (o->recording_time != INT64_MAX) + oc->duration = o->recording_time; output_files[nb_output_files - 1].start_time = o->start_time; output_files[nb_output_files - 1].limit_filesize = o->limit_filesize; av_dict_copy(&output_files[nb_output_files - 1].opts, format_opts, 0); @@ -4091,36 +4168,36 @@ static void show_help(void) static int opt_target(OptionsContext *o, const char *opt, const char *arg) { enum { PAL, NTSC, FILM, UNKNOWN } norm = UNKNOWN; - static const char *const frame_rates[] = {"25", "30000/1001", "24000/1001"}; + static const char *const frame_rates[] = { "25", "30000/1001", "24000/1001" }; - if(!strncmp(arg, "pal-", 4)) { + if (!strncmp(arg, "pal-", 4)) { norm = PAL; arg += 4; - } else if(!strncmp(arg, "ntsc-", 5)) { + } else if (!strncmp(arg, "ntsc-", 5)) { norm = NTSC; arg += 5; - } else if(!strncmp(arg, "film-", 5)) { + } else if (!strncmp(arg, "film-", 5)) { norm = FILM; arg += 5; } else { /* Try to determine PAL/NTSC by peeking in the input files */ - if(nb_input_files) { + if (nb_input_files) { int i, j, fr; for (j = 0; j < nb_input_files; j++) { for (i = 0; i < input_files[j].nb_streams; i++) { AVCodecContext *c = input_files[j].ctx->streams[i]->codec; - if(c->codec_type != AVMEDIA_TYPE_VIDEO) + if (c->codec_type != AVMEDIA_TYPE_VIDEO) continue; fr = c->time_base.den * 1000 / c->time_base.num; - if(fr == 25000) { + if (fr == 25000) { norm = PAL; break; - } else if((fr == 29970) || (fr == 23976)) { + } else if ((fr == 29970) || (fr == 23976)) { norm = NTSC; break; } } - if(norm != UNKNOWN) + if (norm != UNKNOWN) break; } } @@ -4128,14 +4205,14 @@ static int opt_target(OptionsContext *o, const char *opt, const char *arg) av_log(NULL, AV_LOG_INFO, "Assuming %s for target.\n", norm == PAL ? "PAL" : "NTSC"); } - if(norm == UNKNOWN) { + if (norm == UNKNOWN) { av_log(NULL, AV_LOG_FATAL, "Could not determine norm (PAL/NTSC/NTSC-Film) for target.\n"); av_log(NULL, AV_LOG_FATAL, "Please prefix target with \"pal-\", \"ntsc-\" or \"film-\",\n"); av_log(NULL, AV_LOG_FATAL, "or set a framerate with \"-r xxx\".\n"); exit_program(1); } - if(!strcmp(arg, "vcd")) { + if (!strcmp(arg, "vcd")) { opt_video_codec(o, "c:v", "mpeg1video"); opt_audio_codec(o, "c:a", "mp2"); parse_option(o, "f", "vcd", options); @@ -4161,8 +4238,8 @@ static int opt_target(OptionsContext *o, const char *opt, const char *arg) and the first pack from the other stream, respectively, may also have been written before. So the real data starts at SCR 36000+3*1200. */ - o->mux_preload = (36000+3*1200) / 90000.0; //0.44 - } else if(!strcmp(arg, "svcd")) { + o->mux_preload = (36000 + 3 * 1200) / 90000.0; // 0.44 + } else if (!strcmp(arg, "svcd")) { opt_video_codec(o, "c:v", "mpeg2video"); opt_audio_codec(o, "c:a", "mp2"); @@ -4174,8 +4251,8 @@ static int opt_target(OptionsContext *o, const char *opt, const char *arg) opt_default("b", "2040000"); opt_default("maxrate", "2516000"); - opt_default("minrate", "0"); //1145000; - opt_default("bufsize", "1835008"); //224*1024*8; + opt_default("minrate", "0"); // 1145000; + opt_default("bufsize", "1835008"); // 224*1024*8; opt_default("flags", "+scan_offset"); @@ -4184,7 +4261,7 @@ static int opt_target(OptionsContext *o, const char *opt, const char *arg) opt_default("packetsize", "2324"); - } else if(!strcmp(arg, "dvd")) { + } else if (!strcmp(arg, "dvd")) { opt_video_codec(o, "c:v", "mpeg2video"); opt_audio_codec(o, "c:a", "ac3"); @@ -4196,8 +4273,8 @@ static int opt_target(OptionsContext *o, const char *opt, const char *arg) opt_default("b", "6000000"); opt_default("maxrate", "9000000"); - opt_default("minrate", "0"); //1500000; - opt_default("bufsize", "1835008"); //224*1024*8; + opt_default("minrate", "0"); // 1500000; + opt_default("bufsize", "1835008"); // 224*1024*8; opt_default("packetsize", "2048"); // from www.mpucoder.com: DVD sectors contain 2048 bytes of data, this is also the size of one pack. opt_default("muxrate", "10080000"); // from mplex project: data_rate = 1260000. mux_rate = data_rate * 8 @@ -4205,7 +4282,7 @@ static int opt_target(OptionsContext *o, const char *opt, const char *arg) opt_default("b:a", "448000"); parse_option(o, "ar", "48000", options); - } else if(!strncmp(arg, "dv", 2)) { + } else if (!strncmp(arg, "dv", 2)) { parse_option(o, "f", "dv", options); @@ -4227,7 +4304,7 @@ static int opt_target(OptionsContext *o, const char *opt, const char *arg) static int opt_vstats_file(const char *opt, const char *arg) { av_free (vstats_filename); - vstats_filename=av_strdup (arg); + vstats_filename = av_strdup (arg); return 0; } @@ -4277,6 +4354,120 @@ static int opt_video_filters(OptionsContext *o, const char *opt, const char *arg return parse_option(o, "filter:v", arg, options); } +static int opt_vsync(const char *opt, const char *arg) +{ + if (!av_strcasecmp(arg, "cfr")) video_sync_method = VSYNC_CFR; + else if (!av_strcasecmp(arg, "vfr")) video_sync_method = VSYNC_VFR; + else if (!av_strcasecmp(arg, "passthrough")) video_sync_method = VSYNC_PASSTHROUGH; + + if (video_sync_method == VSYNC_AUTO) + video_sync_method = parse_number_or_die("vsync", arg, OPT_INT, VSYNC_AUTO, VSYNC_VFR); + return 0; +} + +static int opt_deinterlace(const char *opt, const char *arg) +{ + av_log(NULL, AV_LOG_WARNING, "-%s is deprecated, use -filter:v yadif instead\n", opt); + do_deinterlace = 1; + return 0; +} + +static int opt_cpuflags(const char *opt, const char *arg) +{ +#define CPUFLAG_MMX2 (AV_CPU_FLAG_MMX | AV_CPU_FLAG_MMX2) +#define CPUFLAG_3DNOW (AV_CPU_FLAG_3DNOW | AV_CPU_FLAG_MMX) +#define CPUFLAG_3DNOWEXT (AV_CPU_FLAG_3DNOWEXT | CPUFLAG_3DNOW) +#define CPUFLAG_SSE (AV_CPU_FLAG_SSE | CPUFLAG_MMX2) +#define CPUFLAG_SSE2 (AV_CPU_FLAG_SSE2 | CPUFLAG_SSE) +#define CPUFLAG_SSE2SLOW (AV_CPU_FLAG_SSE2SLOW | CPUFLAG_SSE2) +#define CPUFLAG_SSE3 (AV_CPU_FLAG_SSE3 | CPUFLAG_SSE2) +#define CPUFLAG_SSE3SLOW (AV_CPU_FLAG_SSE3SLOW | CPUFLAG_SSE3) +#define CPUFLAG_SSSE3 (AV_CPU_FLAG_SSSE3 | CPUFLAG_SSE3) +#define CPUFLAG_SSE4 (AV_CPU_FLAG_SSE4 | CPUFLAG_SSSE3) +#define CPUFLAG_SSE42 (AV_CPU_FLAG_SSE42 | CPUFLAG_SSE4) +#define CPUFLAG_AVX (AV_CPU_FLAG_AVX | CPUFLAG_SSE42) +#define CPUFLAG_XOP (AV_CPU_FLAG_XOP | CPUFLAG_AVX) +#define CPUFLAG_FMA4 (AV_CPU_FLAG_FMA4 | CPUFLAG_AVX) + static const AVOption cpuflags_opts[] = { + { "flags" , NULL, 0, AV_OPT_TYPE_FLAGS, { 0 }, INT64_MIN, INT64_MAX, .unit = "flags" }, + { "altivec" , NULL, 0, AV_OPT_TYPE_CONST, { AV_CPU_FLAG_ALTIVEC }, .unit = "flags" }, + { "mmx" , NULL, 0, AV_OPT_TYPE_CONST, { AV_CPU_FLAG_MMX }, .unit = "flags" }, + { "mmx2" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_MMX2 }, .unit = "flags" }, + { "sse" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSE }, .unit = "flags" }, + { "sse2" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSE2 }, .unit = "flags" }, + { "sse2slow", NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSE2SLOW }, .unit = "flags" }, + { "sse3" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSE3 }, .unit = "flags" }, + { "sse3slow", NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSE3SLOW }, .unit = "flags" }, + { "ssse3" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSSE3 }, .unit = "flags" }, + { "atom" , NULL, 0, AV_OPT_TYPE_CONST, { AV_CPU_FLAG_ATOM }, .unit = "flags" }, + { "sse4.1" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSE4 }, .unit = "flags" }, + { "sse4.2" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_SSE42 }, .unit = "flags" }, + { "avx" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_AVX }, .unit = "flags" }, + { "xop" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_XOP }, .unit = "flags" }, + { "fma4" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_FMA4 }, .unit = "flags" }, + { "3dnow" , NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_3DNOW }, .unit = "flags" }, + { "3dnowext", NULL, 0, AV_OPT_TYPE_CONST, { CPUFLAG_3DNOWEXT }, .unit = "flags" }, + { NULL }, + }; + static const AVClass class = { + .class_name = "cpuflags", + .item_name = av_default_item_name, + .option = cpuflags_opts, + .version = LIBAVUTIL_VERSION_INT, + }; + + int flags = 0, ret; + const AVClass *pclass = &class; + + if ((ret = av_opt_eval_flags(&pclass, &cpuflags_opts[0], arg, &flags)) < 0) + return ret; + + av_set_cpu_flags_mask(flags); + return 0; +} + +static void parse_cpuflags(int argc, char **argv, const OptionDef *options) +{ + int idx = locate_option(argc, argv, options, "cpuflags"); + if (idx && argv[idx + 1]) + opt_cpuflags("cpuflags", argv[idx + 1]); +} + +static int opt_channel_layout(OptionsContext *o, const char *opt, const char *arg) +{ + char layout_str[32]; + char *stream_str; + char *ac_str; + int ret, channels, ac_str_size; + uint64_t layout; + + layout = av_get_channel_layout(arg); + if (!layout) { + av_log(NULL, AV_LOG_ERROR, "Unknown channel layout: %s\n", arg); + return AVERROR(EINVAL); + } + snprintf(layout_str, sizeof(layout_str), "%"PRIu64, layout); + ret = opt_default(opt, layout_str); + if (ret < 0) + return ret; + + /* set 'ac' option based on channel layout */ + channels = av_get_channel_layout_nb_channels(layout); + snprintf(layout_str, sizeof(layout_str), "%d", channels); + stream_str = strchr(opt, ':'); + ac_str_size = 3 + (stream_str ? strlen(stream_str) : 0); + ac_str = av_mallocz(ac_str_size); + if (!ac_str) + return AVERROR(ENOMEM); + av_strlcpy(ac_str, "ac", 3); + if (stream_str) + av_strlcat(ac_str, stream_str, ac_str_size); + ret = parse_option(o, ac_str, layout_str, options); + av_free(ac_str); + + return ret; +} + #define OFFSET(x) offsetof(OptionsContext, x) static const OptionDef options[] = { /* main options */ @@ -4307,7 +4498,7 @@ static const OptionDef options[] = { "when dumping packets, also dump the payload" }, { "re", OPT_BOOL | OPT_EXPERT | OPT_OFFSET, {.off = OFFSET(rate_emu)}, "read input at native frame rate", "" }, { "target", HAS_ARG | OPT_FUNC2, {(void*)opt_target}, "specify target file type (\"vcd\", \"svcd\", \"dvd\", \"dv\", \"dv50\", \"pal-vcd\", \"ntsc-svcd\", ...)", "type" }, - { "vsync", HAS_ARG | OPT_INT | OPT_EXPERT, {(void*)&video_sync_method}, "video sync method", "" }, + { "vsync", HAS_ARG | OPT_EXPERT, {(void*)opt_vsync}, "video sync method", "" }, { "async", HAS_ARG | OPT_INT | OPT_EXPERT, {(void*)&audio_sync_method}, "audio sync method", "" }, { "adrift_threshold", HAS_ARG | OPT_FLOAT | OPT_EXPERT, {(void*)&audio_drift_threshold}, "audio drift threshold", "threshold" }, { "copyts", OPT_BOOL | OPT_EXPERT, {(void*)©_ts}, "copy timestamps" }, @@ -4320,12 +4511,11 @@ static const OptionDef options[] = { { "tag", OPT_STRING | HAS_ARG | OPT_SPEC, {.off = OFFSET(codec_tags)}, "force codec tag/fourcc", "fourcc/tag" }, { "q", HAS_ARG | OPT_EXPERT | OPT_DOUBLE | OPT_SPEC, {.off = OFFSET(qscale)}, "use fixed quality scale (VBR)", "q" }, { "qscale", HAS_ARG | OPT_EXPERT | OPT_DOUBLE | OPT_SPEC, {.off = OFFSET(qscale)}, "use fixed quality scale (VBR)", "q" }, -#if CONFIG_AVFILTER { "filter", HAS_ARG | OPT_STRING | OPT_SPEC, {.off = OFFSET(filters)}, "set stream filterchain", "filter_list" }, -#endif { "stats", OPT_BOOL, {&print_stats}, "print progress report during encoding", }, { "attach", HAS_ARG | OPT_FUNC2, {(void*)opt_attach}, "add an attachment to the output file", "filename" }, { "dump_attachment", HAS_ARG | OPT_STRING | OPT_SPEC, {.off = OFFSET(dump_attachment)}, "extract an attachment into a file", "filename" }, + { "cpuflags", HAS_ARG | OPT_EXPERT, {(void*)opt_cpuflags}, "set CPU flags mask", "mask" }, /* video options */ { "vframes", HAS_ARG | OPT_VIDEO | OPT_FUNC2, {(void*)opt_video_frames}, "set the number of video frames to record", "number" }, @@ -4341,13 +4531,11 @@ static const OptionDef options[] = { "use same quantizer as source (implies VBR)" }, { "pass", HAS_ARG | OPT_VIDEO, {(void*)opt_pass}, "select the pass number (1 or 2)", "n" }, { "passlogfile", HAS_ARG | OPT_STRING | OPT_VIDEO, {(void*)&pass_logfilename_prefix}, "select two pass log file name prefix", "prefix" }, - { "deinterlace", OPT_BOOL | OPT_EXPERT | OPT_VIDEO, {(void*)&do_deinterlace}, - "deinterlace pictures" }, + { "deinterlace", OPT_EXPERT | OPT_VIDEO, {(void*)opt_deinterlace}, + "this option is deprecated, use the yadif filter instead" }, { "vstats", OPT_EXPERT | OPT_VIDEO, {(void*)&opt_vstats}, "dump video coding statistics to file" }, { "vstats_file", HAS_ARG | OPT_EXPERT | OPT_VIDEO, {(void*)opt_vstats_file}, "dump video coding statistics to file", "file" }, -#if CONFIG_AVFILTER { "vf", HAS_ARG | OPT_VIDEO | OPT_FUNC2, {(void*)opt_video_filters}, "video filters", "filter list" }, -#endif { "intra_matrix", HAS_ARG | OPT_EXPERT | OPT_VIDEO | OPT_STRING | OPT_SPEC, {.off = OFFSET(intra_matrices)}, "specify intra matrix coeffs", "matrix" }, { "inter_matrix", HAS_ARG | OPT_EXPERT | OPT_VIDEO | OPT_STRING | OPT_SPEC, {.off = OFFSET(inter_matrices)}, "specify inter matrix coeffs", "matrix" }, { "top", HAS_ARG | OPT_EXPERT | OPT_VIDEO | OPT_INT| OPT_SPEC, {.off = OFFSET(top_field_first)}, "top=1/bottom=0/auto=-1 field first", "" }, @@ -4368,6 +4556,7 @@ static const OptionDef options[] = { { "atag", HAS_ARG | OPT_EXPERT | OPT_AUDIO | OPT_FUNC2, {(void*)opt_audio_tag}, "force audio tag/fourcc", "fourcc/tag" }, { "vol", OPT_INT | HAS_ARG | OPT_AUDIO, {(void*)&audio_volume}, "change audio volume (256=normal)" , "volume" }, // { "sample_fmt", HAS_ARG | OPT_EXPERT | OPT_AUDIO | OPT_SPEC | OPT_STRING, {.off = OFFSET(sample_fmts)}, "set sample format", "format" }, + { "channel_layout", HAS_ARG | OPT_EXPERT | OPT_AUDIO | OPT_FUNC2, {(void*)opt_channel_layout}, "set channel layout", "layout" }, /* subtitle options */ { "sn", OPT_BOOL | OPT_SUBTITLE | OPT_OFFSET, {.off = OFFSET(subtitle_disable)}, "disable subtitle" }, @@ -4404,18 +4593,18 @@ int main(int argc, char **argv) #if CONFIG_AVDEVICE avdevice_register_all(); #endif -#if CONFIG_AVFILTER avfilter_register_all(); -#endif av_register_all(); avformat_network_init(); show_banner(); + parse_cpuflags(argc, argv, options); + /* parse options */ parse_options(&o, argc, argv, options, opt_output_file); - if(nb_output_files <= 0 && nb_input_files == 0) { + if (nb_output_files <= 0 && nb_input_files == 0) { show_usage(); av_log(NULL, AV_LOG_WARNING, "Use -h to get full help or, even better, run 'man %s'\n", program_name); exit_program(1);