X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=avconv.c;h=275c3eb2684dca00f0450a75fa530ad70a0d446a;hb=9fb483fede7207c1f3965e5ae1ac0ee82e4d7a4e;hp=d46c8e0943d2f33615986d0ea087ec066ea770eb;hpb=f154ef1ae5b03f288dd8c025dab1884b4cb20c1a;p=ffmpeg diff --git a/avconv.c b/avconv.c index d46c8e0943d..275c3eb2684 100644 --- a/avconv.c +++ b/avconv.c @@ -27,15 +27,16 @@ #include #include #include +#include + #include "libavformat/avformat.h" #include "libavdevice/avdevice.h" #include "libswscale/swscale.h" #include "libavresample/avresample.h" #include "libavutil/opt.h" -#include "libavutil/audioconvert.h" +#include "libavutil/channel_layout.h" #include "libavutil/parseutils.h" #include "libavutil/samplefmt.h" -#include "libavutil/colorspace.h" #include "libavutil/fifo.h" #include "libavutil/intreadwrite.h" #include "libavutil/dict.h" @@ -48,11 +49,11 @@ #include "libavformat/os_support.h" # include "libavfilter/avfilter.h" -# include "libavfilter/avfiltergraph.h" # include "libavfilter/buffersrc.h" # include "libavfilter/buffersink.h" #if HAVE_SYS_RESOURCE_H +#include #include #include #elif HAVE_GETPROCESSTIMES @@ -83,10 +84,6 @@ const int program_birth_year = 2000; static FILE *vstats_file; -static int64_t video_size = 0; -static int64_t audio_size = 0; -static int64_t extra_size = 0; -static int nb_frames_dup = 0; static int nb_frames_drop = 0; @@ -143,47 +140,59 @@ static int decode_interrupt_cb(void *ctx) const AVIOInterruptCB int_cb = { decode_interrupt_cb, NULL }; -void exit_program(int ret) +static void avconv_cleanup(int ret) { int i, j; for (i = 0; i < nb_filtergraphs; i++) { - avfilter_graph_free(&filtergraphs[i]->graph); - for (j = 0; j < filtergraphs[i]->nb_inputs; j++) { - av_freep(&filtergraphs[i]->inputs[j]->name); - av_freep(&filtergraphs[i]->inputs[j]); + FilterGraph *fg = filtergraphs[i]; + avfilter_graph_free(&fg->graph); + for (j = 0; j < fg->nb_inputs; j++) { + av_freep(&fg->inputs[j]->name); + av_freep(&fg->inputs[j]); } - av_freep(&filtergraphs[i]->inputs); - for (j = 0; j < filtergraphs[i]->nb_outputs; j++) { - av_freep(&filtergraphs[i]->outputs[j]->name); - av_freep(&filtergraphs[i]->outputs[j]); + av_freep(&fg->inputs); + for (j = 0; j < fg->nb_outputs; j++) { + av_freep(&fg->outputs[j]->name); + av_freep(&fg->outputs[j]); } - av_freep(&filtergraphs[i]->outputs); + av_freep(&fg->outputs); + av_freep(&fg->graph_desc); + av_freep(&filtergraphs[i]); } av_freep(&filtergraphs); /* close files */ for (i = 0; i < nb_output_files; i++) { - AVFormatContext *s = output_files[i]->ctx; - if (!(s->oformat->flags & AVFMT_NOFILE) && s->pb) + OutputFile *of = output_files[i]; + AVFormatContext *s = of->ctx; + if (s && s->oformat && !(s->oformat->flags & AVFMT_NOFILE) && s->pb) avio_close(s->pb); avformat_free_context(s); - av_dict_free(&output_files[i]->opts); + av_dict_free(&of->opts); + av_freep(&output_files[i]); } for (i = 0; i < nb_output_streams; i++) { - AVBitStreamFilterContext *bsfc = output_streams[i]->bitstream_filters; + OutputStream *ost = output_streams[i]; + AVBitStreamFilterContext *bsfc = ost->bitstream_filters; while (bsfc) { AVBitStreamFilterContext *next = bsfc->next; av_bitstream_filter_close(bsfc); bsfc = next; } - output_streams[i]->bitstream_filters = NULL; + ost->bitstream_filters = NULL; + av_frame_free(&ost->filtered_frame); + + av_parser_close(ost->parser); + + av_freep(&ost->forced_keyframes); + av_freep(&ost->avfilter); + av_freep(&ost->logfile_prefix); + + avcodec_free_context(&ost->enc_ctx); - av_freep(&output_streams[i]->forced_keyframes); - av_freep(&output_streams[i]->avfilter); - av_freep(&output_streams[i]->filtered_frame); av_freep(&output_streams[i]); } for (i = 0; i < nb_input_files; i++) { @@ -191,10 +200,16 @@ void exit_program(int ret) av_freep(&input_files[i]); } for (i = 0; i < nb_input_streams; i++) { - av_freep(&input_streams[i]->decoded_frame); - av_dict_free(&input_streams[i]->opts); - free_buffer_pool(&input_streams[i]->buffer_pool); - av_freep(&input_streams[i]->filters); + InputStream *ist = input_streams[i]; + + av_frame_free(&ist->decoded_frame); + av_frame_free(&ist->filter_frame); + av_dict_free(&ist->decoder_opts); + av_freep(&ist->filters); + av_freep(&ist->hwaccel_device); + + avcodec_free_context(&ist->dec_ctx); + av_freep(&input_streams[i]); } @@ -209,7 +224,6 @@ void exit_program(int ret) uninit_opts(); - avfilter_uninit(); avformat_network_deinit(); if (received_sigterm) { @@ -217,8 +231,6 @@ void exit_program(int ret) (int) received_sigterm); exit (255); } - - exit(ret); } void assert_avoptions(AVDictionary *m) @@ -230,24 +242,21 @@ void assert_avoptions(AVDictionary *m) } } -static void assert_codec_experimental(AVCodecContext *c, int encoder) +static void abort_codec_experimental(AVCodec *c, int encoder) { const char *codec_string = encoder ? "encoder" : "decoder"; AVCodec *codec; - if (c->codec->capabilities & CODEC_CAP_EXPERIMENTAL && - c->strict_std_compliance > FF_COMPLIANCE_EXPERIMENTAL) { - av_log(NULL, AV_LOG_FATAL, "%s '%s' is experimental and might produce bad " - "results.\nAdd '-strict experimental' if you want to use it.\n", - codec_string, c->codec->name); - codec = encoder ? avcodec_find_encoder(c->codec->id) : avcodec_find_decoder(c->codec->id); - if (!(codec->capabilities & CODEC_CAP_EXPERIMENTAL)) - av_log(NULL, AV_LOG_FATAL, "Or use the non experimental %s '%s'.\n", - codec_string, codec->name); - exit_program(1); - } + av_log(NULL, AV_LOG_FATAL, "%s '%s' is experimental and might produce bad " + "results.\nAdd '-strict experimental' if you want to use it.\n", + codec_string, c->name); + codec = encoder ? avcodec_find_encoder(c->id) : avcodec_find_decoder(c->id); + if (!(codec->capabilities & CODEC_CAP_EXPERIMENTAL)) + av_log(NULL, AV_LOG_FATAL, "Or use the non experimental %s '%s'.\n", + codec_string, codec->name); + exit_program(1); } -/** +/* * Update the requested input sample format based on the output sample format. * This is currently only used to request float output from decoders which * support multiple sample formats, one of which is AV_SAMPLE_FMT_FLT. @@ -268,29 +277,48 @@ static void update_sample_fmt(AVCodecContext *dec, AVCodec *dec_codec, dec_codec->sample_fmts[0] != AV_SAMPLE_FMT_NONE && dec_codec->sample_fmts[1] != AV_SAMPLE_FMT_NONE) { const enum AVSampleFormat *p; - int min_dec = -1, min_inc = -1; + int min_dec = INT_MAX, min_inc = INT_MAX; + enum AVSampleFormat dec_fmt = AV_SAMPLE_FMT_NONE; + enum AVSampleFormat inc_fmt = AV_SAMPLE_FMT_NONE; /* find a matching sample format in the encoder */ for (p = dec_codec->sample_fmts; *p != AV_SAMPLE_FMT_NONE; p++) { if (*p == enc->sample_fmt) { dec->request_sample_fmt = *p; return; - } else if (*p > enc->sample_fmt) { - min_inc = FFMIN(min_inc, *p - enc->sample_fmt); - } else - min_dec = FFMIN(min_dec, enc->sample_fmt - *p); + } else { + enum AVSampleFormat dfmt = av_get_packed_sample_fmt(*p); + enum AVSampleFormat efmt = av_get_packed_sample_fmt(enc->sample_fmt); + int fmt_diff = 32 * abs(dfmt - efmt); + if (av_sample_fmt_is_planar(*p) != + av_sample_fmt_is_planar(enc->sample_fmt)) + fmt_diff++; + if (dfmt == efmt) { + min_inc = fmt_diff; + inc_fmt = *p; + } else if (dfmt > efmt) { + if (fmt_diff < min_inc) { + min_inc = fmt_diff; + inc_fmt = *p; + } + } else { + if (fmt_diff < min_dec) { + min_dec = fmt_diff; + dec_fmt = *p; + } + } + } } /* if none match, provide the one that matches quality closest */ - dec->request_sample_fmt = min_inc > 0 ? enc->sample_fmt + min_inc : - enc->sample_fmt - min_dec; + dec->request_sample_fmt = min_inc != INT_MAX ? inc_fmt : dec_fmt; } } static void write_frame(AVFormatContext *s, AVPacket *pkt, OutputStream *ost) { AVBitStreamFilterContext *bsfc = ost->bitstream_filters; - AVCodecContext *avctx = ost->st->codec; + AVCodecContext *avctx = ost->encoding_needed ? ost->enc_ctx : ost->st->codec; int ret; /* @@ -316,7 +344,10 @@ static void write_frame(AVFormatContext *s, AVPacket *pkt, OutputStream *ost) pkt->flags & AV_PKT_FLAG_KEY); if (a > 0) { av_free_packet(pkt); - new_pkt.destruct = av_destruct_packet; + new_pkt.buf = av_buffer_create(new_pkt.data, new_pkt.size, + av_buffer_default_free, NULL, 0); + if (!new_pkt.buf) + exit_program(1); } else if (a < 0) { av_log(NULL, AV_LOG_ERROR, "%s failed for stream %d, codec %s", bsfc->filter->name, pkt->stream_index, @@ -330,6 +361,28 @@ static void write_frame(AVFormatContext *s, AVPacket *pkt, OutputStream *ost) bsfc = bsfc->next; } + if (!(s->oformat->flags & AVFMT_NOTIMESTAMPS) && + ost->last_mux_dts != AV_NOPTS_VALUE && + pkt->dts < ost->last_mux_dts + !(s->oformat->flags & AVFMT_TS_NONSTRICT)) { + av_log(NULL, AV_LOG_WARNING, "Non-monotonous DTS in output stream " + "%d:%d; previous: %"PRId64", current: %"PRId64"; ", + ost->file_index, ost->st->index, ost->last_mux_dts, pkt->dts); + if (exit_on_error) { + av_log(NULL, AV_LOG_FATAL, "aborting.\n"); + exit_program(1); + } + av_log(NULL, AV_LOG_WARNING, "changing to %"PRId64". This may result " + "in incorrect timestamps in the output file.\n", + ost->last_mux_dts + 1); + pkt->dts = ost->last_mux_dts + 1; + if (pkt->pts != AV_NOPTS_VALUE) + pkt->pts = FFMAX(pkt->pts, pkt->dts); + } + ost->last_mux_dts = pkt->dts; + + ost->data_size += pkt->size; + ost->packets_written++; + pkt->stream_index = ost->index; ret = av_interleaved_write_frame(s, pkt); if (ret < 0) { @@ -343,9 +396,9 @@ 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_compare_ts(ost->sync_opts - ost->first_pts, ost->enc_ctx->time_base, of->recording_time, AV_TIME_BASE_Q) >= 0) { - ost->is_past_recording_time = 1; + ost->finished = 1; return 0; } return 1; @@ -354,7 +407,7 @@ static int check_recording_time(OutputStream *ost) static void do_audio_out(AVFormatContext *s, OutputStream *ost, AVFrame *frame) { - AVCodecContext *enc = ost->st->codec; + AVCodecContext *enc = ost->enc_ctx; AVPacket pkt; int got_packet = 0; @@ -362,71 +415,24 @@ static void do_audio_out(AVFormatContext *s, OutputStream *ost, pkt.data = NULL; pkt.size = 0; - if (!check_recording_time(ost)) - return; - if (frame->pts == AV_NOPTS_VALUE || audio_sync_method < 0) frame->pts = ost->sync_opts; ost->sync_opts = frame->pts + frame->nb_samples; + ost->samples_encoded += frame->nb_samples; + ost->frames_encoded++; + 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); - + av_packet_rescale_ts(&pkt, enc->time_base, ost->st->time_base); write_frame(s, &pkt, ost); - - audio_size += pkt.size; } } -static void pre_process_video_frame(InputStream *ist, AVPicture *picture, void **bufp) -{ - AVCodecContext *dec; - AVPicture *picture2; - AVPicture picture_tmp; - uint8_t *buf = 0; - - dec = ist->st->codec; - - /* deinterlace : must be done before any resize */ - if (do_deinterlace) { - int size; - - /* create temporary picture */ - size = avpicture_get_size(dec->pix_fmt, dec->width, dec->height); - 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, - dec->pix_fmt, dec->width, dec->height) < 0) { - /* if error, do not deinterlace */ - av_log(NULL, AV_LOG_WARNING, "Deinterlacing failed\n"); - av_free(buf); - buf = NULL; - picture2 = picture; - } - } else { - picture2 = picture; - } - - if (picture != picture2) - *picture = *picture2; - *bufp = buf; -} - static void do_subtitle_out(AVFormatContext *s, OutputStream *ost, InputStream *ist, @@ -446,7 +452,7 @@ static void do_subtitle_out(AVFormatContext *s, return; } - enc = ost->st->codec; + enc = ost->enc_ctx; if (!subtitle_out) { subtitle_out = av_malloc(subtitle_out_max_size); @@ -470,6 +476,9 @@ static void do_subtitle_out(AVFormatContext *s, 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; + + ost->frames_encoded++; + subtitle_out_size = avcodec_encode_subtitle(enc, subtitle_out, subtitle_out_max_size, sub); if (subtitle_out_size < 0) { @@ -496,11 +505,11 @@ static void do_subtitle_out(AVFormatContext *s, static void do_video_out(AVFormatContext *s, OutputStream *ost, AVFrame *in_picture, - int *frame_size, float quality) + int *frame_size) { int ret, format_video_sync; AVPacket pkt; - AVCodecContext *enc = ost->st->codec; + AVCodecContext *enc = ost->enc_ctx; *frame_size = 0; @@ -513,7 +522,9 @@ static void do_video_out(AVFormatContext *s, in_picture->pts != AV_NOPTS_VALUE && in_picture->pts < ost->sync_opts) { nb_frames_drop++; - av_log(NULL, AV_LOG_VERBOSE, "*** drop!\n"); + av_log(NULL, AV_LOG_WARNING, + "*** dropping frame %d from stream %d at ts %"PRId64"\n", + ost->frame_number, ost->st->index, in_picture->pts); return; } @@ -529,8 +540,7 @@ static void do_video_out(AVFormatContext *s, pkt.data = NULL; pkt.size = 0; - if (!check_recording_time(ost) || - ost->frame_number >= ost->max_frames) + if (ost->frame_number >= ost->max_frames) return; if (s->oformat->flags & AVFMT_RAWPICTURE && @@ -548,44 +558,31 @@ static void do_video_out(AVFormatContext *s, write_frame(s, &pkt, ost); } else { int got_packet; - AVFrame big_picture; - - big_picture = *in_picture; - /* better than nothing: use input picture interlaced - settings */ - big_picture.interlaced_frame = in_picture->interlaced_frame; - if (ost->st->codec->flags & (CODEC_FLAG_INTERLACED_DCT|CODEC_FLAG_INTERLACED_ME)) { - if (ost->top_field_first == -1) - big_picture.top_field_first = in_picture->top_field_first; - else - big_picture.top_field_first = !!ost->top_field_first; - } - /* handles same_quant here. This is not correct because it may - not be a global option */ - big_picture.quality = quality; - if (!enc->me_threshold) - big_picture.pict_type = 0; + if (enc->flags & (CODEC_FLAG_INTERLACED_DCT|CODEC_FLAG_INTERLACED_ME) && + ost->top_field_first >= 0) + in_picture->top_field_first = !!ost->top_field_first; + + in_picture->quality = enc->global_quality; + in_picture->pict_type = 0; 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; + in_picture->pts >= ost->forced_kf_pts[ost->forced_kf_index]) { + in_picture->pict_type = AV_PICTURE_TYPE_I; ost->forced_kf_index++; } - ret = avcodec_encode_video2(enc, &pkt, &big_picture, &got_packet); + + ost->frames_encoded++; + + ret = avcodec_encode_video2(enc, &pkt, in_picture, &got_packet); if (ret < 0) { av_log(NULL, AV_LOG_FATAL, "Video 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); - + av_packet_rescale_ts(&pkt, 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) { @@ -607,8 +604,7 @@ static double psnr(double d) return -10.0 * log(d) / log(10.0); } -static void do_video_stats(AVFormatContext *os, OutputStream *ost, - int frame_size) +static void do_video_stats(OutputStream *ost, int frame_size) { AVCodecContext *enc; int frame_number; @@ -623,7 +619,7 @@ static void do_video_stats(AVFormatContext *os, OutputStream *ost, } } - enc = ost->st->codec; + enc = ost->enc_ctx; 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); @@ -637,64 +633,55 @@ static void do_video_stats(AVFormatContext *os, OutputStream *ost, ti1 = 0.01; bitrate = (frame_size * 8) / av_q2d(enc->time_base) / 1000.0; - avg_bitrate = (double)(video_size * 8) / ti1 / 1000.0; + avg_bitrate = (double)(ost->data_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)ost->data_size / 1024, ti1, bitrate, avg_bitrate); fprintf(vstats_file, "type= %c\n", av_get_picture_type_char(enc->coded_frame->pict_type)); } } -/** +/* * Read one frame for lavfi output for ost and encode it. */ static int poll_filter(OutputStream *ost) { OutputFile *of = output_files[ost->file_index]; - AVFilterBufferRef *picref; AVFrame *filtered_frame = NULL; int frame_size, ret; - if (!ost->filtered_frame && !(ost->filtered_frame = avcodec_alloc_frame())) { + if (!ost->filtered_frame && !(ost->filtered_frame = av_frame_alloc())) { return AVERROR(ENOMEM); - } else - avcodec_get_frame_defaults(ost->filtered_frame); + } filtered_frame = ost->filtered_frame; if (ost->enc->type == AVMEDIA_TYPE_AUDIO && !(ost->enc->capabilities & CODEC_CAP_VARIABLE_FRAME_SIZE)) - ret = av_buffersink_read_samples(ost->filter->filter, &picref, - ost->st->codec->frame_size); + ret = av_buffersink_get_samples(ost->filter->filter, filtered_frame, + ost->enc_ctx->frame_size); else - ret = av_buffersink_read(ost->filter->filter, &picref); + ret = av_buffersink_get_frame(ost->filter->filter, filtered_frame); if (ret < 0) return ret; - avfilter_copy_buf_props(filtered_frame, picref); - if (picref->pts != AV_NOPTS_VALUE) { - filtered_frame->pts = av_rescale_q(picref->pts, + if (filtered_frame->pts != AV_NOPTS_VALUE) { + int64_t start_time = (of->start_time == AV_NOPTS_VALUE) ? 0 : of->start_time; + filtered_frame->pts = av_rescale_q(filtered_frame->pts, ost->filter->filter->inputs[0]->time_base, - ost->st->codec->time_base) - - av_rescale_q(of->start_time, + ost->enc_ctx->time_base) - + av_rescale_q(start_time, AV_TIME_BASE_Q, - ost->st->codec->time_base); - - if (of->start_time && filtered_frame->pts < 0) { - avfilter_unref_buffer(picref); - return 0; - } + ost->enc_ctx->time_base); } switch (ost->filter->filter->inputs[0]->type) { case AVMEDIA_TYPE_VIDEO: if (!ost->frame_aspect_ratio) - ost->st->codec->sample_aspect_ratio = picref->video->pixel_aspect; + ost->enc_ctx->sample_aspect_ratio = filtered_frame->sample_aspect_ratio; - do_video_out(of->ctx, ost, filtered_frame, &frame_size, - same_quant ? ost->last_quality : - ost->st->codec->global_quality); + do_video_out(of->ctx, ost, filtered_frame, &frame_size); if (vstats_filename && frame_size) - do_video_stats(of->ctx, ost, frame_size); + do_video_stats(ost, frame_size); break; case AVMEDIA_TYPE_AUDIO: do_audio_out(of->ctx, ost, filtered_frame); @@ -704,12 +691,25 @@ static int poll_filter(OutputStream *ost) av_assert0(0); } - avfilter_unref_buffer(picref); + av_frame_unref(filtered_frame); return 0; } -/** +static void finish_output_stream(OutputStream *ost) +{ + OutputFile *of = output_files[ost->file_index]; + int i; + + ost->finished = 1; + + if (of->shortest) { + for (i = 0; i < of->ctx->nb_streams; i++) + output_streams[of->ost_index + i]->finished = 1; + } +} + +/* * Read as many frames from possible from lavfi and encode them. * * Always read from the active stream with the lowest timestamp. If no frames @@ -729,11 +729,10 @@ static int poll_filters(void) for (i = 0; i < nb_output_streams; i++) { int64_t pts = output_streams[i]->sync_opts; - if (!output_streams[i]->filter || - output_streams[i]->is_past_recording_time) + if (!output_streams[i]->filter || output_streams[i]->finished) continue; - pts = av_rescale_q(pts, output_streams[i]->st->codec->time_base, + pts = av_rescale_q(pts, output_streams[i]->enc_ctx->time_base, AV_TIME_BASE_Q); if (pts < min_pts) { min_pts = pts; @@ -747,11 +746,7 @@ static int poll_filters(void) ret = poll_filter(ost); if (ret == AVERROR_EOF) { - ost->is_past_recording_time = 1; - - if (opt_shortest) - return ret; - + finish_output_stream(ost); ret = 0; } else if (ret == AVERROR(EAGAIN)) return 0; @@ -760,6 +755,109 @@ static int poll_filters(void) return ret; } +static void print_final_stats(int64_t total_size) +{ + uint64_t video_size = 0, audio_size = 0, extra_size = 0, other_size = 0; + uint64_t data_size = 0; + float percent = -1.0; + int i, j; + + for (i = 0; i < nb_output_streams; i++) { + OutputStream *ost = output_streams[i]; + switch (ost->enc_ctx->codec_type) { + case AVMEDIA_TYPE_VIDEO: video_size += ost->data_size; break; + case AVMEDIA_TYPE_AUDIO: audio_size += ost->data_size; break; + default: other_size += ost->data_size; break; + } + extra_size += ost->enc_ctx->extradata_size; + data_size += ost->data_size; + } + + if (data_size && total_size >= data_size) + percent = 100.0 * (total_size - data_size) / data_size; + + av_log(NULL, AV_LOG_INFO, "\n"); + av_log(NULL, AV_LOG_INFO, "video:%1.0fkB audio:%1.0fkB other streams:%1.0fkB global headers:%1.0fkB muxing overhead: ", + video_size / 1024.0, + audio_size / 1024.0, + other_size / 1024.0, + extra_size / 1024.0); + if (percent >= 0.0) + av_log(NULL, AV_LOG_INFO, "%f%%", percent); + else + av_log(NULL, AV_LOG_INFO, "unknown"); + av_log(NULL, AV_LOG_INFO, "\n"); + + /* print verbose per-stream stats */ + for (i = 0; i < nb_input_files; i++) { + InputFile *f = input_files[i]; + uint64_t total_packets = 0, total_size = 0; + + av_log(NULL, AV_LOG_VERBOSE, "Input file #%d (%s):\n", + i, f->ctx->filename); + + for (j = 0; j < f->nb_streams; j++) { + InputStream *ist = input_streams[f->ist_index + j]; + enum AVMediaType type = ist->dec_ctx->codec_type; + + total_size += ist->data_size; + total_packets += ist->nb_packets; + + av_log(NULL, AV_LOG_VERBOSE, " Input stream #%d:%d (%s): ", + i, j, media_type_string(type)); + av_log(NULL, AV_LOG_VERBOSE, "%"PRIu64" packets read (%"PRIu64" bytes); ", + ist->nb_packets, ist->data_size); + + if (ist->decoding_needed) { + av_log(NULL, AV_LOG_VERBOSE, "%"PRIu64" frames decoded", + ist->frames_decoded); + if (type == AVMEDIA_TYPE_AUDIO) + av_log(NULL, AV_LOG_VERBOSE, " (%"PRIu64" samples)", ist->samples_decoded); + av_log(NULL, AV_LOG_VERBOSE, "; "); + } + + av_log(NULL, AV_LOG_VERBOSE, "\n"); + } + + av_log(NULL, AV_LOG_VERBOSE, " Total: %"PRIu64" packets (%"PRIu64" bytes) demuxed\n", + total_packets, total_size); + } + + for (i = 0; i < nb_output_files; i++) { + OutputFile *of = output_files[i]; + uint64_t total_packets = 0, total_size = 0; + + av_log(NULL, AV_LOG_VERBOSE, "Output file #%d (%s):\n", + i, of->ctx->filename); + + for (j = 0; j < of->ctx->nb_streams; j++) { + OutputStream *ost = output_streams[of->ost_index + j]; + enum AVMediaType type = ost->enc_ctx->codec_type; + + total_size += ost->data_size; + total_packets += ost->packets_written; + + av_log(NULL, AV_LOG_VERBOSE, " Output stream #%d:%d (%s): ", + i, j, media_type_string(type)); + if (ost->encoding_needed) { + av_log(NULL, AV_LOG_VERBOSE, "%"PRIu64" frames encoded", + ost->frames_encoded); + if (type == AVMEDIA_TYPE_AUDIO) + av_log(NULL, AV_LOG_VERBOSE, " (%"PRIu64" samples)", ost->samples_encoded); + av_log(NULL, AV_LOG_VERBOSE, "; "); + } + + av_log(NULL, AV_LOG_VERBOSE, "%"PRIu64" packets muxed (%"PRIu64" bytes); ", + ost->packets_written, ost->data_size); + + av_log(NULL, AV_LOG_VERBOSE, "\n"); + } + + av_log(NULL, AV_LOG_VERBOSE, " Total: %"PRIu64" packets (%"PRIu64" bytes) muxed\n", + total_packets, total_size); + } +} + static void print_report(int is_last_report, int64_t timer_start) { char buf[1024]; @@ -778,7 +876,7 @@ static void print_report(int is_last_report, int64_t timer_start) if (!is_last_report) { int64_t cur_time; /* display the report every 0.5 seconds */ - cur_time = av_gettime(); + cur_time = av_gettime_relative(); if (last_time == -1) { last_time = cur_time; return; @@ -792,8 +890,15 @@ static void print_report(int is_last_report, int64_t timer_start) 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 + 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) { + char errbuf[128]; + av_strerror(total_size, errbuf, sizeof(errbuf)); + av_log(NULL, AV_LOG_VERBOSE, "Bitrate not available, " + "avio_tell() failed: %s\n", errbuf); + total_size = 0; + } buf[0] = '\0'; ti1 = 1e10; @@ -801,14 +906,14 @@ static void print_report(int is_last_report, int64_t timer_start) for (i = 0; i < nb_output_streams; i++) { float q = -1; ost = output_streams[i]; - enc = ost->st->codec; + enc = ost->enc_ctx; if (!ost->stream_copy && enc->coded_frame) 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_relative() - timer_start) / 1000000.0; frame_number = ost->frame_number; snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "frame=%5d fps=%3d q=%3.1f ", @@ -821,7 +926,7 @@ static void print_report(int is_last_report, int64_t timer_start) if (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))); + snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "%X", (int)lrintf(log2(qp_histogram[j] + 1))); } if (enc->flags&CODEC_FLAG_PSNR) { int j; @@ -848,7 +953,7 @@ static void print_report(int is_last_report, int64_t timer_start) vid = 1; } /* compute min output value */ - pts = (double)ost->st->pts.val * av_q2d(ost->st->time_base); + pts = (double)ost->last_mux_dts * av_q2d(ost->st->time_base); if ((pts < ti1) && (pts > 0)) ti1 = pts; } @@ -861,24 +966,17 @@ static void print_report(int is_last_report, int64_t timer_start) "size=%8.0fkB time=%0.2f bitrate=%6.1fkbits/s", (double)total_size / 1024, ti1, bitrate); - if (nb_frames_dup || nb_frames_drop) - snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), " dup=%d drop=%d", - nb_frames_dup, nb_frames_drop); + if (nb_frames_drop) + snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), " drop=%d", + nb_frames_drop); av_log(NULL, AV_LOG_INFO, "%s \r", buf); fflush(stderr); - if (is_last_report) { - 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 - ); - } + if (is_last_report) + print_final_stats(total_size); + } static void flush_encoders(void) @@ -887,33 +985,30 @@ static void flush_encoders(void) for (i = 0; i < nb_output_streams; i++) { OutputStream *ost = output_streams[i]; - AVCodecContext *enc = ost->st->codec; + AVCodecContext *enc = ost->enc_ctx; 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 (enc->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 == AV_CODEC_ID_RAWVIDEO) + if (enc->codec_type == AVMEDIA_TYPE_VIDEO && (os->oformat->flags & AVFMT_RAWPICTURE) && enc->codec->id == AV_CODEC_ID_RAWVIDEO) continue; for (;;) { int (*encode)(AVCodecContext*, AVPacket*, const AVFrame*, int*) = NULL; const char *desc; - int64_t *size; - switch (ost->st->codec->codec_type) { + switch (enc->codec_type) { case AVMEDIA_TYPE_AUDIO: encode = avcodec_encode_audio2; desc = "Audio"; - size = &audio_size; break; case AVMEDIA_TYPE_VIDEO: encode = avcodec_encode_video2; desc = "Video"; - size = &video_size; break; default: stop_encoding = 1; @@ -931,7 +1026,6 @@ static void flush_encoders(void) av_log(NULL, AV_LOG_FATAL, "%s encoding failed\n", desc); exit_program(1); } - *size += ret; if (ost->logfile && enc->stats_out) { fprintf(ost->logfile, "%s", enc->stats_out); } @@ -939,10 +1033,7 @@ static void flush_encoders(void) 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); + av_packet_rescale_ts(&pkt, enc->time_base, ost->st->time_base); write_frame(os, &pkt, ost); } @@ -963,7 +1054,7 @@ static int check_output_constraints(InputStream *ist, OutputStream *ost) if (ost->source_index != ist_index) return 0; - if (of->start_time && ist->last_dts < of->start_time) + if (of->start_time != AV_NOPTS_VALUE && ist->last_dts < of->start_time) return 0; return 1; @@ -972,7 +1063,9 @@ static int check_output_constraints(InputStream *ist, OutputStream *ost) static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *pkt) { OutputFile *of = output_files[ost->file_index]; - int64_t ost_tb_start_time = av_rescale_q(of->start_time, AV_TIME_BASE_Q, ost->st->time_base); + InputFile *f = input_files [ist->file_index]; + int64_t start_time = (of->start_time == AV_NOPTS_VALUE) ? 0 : of->start_time; + int64_t ost_tb_start_time = av_rescale_q(start_time, AV_TIME_BASE_Q, ost->st->time_base); AVPacket opkt; av_init_packet(&opkt); @@ -982,18 +1075,24 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p return; if (of->recording_time != INT64_MAX && - ist->last_dts >= of->recording_time + of->start_time) { - ost->is_past_recording_time = 1; + ist->last_dts >= of->recording_time + start_time) { + ost->finished = 1; return; } + if (f->recording_time != INT64_MAX) { + start_time = f->ctx->start_time; + if (f->start_time != AV_NOPTS_VALUE) + start_time += f->start_time; + if (ist->last_dts >= f->recording_time + start_time) { + ost->finished = 1; + return; + } + } + /* force the input stream PTS */ - if (ost->st->codec->codec_type == AVMEDIA_TYPE_AUDIO) - audio_size += pkt->size; - else if (ost->st->codec->codec_type == AVMEDIA_TYPE_VIDEO) { - video_size += pkt->size; + if (ost->enc_ctx->codec_type == AVMEDIA_TYPE_VIDEO) ost->sync_opts++; - } 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; @@ -1010,36 +1109,30 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p opkt.flags = pkt->flags; // FIXME remove the following 2 lines they shall be replaced by the bitstream filters - if ( ost->st->codec->codec_id != AV_CODEC_ID_H264 - && ost->st->codec->codec_id != AV_CODEC_ID_MPEG1VIDEO - && ost->st->codec->codec_id != AV_CODEC_ID_MPEG2VIDEO - && ost->st->codec->codec_id != AV_CODEC_ID_VC1 + if ( ost->enc_ctx->codec_id != AV_CODEC_ID_H264 + && ost->enc_ctx->codec_id != AV_CODEC_ID_MPEG1VIDEO + && ost->enc_ctx->codec_id != AV_CODEC_ID_MPEG2VIDEO + && ost->enc_ctx->codec_id != AV_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; + if (av_parser_change(ost->parser, ost->st->codec, + &opkt.data, &opkt.size, + pkt->data, pkt->size, + pkt->flags & AV_PKT_FLAG_KEY)) { + opkt.buf = av_buffer_create(opkt.data, opkt.size, av_buffer_default_free, NULL, 0); + if (!opkt.buf) + exit_program(1); + } } else { opkt.data = pkt->data; opkt.size = pkt->size; } write_frame(of->ctx, &opkt, ost); - ost->st->codec->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->last_dts, 1000000, AV_TIME_BASE); - int64_t now = av_gettime() - ist->start; - if (pts > now) - av_usleep(pts - now); - } } int guess_input_channel_layout(InputStream *ist) { - AVCodecContext *dec = ist->st->codec; + AVCodecContext *dec = ist->dec_ctx; if (!dec->channel_layout) { char layout_name[256]; @@ -1057,94 +1150,35 @@ int guess_input_channel_layout(InputStream *ist) static int decode_audio(InputStream *ist, AVPacket *pkt, int *got_output) { - AVFrame *decoded_frame; - AVCodecContext *avctx = ist->st->codec; - int bps = av_get_bytes_per_sample(ist->st->codec->sample_fmt); - int i, ret, resample_changed; + AVFrame *decoded_frame, *f; + AVCodecContext *avctx = ist->dec_ctx; + int i, ret, err = 0, resample_changed; - if (!ist->decoded_frame && !(ist->decoded_frame = avcodec_alloc_frame())) + if (!ist->decoded_frame && !(ist->decoded_frame = av_frame_alloc())) + return AVERROR(ENOMEM); + if (!ist->filter_frame && !(ist->filter_frame = av_frame_alloc())) return AVERROR(ENOMEM); - else - avcodec_get_frame_defaults(ist->decoded_frame); decoded_frame = ist->decoded_frame; ret = avcodec_decode_audio4(avctx, decoded_frame, got_output, pkt); if (!*got_output || ret < 0) { if (!pkt->size) { for (i = 0; i < ist->nb_filters; i++) - av_buffersrc_buffer(ist->filters[i]->filter, NULL); + av_buffersrc_add_frame(ist->filters[i]->filter, NULL); } return ret; } + ist->samples_decoded += decoded_frame->nb_samples; + ist->frames_decoded++; + /* 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_dts = decoded_frame->pts; - else if (pkt->pts != AV_NOPTS_VALUE) { + else if (pkt->pts != AV_NOPTS_VALUE) decoded_frame->pts = pkt->pts; - pkt->pts = AV_NOPTS_VALUE; - } - - // preprocess audio (volume) - if (audio_volume != 256) { - int decoded_data_size = decoded_frame->nb_samples * avctx->channels * bps; - void *samples = decoded_frame->data[0]; - switch (avctx->sample_fmt) { - case AV_SAMPLE_FMT_U8: - { - uint8_t *volp = samples; - for (i = 0; i < (decoded_data_size / sizeof(*volp)); i++) { - int v = (((*volp - 128) * audio_volume + 128) >> 8) + 128; - *volp++ = av_clip_uint8(v); - } - break; - } - case AV_SAMPLE_FMT_S16: - { - int16_t *volp = samples; - for (i = 0; i < (decoded_data_size / sizeof(*volp)); i++) { - int v = ((*volp) * audio_volume + 128) >> 8; - *volp++ = av_clip_int16(v); - } - break; - } - case AV_SAMPLE_FMT_S32: - { - int32_t *volp = samples; - for (i = 0; i < (decoded_data_size / sizeof(*volp)); i++) { - int64_t v = (((int64_t)*volp * audio_volume + 128) >> 8); - *volp++ = av_clipl_int32(v); - } - break; - } - case AV_SAMPLE_FMT_FLT: - { - float *volp = samples; - float scale = audio_volume / 256.f; - for (i = 0; i < (decoded_data_size / sizeof(*volp)); i++) { - *volp++ *= scale; - } - break; - } - case AV_SAMPLE_FMT_DBL: - { - double *volp = samples; - double scale = audio_volume / 256.; - for (i = 0; i < (decoded_data_size / sizeof(*volp)); i++) { - *volp++ *= scale; - } - break; - } - default: - av_log(NULL, AV_LOG_FATAL, - "Audio volume adjustment on sample format %s is not supported.\n", - av_get_sample_fmt_name(ist->st->codec->sample_fmt)); - exit_program(1); - } - } - - rate_emu_sleep(ist); + pkt->pts = AV_NOPTS_VALUE; resample_changed = ist->resample_sample_fmt != decoded_frame->format || ist->resample_channels != avctx->channels || @@ -1190,43 +1224,59 @@ static int decode_audio(InputStream *ist, AVPacket *pkt, int *got_output) if (decoded_frame->pts != AV_NOPTS_VALUE) decoded_frame->pts = av_rescale_q(decoded_frame->pts, ist->st->time_base, - (AVRational){1, ist->st->codec->sample_rate}); - for (i = 0; i < ist->nb_filters; i++) - av_buffersrc_write_frame(ist->filters[i]->filter, decoded_frame); + (AVRational){1, avctx->sample_rate}); + for (i = 0; i < ist->nb_filters; i++) { + if (i < ist->nb_filters - 1) { + f = ist->filter_frame; + err = av_frame_ref(f, decoded_frame); + if (err < 0) + break; + } else + f = decoded_frame; - return ret; + err = av_buffersrc_add_frame(ist->filters[i]->filter, f); + if (err < 0) + break; + } + + av_frame_unref(ist->filter_frame); + av_frame_unref(decoded_frame); + return err < 0 ? err : ret; } static int decode_video(InputStream *ist, AVPacket *pkt, int *got_output) { - AVFrame *decoded_frame; - void *buffer_to_free = NULL; - int i, ret = 0, resample_changed; - float quality; + AVFrame *decoded_frame, *f; + int i, ret = 0, err = 0, resample_changed; - if (!ist->decoded_frame && !(ist->decoded_frame = avcodec_alloc_frame())) + if (!ist->decoded_frame && !(ist->decoded_frame = av_frame_alloc())) + return AVERROR(ENOMEM); + if (!ist->filter_frame && !(ist->filter_frame = av_frame_alloc())) return AVERROR(ENOMEM); - else - avcodec_get_frame_defaults(ist->decoded_frame); decoded_frame = ist->decoded_frame; - ret = avcodec_decode_video2(ist->st->codec, + ret = avcodec_decode_video2(ist->dec_ctx, decoded_frame, got_output, pkt); if (!*got_output || ret < 0) { if (!pkt->size) { for (i = 0; i < ist->nb_filters; i++) - av_buffersrc_buffer(ist->filters[i]->filter, NULL); + av_buffersrc_add_frame(ist->filters[i]->filter, NULL); } return ret; } - quality = same_quant ? decoded_frame->quality : 0; + ist->frames_decoded++; + + if (ist->hwaccel_retrieve_data && decoded_frame->format == ist->hwaccel_pix_fmt) { + err = ist->hwaccel_retrieve_data(ist->dec_ctx, decoded_frame); + if (err < 0) + goto fail; + } + ist->hwaccel_retrieved_pix_fmt = decoded_frame->format; + 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; @@ -1241,6 +1291,14 @@ static int decode_video(InputStream *ist, AVPacket *pkt, int *got_output) ist->resample_width, ist->resample_height, av_get_pix_fmt_name(ist->resample_pix_fmt), decoded_frame->width, decoded_frame->height, av_get_pix_fmt_name(decoded_frame->format)); + ret = poll_filters(); + if (ret < 0 && (ret != AVERROR_EOF && ret != AVERROR(EAGAIN))) { + char errbuf[128]; + av_strerror(ret, errbuf, sizeof(errbuf)); + + av_log(NULL, AV_LOG_ERROR, "Error while filtering: %s\n", errbuf); + } + ist->resample_width = decoded_frame->width; ist->resample_height = decoded_frame->height; ist->resample_pix_fmt = decoded_frame->format; @@ -1254,43 +1312,36 @@ static int decode_video(InputStream *ist, AVPacket *pkt, int *got_output) } for (i = 0; i < ist->nb_filters; i++) { - // XXX what an ugly hack - if (ist->filters[i]->graph->nb_outputs == 1) - ist->filters[i]->graph->outputs[0]->ost->last_quality = quality; - - 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(ist->filters[i]->filter, fb); + if (i < ist->nb_filters - 1) { + f = ist->filter_frame; + err = av_frame_ref(f, decoded_frame); + if (err < 0) + break; } else - av_buffersrc_write_frame(ist->filters[i]->filter, decoded_frame); + f = decoded_frame; + + err = av_buffersrc_add_frame(ist->filters[i]->filter, f); + if (err < 0) + break; } - av_free(buffer_to_free); - return ret; +fail: + av_frame_unref(ist->filter_frame); + av_frame_unref(decoded_frame); + return err < 0 ? err : ret; } static int transcode_subtitles(InputStream *ist, AVPacket *pkt, int *got_output) { AVSubtitle subtitle; - int i, ret = avcodec_decode_subtitle2(ist->st->codec, + int i, ret = avcodec_decode_subtitle2(ist->dec_ctx, &subtitle, got_output, pkt); if (ret < 0) return ret; if (!*got_output) return ret; - rate_emu_sleep(ist); + ist->frames_decoded++; for (i = 0; i < nb_output_streams; i++) { OutputStream *ost = output_streams[i]; @@ -1306,7 +1357,7 @@ static int transcode_subtitles(InputStream *ist, AVPacket *pkt, int *got_output) } /* pkt = NULL means EOF (needed to flush decoder buffers) */ -static int output_packet(InputStream *ist, const AVPacket *pkt) +static int process_input_packet(InputStream *ist, const AVPacket *pkt) { int i; int got_output; @@ -1315,7 +1366,7 @@ static int output_packet(InputStream *ist, const AVPacket *pkt) if (ist->next_dts == AV_NOPTS_VALUE) ist->next_dts = ist->last_dts; - if (pkt == NULL) { + if (!pkt) { /* EOF handling */ av_init_packet(&avpkt); avpkt.data = NULL; @@ -1335,13 +1386,14 @@ static int output_packet(InputStream *ist, const AVPacket *pkt) ist->last_dts = ist->next_dts; - if (avpkt.size && avpkt.size != pkt->size) { + if (avpkt.size && avpkt.size != pkt->size && + !(ist->dec->capabilities & CODEC_CAP_SUBFRAMES)) { av_log(NULL, ist->showed_multi_packet_warning ? AV_LOG_VERBOSE : AV_LOG_WARNING, "Multiple frames in a packet from stream %d\n", pkt->stream_index); ist->showed_multi_packet_warning = 1; } - switch (ist->st->codec->codec_type) { + switch (ist->dec_ctx->codec_type) { case AVMEDIA_TYPE_AUDIO: ret = decode_audio (ist, &avpkt, &got_output); break; @@ -1352,10 +1404,10 @@ static int output_packet(InputStream *ist, const AVPacket *pkt) else if (ist->st->avg_frame_rate.num) ist->next_dts += av_rescale_q(1, av_inv_q(ist->st->avg_frame_rate), AV_TIME_BASE_Q); - else if (ist->st->codec->time_base.num != 0) { + else if (ist->dec_ctx->framerate.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); + ist->dec_ctx->ticks_per_frame; + ist->next_dts += av_rescale_q(ticks, ist->dec_ctx->framerate, AV_TIME_BASE_Q); } break; case AVMEDIA_TYPE_SUBTITLE: @@ -1379,19 +1431,18 @@ static int output_packet(InputStream *ist, const AVPacket *pkt) /* handle stream copy */ if (!ist->decoding_needed) { - rate_emu_sleep(ist); ist->last_dts = ist->next_dts; - switch (ist->st->codec->codec_type) { + switch (ist->dec_ctx->codec_type) { case AVMEDIA_TYPE_AUDIO: - ist->next_dts += ((int64_t)AV_TIME_BASE * ist->st->codec->frame_size) / - ist->st->codec->sample_rate; + ist->next_dts += ((int64_t)AV_TIME_BASE * ist->dec_ctx->frame_size) / + ist->dec_ctx->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; + if (ist->dec_ctx->framerate.num != 0) { + int ticks = ist->st->parser ? ist->st->parser->repeat_pict + 1 : ist->dec_ctx->ticks_per_frame; ist->next_dts += ((int64_t)AV_TIME_BASE * - ist->st->codec->time_base.num * ticks) / - ist->st->codec->time_base.den; + ist->dec_ctx->framerate.den * ticks) / + ist->dec_ctx->framerate.num; } break; } @@ -1410,7 +1461,7 @@ static int output_packet(InputStream *ist, const AVPacket *pkt) static void print_sdp(void) { - char sdp[2048]; + char sdp[16384]; int i; AVFormatContext **avc = av_malloc(sizeof(*avc) * nb_output_files); @@ -1425,15 +1476,72 @@ static void print_sdp(void) av_freep(&avc); } -static int init_input_stream(int ist_index, char *error, int error_len) +static const HWAccel *get_hwaccel(enum AVPixelFormat pix_fmt) { int i; + for (i = 0; hwaccels[i].name; i++) + if (hwaccels[i].pix_fmt == pix_fmt) + return &hwaccels[i]; + return NULL; +} + +static enum AVPixelFormat get_format(AVCodecContext *s, const enum AVPixelFormat *pix_fmts) +{ + InputStream *ist = s->opaque; + const enum AVPixelFormat *p; + int ret; + + for (p = pix_fmts; *p != -1; p++) { + const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(*p); + const HWAccel *hwaccel; + + if (!(desc->flags & AV_PIX_FMT_FLAG_HWACCEL)) + break; + + hwaccel = get_hwaccel(*p); + if (!hwaccel || + (ist->active_hwaccel_id && ist->active_hwaccel_id != hwaccel->id) || + (ist->hwaccel_id != HWACCEL_AUTO && ist->hwaccel_id != hwaccel->id)) + continue; + + ret = hwaccel->init(s); + if (ret < 0) { + if (ist->hwaccel_id == hwaccel->id) { + av_log(NULL, AV_LOG_FATAL, + "%s hwaccel requested for input stream #%d:%d, " + "but cannot be initialized.\n", hwaccel->name, + ist->file_index, ist->st->index); + return AV_PIX_FMT_NONE; + } + continue; + } + ist->active_hwaccel_id = hwaccel->id; + ist->hwaccel_pix_fmt = *p; + break; + } + + return *p; +} + +static int get_buffer(AVCodecContext *s, AVFrame *frame, int flags) +{ + InputStream *ist = s->opaque; + + if (ist->hwaccel_get_buffer && frame->format == ist->hwaccel_pix_fmt) + return ist->hwaccel_get_buffer(s, frame, flags); + + return avcodec_default_get_buffer2(s, frame, flags); +} + +static int init_input_stream(int ist_index, char *error, int error_len) +{ + int i, ret; InputStream *ist = input_streams[ist_index]; if (ist->decoding_needed) { AVCodec *codec = ist->dec; if (!codec) { snprintf(error, error_len, "Decoder (codec id %d) not found for input stream #%d:%d", - ist->st->codec->codec_id, ist->file_index, ist->st->index); + ist->dec_ctx->codec_id, ist->file_index, ist->st->index); return AVERROR(EINVAL); } @@ -1442,32 +1550,39 @@ static int init_input_stream(int ist_index, char *error, int error_len) for (i = 0; i < nb_output_streams; i++) { OutputStream *ost = output_streams[i]; if (ost->source_index == ist_index) { - update_sample_fmt(ist->st->codec, codec, ost->st->codec); + update_sample_fmt(ist->dec_ctx, codec, ost->enc_ctx); break; } } - if (codec->type == AVMEDIA_TYPE_VIDEO && codec->capabilities & CODEC_CAP_DR1) { - ist->st->codec->get_buffer = codec_get_buffer; - ist->st->codec->release_buffer = codec_release_buffer; - ist->st->codec->opaque = &ist->buffer_pool; - } + ist->dec_ctx->opaque = ist; + ist->dec_ctx->get_format = get_format; + ist->dec_ctx->get_buffer2 = get_buffer; + ist->dec_ctx->thread_safe_callbacks = 1; - 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); - return AVERROR(EINVAL); + av_opt_set_int(ist->dec_ctx, "refcounted_frames", 1, 0); + + if (!av_dict_get(ist->decoder_opts, "threads", NULL, 0)) + av_dict_set(&ist->decoder_opts, "threads", "auto", 0); + if ((ret = avcodec_open2(ist->dec_ctx, codec, &ist->decoder_opts)) < 0) { + char errbuf[128]; + if (ret == AVERROR_EXPERIMENTAL) + abort_codec_experimental(codec, 0); + + av_strerror(ret, errbuf, sizeof(errbuf)); + + snprintf(error, error_len, + "Error while opening decoder for input stream " + "#%d:%d : %s", + ist->file_index, ist->st->index, errbuf); + return ret; } - assert_codec_experimental(ist->st->codec, 0); - assert_avoptions(ist->opts); + assert_avoptions(ist->decoder_opts); } - 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->last_dts = ist->st->avg_frame_rate.num ? - ist->dec_ctx->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; return 0; } @@ -1482,7 +1597,7 @@ static InputStream *get_input_stream(OutputStream *ost) int i; for (i = 0; i < fg->nb_inputs; i++) - if (fg->inputs[i]->ist->st->codec->codec_type == ost->st->codec->codec_type) + if (fg->inputs[i]->ist->dec_ctx->codec_type == ost->enc_ctx->codec_type) return fg->inputs[i]->ist; } @@ -1520,11 +1635,38 @@ static void parse_forced_key_frames(char *kf, OutputStream *ost, } } +static void set_encoder_id(OutputFile *of, OutputStream *ost) +{ + AVDictionaryEntry *e; + + uint8_t *encoder_string; + int encoder_string_len; + int format_flags = 0; + + e = av_dict_get(of->opts, "fflags", NULL, 0); + if (e) { + const AVOption *o = av_opt_find(of->ctx, "fflags", NULL, 0, 0); + if (!o) + return; + av_opt_eval_flags(of->ctx, o, e->value, &format_flags); + } + + encoder_string_len = sizeof(LIBAVCODEC_IDENT) + strlen(ost->enc->name) + 2; + encoder_string = av_mallocz(encoder_string_len); + if (!encoder_string) + exit_program(1); + + if (!(format_flags & AVFMT_FLAG_BITEXACT)) + av_strlcpy(encoder_string, LIBAVCODEC_IDENT " ", encoder_string_len); + av_strlcat(encoder_string, ost->enc->name, encoder_string_len); + av_dict_set(&ost->st->metadata, "encoder", encoder_string, + AV_DICT_DONT_STRDUP_VAL | AV_DICT_DONT_OVERWRITE); +} + static int transcode_init(void) { int ret = 0, i, j, k; AVFormatContext *oc; - AVCodecContext *codec, *icodec; OutputStream *ost; InputStream *ist; char error[1024]; @@ -1535,7 +1677,7 @@ static int transcode_init(void) InputFile *ifile = input_files[i]; if (ifile->rate_emu) for (j = 0; j < ifile->nb_streams; j++) - input_streams[j + ifile->ist_index]->start = av_gettime(); + input_streams[j + ifile->ist_index]->start = av_gettime_relative(); } /* output stream init */ @@ -1555,6 +1697,8 @@ static int transcode_init(void) /* for each output stream, we compute the right encoding parameters */ for (i = 0; i < nb_output_streams; i++) { + AVCodecContext *enc_ctx; + AVCodecContext *dec_ctx = NULL; ost = output_streams[i]; oc = output_files[ost->file_index]->ctx; ist = get_input_stream(ost); @@ -1562,85 +1706,108 @@ static int transcode_init(void) if (ost->attachment_filename) continue; - codec = ost->st->codec; + enc_ctx = ost->stream_copy ? ost->st->codec : ost->enc_ctx; if (ist) { - icodec = ist->st->codec; + dec_ctx = ist->dec_ctx; ost->st->disposition = ist->st->disposition; - codec->bits_per_raw_sample = icodec->bits_per_raw_sample; - codec->chroma_sample_location = icodec->chroma_sample_location; + enc_ctx->bits_per_raw_sample = dec_ctx->bits_per_raw_sample; + enc_ctx->chroma_sample_location = dec_ctx->chroma_sample_location; } if (ost->stream_copy) { + AVRational sar; uint64_t extra_size; av_assert0(ist && !ost->filter); - extra_size = (uint64_t)icodec->extradata_size + FF_INPUT_BUFFER_PADDING_SIZE; + extra_size = (uint64_t)dec_ctx->extradata_size + FF_INPUT_BUFFER_PADDING_SIZE; if (extra_size > INT_MAX) { return AVERROR(EINVAL); } /* if stream_copy is selected, no need to decode or encode */ - codec->codec_id = icodec->codec_id; - codec->codec_type = icodec->codec_type; + enc_ctx->codec_id = dec_ctx->codec_id; + enc_ctx->codec_type = dec_ctx->codec_type; - if (!codec->codec_tag) { + if (!enc_ctx->codec_tag) { if (!oc->oformat->codec_tag || - av_codec_get_id (oc->oformat->codec_tag, icodec->codec_tag) == codec->codec_id || - av_codec_get_tag(oc->oformat->codec_tag, icodec->codec_id) <= 0) - codec->codec_tag = icodec->codec_tag; + av_codec_get_id (oc->oformat->codec_tag, dec_ctx->codec_tag) == enc_ctx->codec_id || + av_codec_get_tag(oc->oformat->codec_tag, dec_ctx->codec_id) <= 0) + enc_ctx->codec_tag = dec_ctx->codec_tag; } - codec->bit_rate = icodec->bit_rate; - codec->rc_max_rate = icodec->rc_max_rate; - codec->rc_buffer_size = icodec->rc_buffer_size; - codec->field_order = icodec->field_order; - codec->extradata = av_mallocz(extra_size); - if (!codec->extradata) { + enc_ctx->bit_rate = dec_ctx->bit_rate; + enc_ctx->rc_max_rate = dec_ctx->rc_max_rate; + enc_ctx->rc_buffer_size = dec_ctx->rc_buffer_size; + enc_ctx->field_order = dec_ctx->field_order; + enc_ctx->extradata = av_mallocz(extra_size); + if (!enc_ctx->extradata) { return AVERROR(ENOMEM); } - memcpy(codec->extradata, icodec->extradata, icodec->extradata_size); - codec->extradata_size = icodec->extradata_size; + memcpy(enc_ctx->extradata, dec_ctx->extradata, dec_ctx->extradata_size); + enc_ctx->extradata_size = dec_ctx->extradata_size; if (!copy_tb) { - codec->time_base = icodec->time_base; - codec->time_base.num *= icodec->ticks_per_frame; - av_reduce(&codec->time_base.num, &codec->time_base.den, - codec->time_base.num, codec->time_base.den, INT_MAX); + enc_ctx->time_base = dec_ctx->time_base; + enc_ctx->time_base.num *= dec_ctx->ticks_per_frame; + av_reduce(&enc_ctx->time_base.num, &enc_ctx->time_base.den, + enc_ctx->time_base.num, enc_ctx->time_base.den, INT_MAX); } else - codec->time_base = ist->st->time_base; + enc_ctx->time_base = ist->st->time_base; + + if (ist->st->nb_side_data) { + ost->st->side_data = av_realloc_array(NULL, ist->st->nb_side_data, + sizeof(*ist->st->side_data)); + if (!ost->st->side_data) + return AVERROR(ENOMEM); + + for (j = 0; j < ist->st->nb_side_data; j++) { + const AVPacketSideData *sd_src = &ist->st->side_data[j]; + AVPacketSideData *sd_dst = &ost->st->side_data[j]; + + sd_dst->data = av_malloc(sd_src->size); + if (!sd_dst->data) + return AVERROR(ENOMEM); + memcpy(sd_dst->data, sd_src->data, sd_src->size); + sd_dst->size = sd_src->size; + sd_dst->type = sd_src->type; + ost->st->nb_side_data++; + } + } + + ost->parser = av_parser_init(enc_ctx->codec_id); - switch (codec->codec_type) { + switch (enc_ctx->codec_type) { case AVMEDIA_TYPE_AUDIO: if (audio_volume != 256) { av_log(NULL, AV_LOG_FATAL, "-acodec copy and -vol are incompatible (frames are not decoded)\n"); exit_program(1); } - codec->channel_layout = icodec->channel_layout; - codec->sample_rate = icodec->sample_rate; - codec->channels = icodec->channels; - codec->frame_size = icodec->frame_size; - codec->audio_service_type = icodec->audio_service_type; - codec->block_align = icodec->block_align; + enc_ctx->channel_layout = dec_ctx->channel_layout; + enc_ctx->sample_rate = dec_ctx->sample_rate; + enc_ctx->channels = dec_ctx->channels; + enc_ctx->frame_size = dec_ctx->frame_size; + enc_ctx->audio_service_type = dec_ctx->audio_service_type; + enc_ctx->block_align = dec_ctx->block_align; break; case AVMEDIA_TYPE_VIDEO: - codec->pix_fmt = icodec->pix_fmt; - codec->width = icodec->width; - codec->height = icodec->height; - codec->has_b_frames = icodec->has_b_frames; - if (!codec->sample_aspect_ratio.num) { - codec->sample_aspect_ratio = - ost->st->sample_aspect_ratio = - ist->st->sample_aspect_ratio.num ? ist->st->sample_aspect_ratio : - ist->st->codec->sample_aspect_ratio.num ? - ist->st->codec->sample_aspect_ratio : (AVRational){0, 1}; - } + enc_ctx->pix_fmt = dec_ctx->pix_fmt; + enc_ctx->width = dec_ctx->width; + enc_ctx->height = dec_ctx->height; + enc_ctx->has_b_frames = dec_ctx->has_b_frames; + if (ost->frame_aspect_ratio) + sar = av_d2q(ost->frame_aspect_ratio * enc_ctx->height / enc_ctx->width, 255); + else if (ist->st->sample_aspect_ratio.num) + sar = ist->st->sample_aspect_ratio; + else + sar = dec_ctx->sample_aspect_ratio; + ost->st->sample_aspect_ratio = enc_ctx->sample_aspect_ratio = sar; break; case AVMEDIA_TYPE_SUBTITLE: - codec->width = icodec->width; - codec->height = icodec->height; + enc_ctx->width = dec_ctx->width; + enc_ctx->height = dec_ctx->height; break; case AVMEDIA_TYPE_DATA: case AVMEDIA_TYPE_ATTACHMENT: @@ -1664,6 +1831,8 @@ static int transcode_init(void) ist->decoding_needed = 1; ost->encoding_needed = 1; + set_encoder_id(output_files[ost->file_index], ost); + /* * We want CFR output if and only if one of those is true: * 1) user specified output framerate with -r @@ -1673,12 +1842,25 @@ static int transcode_init(void) * * in such a case, set ost->frame_rate */ - if (codec->codec_type == AVMEDIA_TYPE_VIDEO && + if (enc_ctx->codec_type == AVMEDIA_TYPE_VIDEO && !ost->frame_rate.num && ist && (video_sync_method == VSYNC_CFR || (video_sync_method == VSYNC_AUTO && !(oc->oformat->flags & (AVFMT_NOTIMESTAMPS | AVFMT_VARIABLE_FPS))))) { - ost->frame_rate = ist->st->avg_frame_rate.num ? ist->st->avg_frame_rate : (AVRational){25, 1}; + if (ist->framerate.num) + ost->frame_rate = ist->framerate; + else if (ist->st->avg_frame_rate.num) + ost->frame_rate = ist->st->avg_frame_rate; + else { + av_log(NULL, AV_LOG_WARNING, "Constant framerate requested " + "for the output stream #%d:%d, but no information " + "about the input framerate is available. Falling " + "back to a default value of 25fps. Use the -r option " + "if you want a different framerate.\n", + ost->file_index, ost->index); + ost->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]; @@ -1686,64 +1868,68 @@ static int transcode_init(void) } if (!ost->filter && - (codec->codec_type == AVMEDIA_TYPE_VIDEO || - codec->codec_type == AVMEDIA_TYPE_AUDIO)) { + (enc_ctx->codec_type == AVMEDIA_TYPE_VIDEO || + enc_ctx->codec_type == AVMEDIA_TYPE_AUDIO)) { FilterGraph *fg; fg = init_simple_filtergraph(ist, ost); if (configure_filtergraph(fg)) { av_log(NULL, AV_LOG_FATAL, "Error opening filters!\n"); - exit(1); + exit_program(1); } } - switch (codec->codec_type) { + switch (enc_ctx->codec_type) { case AVMEDIA_TYPE_AUDIO: - codec->sample_fmt = ost->filter->filter->inputs[0]->format; - codec->sample_rate = ost->filter->filter->inputs[0]->sample_rate; - codec->channel_layout = ost->filter->filter->inputs[0]->channel_layout; - codec->channels = av_get_channel_layout_nb_channels(codec->channel_layout); - codec->time_base = (AVRational){ 1, codec->sample_rate }; + enc_ctx->sample_fmt = ost->filter->filter->inputs[0]->format; + enc_ctx->sample_rate = ost->filter->filter->inputs[0]->sample_rate; + enc_ctx->channel_layout = ost->filter->filter->inputs[0]->channel_layout; + enc_ctx->channels = av_get_channel_layout_nb_channels(enc_ctx->channel_layout); + enc_ctx->time_base = (AVRational){ 1, enc_ctx->sample_rate }; break; case AVMEDIA_TYPE_VIDEO: - codec->time_base = ost->filter->filter->inputs[0]->time_base; + enc_ctx->time_base = ost->filter->filter->inputs[0]->time_base; - codec->width = ost->filter->filter->inputs[0]->w; - codec->height = ost->filter->filter->inputs[0]->h; - codec->sample_aspect_ratio = ost->st->sample_aspect_ratio = + enc_ctx->width = ost->filter->filter->inputs[0]->w; + enc_ctx->height = ost->filter->filter->inputs[0]->h; + enc_ctx->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 * enc_ctx->height/enc_ctx->width, 255) : ost->filter->filter->inputs[0]->sample_aspect_ratio; - codec->pix_fmt = ost->filter->filter->inputs[0]->format; + enc_ctx->pix_fmt = ost->filter->filter->inputs[0]->format; + + ost->st->avg_frame_rate = ost->frame_rate; - if (codec->width != icodec->width || - codec->height != icodec->height || - codec->pix_fmt != icodec->pix_fmt) { - codec->bits_per_raw_sample = 0; + if (dec_ctx && + (enc_ctx->width != dec_ctx->width || + enc_ctx->height != dec_ctx->height || + enc_ctx->pix_fmt != dec_ctx->pix_fmt)) { + enc_ctx->bits_per_raw_sample = 0; } if (ost->forced_keyframes) parse_forced_key_frames(ost->forced_keyframes, ost, - ost->st->codec); + ost->enc_ctx); break; case AVMEDIA_TYPE_SUBTITLE: - codec->time_base = (AVRational){1, 1000}; + enc_ctx->time_base = (AVRational){1, 1000}; break; default: abort(); break; } /* two pass mode */ - if ((codec->flags & (CODEC_FLAG_PASS1 | CODEC_FLAG_PASS2))) { + if ((enc_ctx->flags & (CODEC_FLAG_PASS1 | CODEC_FLAG_PASS2))) { char logfilename[1024]; FILE *f; snprintf(logfilename, sizeof(logfilename), "%s-%d.log", - pass_logfilename_prefix ? pass_logfilename_prefix : DEFAULT_PASS_LOGFILENAME_PREFIX, + ost->logfile_prefix ? ost->logfile_prefix : + DEFAULT_PASS_LOGFILENAME_PREFIX, i); if (!strcmp(ost->enc->name, "libx264")) { - av_dict_set(&ost->opts, "stats", logfilename, AV_DICT_DONT_OVERWRITE); + av_dict_set(&ost->encoder_opts, "stats", logfilename, AV_DICT_DONT_OVERWRITE); } else { - if (codec->flags & CODEC_FLAG_PASS1) { + if (enc_ctx->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", @@ -1759,7 +1945,7 @@ static int transcode_init(void) logfilename); exit_program(1); } - codec->stats_in = logbuffer; + enc_ctx->stats_in = logbuffer; } } } @@ -1774,33 +1960,45 @@ static int transcode_init(void) AVCodecContext *dec = NULL; if ((ist = get_input_stream(ost))) - dec = ist->st->codec; + dec = ist->dec_ctx; if (dec && dec->subtitle_header) { - ost->st->codec->subtitle_header = av_malloc(dec->subtitle_header_size); - if (!ost->st->codec->subtitle_header) { + ost->enc_ctx->subtitle_header = av_malloc(dec->subtitle_header_size); + if (!ost->enc_ctx->subtitle_header) { ret = AVERROR(ENOMEM); goto dump_format; } - memcpy(ost->st->codec->subtitle_header, dec->subtitle_header, dec->subtitle_header_size); - ost->st->codec->subtitle_header_size = dec->subtitle_header_size; + memcpy(ost->enc_ctx->subtitle_header, dec->subtitle_header, dec->subtitle_header_size); + ost->enc_ctx->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) { + if (!av_dict_get(ost->encoder_opts, "threads", NULL, 0)) + av_dict_set(&ost->encoder_opts, "threads", "auto", 0); + av_dict_set(&ost->encoder_opts, "side_data_only_packets", "1", 0); + + if ((ret = avcodec_open2(ost->enc_ctx, codec, &ost->encoder_opts)) < 0) { + if (ret == AVERROR_EXPERIMENTAL) + abort_codec_experimental(codec, 1); 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); - ret = AVERROR(EINVAL); goto dump_format; } - assert_codec_experimental(ost->st->codec, 1); - assert_avoptions(ost->opts); - if (ost->st->codec->bit_rate && ost->st->codec->bit_rate < 1000) + assert_avoptions(ost->encoder_opts); + if (ost->enc_ctx->bit_rate && ost->enc_ctx->bit_rate < 1000) av_log(NULL, AV_LOG_WARNING, "The bitrate parameter is set too low." "It takes bits/s as argument, not kbits/s\n"); - extra_size += ost->st->codec->extradata_size; - if (ost->st->codec->me_threshold) - input_streams[ost->source_index]->st->codec->debug |= FF_DEBUG_MV; + ret = avcodec_copy_context(ost->st->codec, ost->enc_ctx); + if (ret < 0) { + av_log(NULL, AV_LOG_FATAL, + "Error initializing the output stream codec context.\n"); + exit_program(1); + } + + ost->st->time_base = ost->enc_ctx->time_base; + } else { + ret = av_opt_set_dict(ost->enc_ctx, &ost->encoder_opts); + if (ret < 0) + return ret; + ost->st->time_base = ost->st->codec->time_base; } } @@ -1831,10 +2029,11 @@ static int transcode_init(void) oc->interrupt_callback = int_cb; if ((ret = avformat_write_header(oc, &output_files[i]->opts)) < 0) { char errbuf[128]; - const char *errbuf_ptr = errbuf; - if (av_strerror(ret, errbuf, sizeof(errbuf)) < 0) - errbuf_ptr = strerror(AVUNERROR(ret)); - snprintf(error, sizeof(error), "Could not write header for output file #%d (incorrect codec parameters ?): %s", i, errbuf_ptr); + av_strerror(ret, errbuf, sizeof(errbuf)); + snprintf(error, sizeof(error), + "Could not write header for output file #%d " + "(incorrect codec parameters ?): %s", + i, errbuf); ret = AVERROR(EINVAL); goto dump_format; } @@ -1900,10 +2099,37 @@ static int transcode_init(void) ost->sync_ist->st->index); if (ost->stream_copy) av_log(NULL, AV_LOG_INFO, " (copy)"); - else - av_log(NULL, AV_LOG_INFO, " (%s -> %s)", input_streams[ost->source_index]->dec ? - input_streams[ost->source_index]->dec->name : "?", - ost->enc ? ost->enc->name : "?"); + else { + const AVCodec *in_codec = input_streams[ost->source_index]->dec; + const AVCodec *out_codec = ost->enc; + const char *decoder_name = "?"; + const char *in_codec_name = "?"; + const char *encoder_name = "?"; + const char *out_codec_name = "?"; + const AVCodecDescriptor *desc; + + if (in_codec) { + decoder_name = in_codec->name; + desc = avcodec_descriptor_get(in_codec->id); + if (desc) + in_codec_name = desc->name; + if (!strcmp(decoder_name, in_codec_name)) + decoder_name = "native"; + } + + if (out_codec) { + encoder_name = out_codec->name; + desc = avcodec_descriptor_get(out_codec->id); + if (desc) + out_codec_name = desc->name; + if (!strcmp(encoder_name, out_codec_name)) + encoder_name = "native"; + } + + av_log(NULL, AV_LOG_INFO, " (%s (%s) -> %s (%s))", + in_codec_name, decoder_name, + out_codec_name, encoder_name); + } av_log(NULL, AV_LOG_INFO, "\n"); } @@ -1919,10 +2145,7 @@ static int transcode_init(void) return 0; } -/** - * @return 1 if there are still streams where more output is wanted, - * 0 otherwise - */ +/* Return 1 if there remain streams where more output is wanted, 0 otherwise. */ static int need_output(void) { int i; @@ -1932,13 +2155,13 @@ static int need_output(void) OutputFile *of = output_files[ost->file_index]; AVFormatContext *os = output_files[ost->file_index]->ctx; - if (ost->is_past_recording_time || + if (ost->finished || (os->pb && avio_tell(os->pb) >= of->limit_filesize)) continue; if (ost->frame_number >= ost->max_frames) { int j; for (j = 0; j < of->ctx->nb_streams; j++) - output_streams[of->ost_index + j]->is_past_recording_time = 1; + output_streams[of->ost_index + j]->finished = 1; continue; } @@ -2083,6 +2306,17 @@ static int get_input_packet_mt(InputFile *f, AVPacket *pkt) static int get_input_packet(InputFile *f, AVPacket *pkt) { + if (f->rate_emu) { + int i; + for (i = 0; i < f->nb_streams; i++) { + InputStream *ist = input_streams[f->ist_index + i]; + int64_t pts = av_rescale(ist->last_dts, 1000000, AV_TIME_BASE); + int64_t now = av_gettime_relative() - ist->start; + if (pts > now) + return AVERROR(EAGAIN); + } + } + #if HAVE_PTHREADS if (nb_input_files > 1) return get_input_packet_mt(f, pkt); @@ -2106,13 +2340,158 @@ static void reset_eagain(void) input_files[i]->eagain = 0; } +/* + * Read one packet from an input file and send it for + * - decoding -> lavfi (audio/video) + * - decoding -> encoding -> muxing (subtitles) + * - muxing (streamcopy) + * + * Return + * - 0 -- one packet was read and processed + * - AVERROR(EAGAIN) -- no packets were available for selected file, + * this function should be called again + * - AVERROR_EOF -- this function should not be called again + */ +static int process_input(void) +{ + InputFile *ifile; + AVFormatContext *is; + InputStream *ist; + AVPacket pkt; + int ret, i, j; + + /* select the stream that we must read now */ + ifile = select_input_file(); + /* if none, if is finished */ + if (!ifile) { + if (got_eagain()) { + reset_eagain(); + av_usleep(10000); + return AVERROR(EAGAIN); + } + av_log(NULL, AV_LOG_VERBOSE, "No more inputs to read from.\n"); + return AVERROR_EOF; + } + + is = ifile->ctx; + ret = get_input_packet(ifile, &pkt); + + if (ret == AVERROR(EAGAIN)) { + ifile->eagain = 1; + return ret; + } + if (ret < 0) { + if (ret != AVERROR_EOF) { + print_error(is->filename, ret); + if (exit_on_error) + exit_program(1); + } + ifile->eof_reached = 1; + + for (i = 0; i < ifile->nb_streams; i++) { + ist = input_streams[ifile->ist_index + i]; + if (ist->decoding_needed) + process_input_packet(ist, NULL); + + /* mark all outputs that don't go through lavfi as finished */ + for (j = 0; j < nb_output_streams; j++) { + OutputStream *ost = output_streams[j]; + + if (ost->source_index == ifile->ist_index + i && + (ost->stream_copy || ost->enc->type == AVMEDIA_TYPE_SUBTITLE)) + finish_output_stream(ost); + } + } + + return AVERROR(EAGAIN); + } + + reset_eagain(); + + if (do_pkt_dump) { + av_pkt_dump_log2(NULL, AV_LOG_DEBUG, &pkt, do_hex_dump, + is->streams[pkt.stream_index]); + } + /* the following test is needed in case new streams appear + dynamically in stream : we ignore them */ + if (pkt.stream_index >= ifile->nb_streams) + goto discard_packet; + + ist = input_streams[ifile->ist_index + pkt.stream_index]; + + ist->data_size += pkt.size; + ist->nb_packets++; + + if (ist->discard) + goto discard_packet; + + /* add the stream-global side data to the first packet */ + if (ist->nb_packets == 1) + for (i = 0; i < ist->st->nb_side_data; i++) { + AVPacketSideData *src_sd = &ist->st->side_data[i]; + uint8_t *dst_data; + + if (av_packet_get_side_data(&pkt, src_sd->type, NULL)) + continue; + if (ist->autorotate && src_sd->type == AV_PKT_DATA_DISPLAYMATRIX) + continue; + + dst_data = av_packet_new_side_data(&pkt, src_sd->type, src_sd->size); + if (!dst_data) + exit_program(1); + + memcpy(dst_data, src_sd->data, src_sd->size); + } + + if (pkt.dts != AV_NOPTS_VALUE) + pkt.dts += av_rescale_q(ifile->ts_offset, AV_TIME_BASE_Q, ist->st->time_base); + if (pkt.pts != AV_NOPTS_VALUE) + pkt.pts += av_rescale_q(ifile->ts_offset, AV_TIME_BASE_Q, ist->st->time_base); + + if (pkt.pts != AV_NOPTS_VALUE) + pkt.pts *= ist->ts_scale; + if (pkt.dts != AV_NOPTS_VALUE) + pkt.dts *= ist->ts_scale; + + if ((ist->dec_ctx->codec_type == AVMEDIA_TYPE_VIDEO || + ist->dec_ctx->codec_type == AVMEDIA_TYPE_AUDIO) && + 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_dts; + + if ((FFABS(delta) > 1LL * dts_delta_threshold * AV_TIME_BASE || pkt_dts + 1 < ist->last_dts) && !copy_ts) { + ifile->ts_offset -= delta; + av_log(NULL, AV_LOG_DEBUG, + "timestamp discontinuity %"PRId64", new offset= %"PRId64"\n", + delta, ifile->ts_offset); + pkt.dts -= av_rescale_q(delta, AV_TIME_BASE_Q, ist->st->time_base); + if (pkt.pts != AV_NOPTS_VALUE) + pkt.pts -= av_rescale_q(delta, AV_TIME_BASE_Q, ist->st->time_base); + } + } + + ret = process_input_packet(ist, &pkt); + if (ret < 0) { + av_log(NULL, AV_LOG_ERROR, "Error while decoding stream #%d:%d\n", + ist->file_index, ist->st->index); + if (exit_on_error) + exit_program(1); + } + +discard_packet: + av_free_packet(&pkt); + + return 0; +} + /* * The following code is the main loop of the file converter */ static int transcode(void) { - int ret, i; - AVFormatContext *is, *os; + int ret, i, need_input = 1; + AVFormatContext *os; OutputStream *ost; InputStream *ist; int64_t timer_start; @@ -2124,7 +2503,7 @@ static int transcode(void) av_log(NULL, AV_LOG_INFO, "Press ctrl-c to stop encoding\n"); term_init(); - timer_start = av_gettime(); + timer_start = av_gettime_relative(); #if HAVE_PTHREADS if ((ret = init_input_threads()) < 0) @@ -2132,107 +2511,32 @@ static int transcode(void) #endif while (!received_sigterm) { - InputFile *ifile; - AVPacket pkt; - /* check if there's any stream where output is still needed */ if (!need_output()) { av_log(NULL, AV_LOG_VERBOSE, "No more output streams to write to, finishing.\n"); break; } - /* select the stream that we must read now */ - ifile = select_input_file(); - /* if none, if is finished */ - if (!ifile) { - if (got_eagain()) { - reset_eagain(); - av_usleep(10000); - continue; - } - av_log(NULL, AV_LOG_VERBOSE, "No more inputs to read from, finishing.\n"); - break; + /* read and process one input packet if needed */ + if (need_input) { + ret = process_input(); + if (ret == AVERROR_EOF) + need_input = 0; } - is = ifile->ctx; - ret = get_input_packet(ifile, &pkt); - - if (ret == AVERROR(EAGAIN)) { - ifile->eagain = 1; - continue; - } + ret = poll_filters(); if (ret < 0) { - if (ret != AVERROR_EOF) { - print_error(is->filename, ret); - if (exit_on_error) - exit_program(1); - } - ifile->eof_reached = 1; - - for (i = 0; i < ifile->nb_streams; i++) { - ist = input_streams[ifile->ist_index + i]; - if (ist->decoding_needed) - output_packet(ist, NULL); - } - - if (opt_shortest) - break; - else + if (ret == AVERROR_EOF || ret == AVERROR(EAGAIN)) { continue; - } + } else { + char errbuf[128]; + av_strerror(ret, errbuf, sizeof(errbuf)); - reset_eagain(); - - if (do_pkt_dump) { - av_pkt_dump_log2(NULL, AV_LOG_DEBUG, &pkt, do_hex_dump, - is->streams[pkt.stream_index]); - } - /* the following test is needed in case new streams appear - dynamically in stream : we ignore them */ - if (pkt.stream_index >= ifile->nb_streams) - goto discard_packet; - - ist = input_streams[ifile->ist_index + pkt.stream_index]; - if (ist->discard) - goto discard_packet; - - if (pkt.dts != AV_NOPTS_VALUE) - pkt.dts += av_rescale_q(ifile->ts_offset, AV_TIME_BASE_Q, ist->st->time_base); - if (pkt.pts != AV_NOPTS_VALUE) - pkt.pts += av_rescale_q(ifile->ts_offset, AV_TIME_BASE_Q, ist->st->time_base); - - if (pkt.pts != AV_NOPTS_VALUE) - pkt.pts *= ist->ts_scale; - if (pkt.dts != AV_NOPTS_VALUE) - pkt.dts *= ist->ts_scale; - - 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_dts; - if ((FFABS(delta) > 1LL * dts_delta_threshold * AV_TIME_BASE || pkt_dts + 1 < ist->last_dts) && !copy_ts) { - ifile->ts_offset -= delta; - av_log(NULL, AV_LOG_DEBUG, - "timestamp discontinuity %"PRId64", new offset= %"PRId64"\n", - delta, ifile->ts_offset); - pkt.dts-= av_rescale_q(delta, AV_TIME_BASE_Q, ist->st->time_base); - if (pkt.pts != AV_NOPTS_VALUE) - pkt.pts-= av_rescale_q(delta, AV_TIME_BASE_Q, ist->st->time_base); + av_log(NULL, AV_LOG_ERROR, "Error while filtering: %s\n", errbuf); + break; } } - if (output_packet(ist, &pkt) < 0 || poll_filters() < 0) { - av_log(NULL, AV_LOG_ERROR, "Error while decoding stream #%d:%d\n", - ist->file_index, ist->st->index); - if (exit_on_error) - exit_program(1); - av_free_packet(&pkt); - continue; - } - - discard_packet: - av_free_packet(&pkt); - /* dump report by using the output first video and audio streams */ print_report(0, timer_start); } @@ -2244,7 +2548,7 @@ static int transcode(void) for (i = 0; i < nb_input_streams; i++) { ist = input_streams[i]; if (!input_files[ist->file_index]->eof_reached && ist->decoding_needed) { - output_packet(ist, NULL); + process_input_packet(ist, NULL); } } poll_filters(); @@ -2265,8 +2569,7 @@ static int transcode(void) for (i = 0; i < nb_output_streams; i++) { ost = output_streams[i]; if (ost->encoding_needed) { - av_freep(&ost->st->codec->stats_in); - avcodec_close(ost->st->codec); + av_freep(&ost->enc_ctx->stats_in); } } @@ -2274,7 +2577,9 @@ static int transcode(void) for (i = 0; i < nb_input_streams; i++) { ist = input_streams[i]; if (ist->decoding_needed) { - avcodec_close(ist->st->codec); + avcodec_close(ist->dec_ctx); + if (ist->hwaccel_uninit) + ist->hwaccel_uninit(ist->dec_ctx); } } @@ -2290,15 +2595,13 @@ static int transcode(void) for (i = 0; i < nb_output_streams; i++) { ost = output_streams[i]; if (ost) { - if (ost->stream_copy) - av_freep(&ost->st->codec->extradata); if (ost->logfile) { fclose(ost->logfile); ost->logfile = NULL; } - av_freep(&ost->st->codec->subtitle_header); av_free(ost->forced_kf_pts); - av_dict_free(&ost->opts); + av_dict_free(&ost->encoder_opts); + av_dict_free(&ost->resample_opts); } } } @@ -2319,7 +2622,7 @@ static int64_t getutime(void) GetProcessTimes(proc, &c, &e, &k, &u); return ((int64_t) u.dwHighDateTime << 32 | u.dwLowDateTime) / 10; #else - return av_gettime(); + return av_gettime_relative(); #endif } @@ -2341,19 +2644,12 @@ static int64_t getmaxrss(void) #endif } -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]); -} - int main(int argc, char **argv) { - OptionsContext o = { 0 }; + int ret; int64_t ti; - reset_options(&o); + register_exit(avconv_cleanup); av_log_set_flags(AV_LOG_SKIP_REPEATED); parse_loglevel(argc, argv, options); @@ -2368,10 +2664,10 @@ int main(int argc, char **argv) show_banner(); - parse_cpuflags(argc, argv, options); - - /* parse options */ - parse_options(&o, argc, argv, options, opt_output_file); + /* parse options and open all input/output files */ + ret = avconv_parse_options(argc, argv); + if (ret < 0) + exit_program(1); if (nb_output_files <= 0 && nb_input_files == 0) { show_usage(); @@ -2385,11 +2681,6 @@ int main(int argc, char **argv) exit_program(1); } - if (nb_input_files == 0) { - av_log(NULL, AV_LOG_FATAL, "At least one input file must be specified\n"); - exit_program(1); - } - ti = getutime(); if (transcode() < 0) exit_program(1);