April 2014 - ffmpeg-devel - lists.ffmpeg.org

[PATCH] cdxl: fix duration
by Piotr Bandurski 22 Dec '14

22 Dec '14

fixes ticket #1937

4 10

[PATCH] avcodec/libxavs: remove global header code
by Michael Niedermayer 06 Dec '14

06 Dec '14

The code does not work, also the resulting files cannot be played when one fixes the code so it doesnt crash Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> --- libavcodec/libxavs.c | 30 ------------------------------ 1 file changed, 30 deletions(-) diff --git a/libavcodec/libxavs.c b/libavcodec/libxavs.c index ffe3411..91507f4 100644 --- a/libavcodec/libxavs.c +++ b/libavcodec/libxavs.c @@ -200,7 +200,6 @@ static av_cold int XAVS_close(AVCodecContext *avctx) { XavsContext *x4 = avctx->priv_data; - av_freep(&avctx->extradata); av_free(x4->sei); av_freep(&x4->pts_buffer); @@ -346,9 +345,6 @@ static av_cold int XAVS_init(AVCodecContext *avctx) x4->params.i_threads = avctx->thread_count; x4->params.b_interlaced = avctx->flags & CODEC_FLAG_INTERLACED_DCT; - if (avctx->flags & CODEC_FLAG_GLOBAL_HEADER) - x4->params.b_repeat_headers = 0; - x4->enc = xavs_encoder_open(&x4->params); if (!x4->enc) return -1; @@ -360,32 +356,6 @@ static av_cold int XAVS_init(AVCodecContext *avctx) if (!avctx->coded_frame) return AVERROR(ENOMEM); - /* TAG: Do we have GLOBAL HEADER in AVS */ - /* We Have PPS and SPS in AVS */ - if (avctx->flags & CODEC_FLAG_GLOBAL_HEADER) { - xavs_nal_t *nal; - int nnal, s, i, size; - uint8_t *p; - - s = xavs_encoder_headers(x4->enc, &nal, &nnal); - - avctx->extradata = p = av_malloc(s); - for (i = 0; i < nnal; i++) { - /* Don't put the SEI in extradata. */ - if (nal[i].i_type == NAL_SEI) { - x4->sei = av_malloc( 5 + nal[i].i_payload * 4 / 3 ); - if (xavs_nal_encode(x4->sei, &x4->sei_size, 1, nal + i) < 0) - return -1; - - continue; - } - size = xavs_nal_encode(p, &s, 1, nal + i); - if (size < 0) - return -1; - p += size; - } - avctx->extradata_size = p - avctx->extradata; - } return 0; } -- 1.7.9.5

1 1

Use ffmpeg as an Android stagefright plugin
by Chih-Wei Huang 20 Nov '14

20 Nov '14

Hello ffmpeg list, I'm planning to integrate the latest ffmpeg as a stagefright OMXPlugin with Android-x86 4.3. I found the stagefright-plugins developed by Michael Chen is a good start. (repo: https://github.com/omxcodec/stagefright-plugins.git) But it's based on android 4.0 (ICS). My plan is: * Update stagefright-plugins to work with stagefright of android 4.3. * Make it to be a libstagefrighthw.so plugin. * Create Android.mk to build ffmpeg libs. Before I go ahead, I'd like to know if anyone is doing a similar project to avoid duplicate work. Thanks a lot! -- Chih-Wei Android-x86 project http://www.android-x86.org

6 20

[PATCH] v4l2: setting device parameters early
by Federico Simoncelli 08 Nov '14

08 Nov '14

Setting the v4l2 device parameters may influence video properties such as width, height and pixel format. This patch moves v4l2_set_parameters to be executed before fetching the video properties, avoiding errors such as (pal vs ntsc): The v4l2 frame is 829440 bytes, but 691200 bytes are expected Signed-off-by: Federico Simoncelli <fsimonce(a)redhat.com> --- libavdevice/v4l2.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/libavdevice/v4l2.c b/libavdevice/v4l2.c index cb962b7..1ab6f1c 100644 --- a/libavdevice/v4l2.c +++ b/libavdevice/v4l2.c @@ -866,6 +866,9 @@ static int v4l2_read_header(AVFormatContext *s1) avpriv_set_pts_info(st, 64, 1, 1000000); /* 64 bits pts in us */ + if ((res = v4l2_set_parameters(s1)) < 0) + return res; + if (s->pixel_format) { AVCodec *codec = avcodec_find_decoder_by_name(s->pixel_format); @@ -917,9 +920,6 @@ static int v4l2_read_header(AVFormatContext *s1) s->frame_format = desired_format; - if ((res = v4l2_set_parameters(s1)) < 0) - return res; - st->codec->pix_fmt = avpriv_fmt_v4l2ff(desired_format, codec_id); s->frame_size = avpicture_get_size(st->codec->pix_fmt, s->width, s->height); -- 1.8.4.2

3 7

[PATCH] Include content of the news article in the website RSS
by Gerion Entrup 30 Aug '14

30 Aug '14

Fixes Trac ticket #844 --- Makefile | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/Makefile b/Makefile index db2ac28..a22232c 100644 --- a/Makefile +++ b/Makefile @@ -19,19 +19,35 @@ htdocs/%.html: src/% src/%_title $(PAGE_DEPS) htdocs/main.rss: htdocs/index.html echo '<?xml version="1.0" encoding="UTF-8" ?>' > $@ - echo '<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">' >> $@ + echo '<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">' >> $@ echo '<channel>' >> $@ echo ' <title>FFmpeg RSS</title>' >> $@ echo ' <link>http://ffmpeg.org</link>' >> $@ echo ' <description>FFmpeg RSS</description>' >> $@ echo ' <atom:link href="http://ffmpeg.org/main.rss" rel="self" type="application/rss+xml" />' >> $@ - grep '<a *id=".*" *></a><h3>.*20..,.*</h3>' $< | sed 'sX<a *id="$.*$" *> *</a> *<h3>$.*20..$, *$.*$</h3>X\ + NSTART='' + while read line; do \ + if echo "$$line" | grep '<h1>Older entries are in the .*news archive' -q ; then \ + break; \ + fi; \ + if echo "$$line" | grep '<a *id=".*" *></a><h3>.*20..,.*</h3>' -q ; then \ + if [ $$NSTART ]; then \ + echo " ]]></content:encoded>" >> $@; \ + echo " </item>" >> $@ ; echo >> $@; \ + fi; \ + echo "$$line" | sed 'sX<a *id="$.*$" *> *</a> *<h3>$.*20..$, *$.*$</h3>X\ <item>\ <title>\2, \3</title>\ <link>http://ffmpeg.org/index.html#\1</link>\ - <guid>http://ffmpeg.org/index.html#\1</guid>\ - </item>\ -X' >> $@ + <guid>http://ffmpeg.org/index.html#\1</guid>X' >> $@; \ + echo -n ' <content:encoded><![CDATA[' >> $@; \ + NSTART='yes'; \ + elif [ $$NSTART -a "$$line" ]; then \ + echo " $$line" >> $@; \ + fi; \ + done < $< + echo ' ]]></content:encoded>' >> $@ + echo ' </item>' >> $@ echo '</channel>' >> $@ echo '</rss>' >> $@ -- 1.8.4.5

4 7

[PATCH 1/4] lavc/flacenc: add sse4 version of the 16-bit lpc encoder
by James Darnley 14 Aug '14

14 Aug '14

>From 1.8 to 2.4 times faster. Runtime is reduced by 2 to 39%. The speed-up generally increases with compression_level. This lpc encoder is not used with levels < 3 so it provides no speed-up in these cases. --- LICENSE | 1 + libavcodec/flacenc.c | 2 +- libavcodec/x86/Makefile | 3 + libavcodec/x86/flac_dsp_gpl.asm | 79 +++++++++++++++++++++++++++++++++++++++ libavcodec/x86/flacdsp_init.c | 4 ++ 5 files changed, 88 insertions(+), 1 deletions(-) create mode 100644 libavcodec/x86/flac_dsp_gpl.asm diff --git a/LICENSE b/LICENSE index 1f757aa..c194087 100644 --- a/LICENSE +++ b/LICENSE @@ -15,6 +15,7 @@ Specifically, the GPL parts of FFmpeg are - libpostproc - libmpcodecs - optional x86 optimizations in the files + libavcodec/x86/flac_dsp_gpl.asm libavcodec/x86/idct_mmx.c - libutvideo encoding/decoding wrappers in libavcodec/libutvideo*.cpp diff --git a/libavcodec/flacenc.c b/libavcodec/flacenc.c index 1fc8c4c..e958cd8 100644 --- a/libavcodec/flacenc.c +++ b/libavcodec/flacenc.c @@ -80,7 +80,7 @@ typedef struct FlacSubframe { int shift; RiceContext rc; int32_t samples[FLAC_MAX_BLOCKSIZE]; - int32_t residual[FLAC_MAX_BLOCKSIZE+1]; + int32_t residual[FLAC_MAX_BLOCKSIZE+3]; } FlacSubframe; typedef struct FlacFrame { diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile index 0d3594f..374b1d2 100644 --- a/libavcodec/x86/Makefile +++ b/libavcodec/x86/Makefile @@ -76,6 +76,9 @@ YASM-OBJS-$(CONFIG_DSPUTIL) += x86/dsputil.o \ YASM-OBJS-$(CONFIG_ENCODERS) += x86/dsputilenc.o YASM-OBJS-$(CONFIG_FFT) += x86/fft.o YASM-OBJS-$(CONFIG_FLAC_DECODER) += x86/flacdsp.o +ifdef CONFIG_GPL +YASM-OBJS-$(CONFIG_FLAC_ENCODER) += x86/flac_dsp_gpl.o +endif YASM-OBJS-$(CONFIG_H263DSP) += x86/h263_loopfilter.o YASM-OBJS-$(CONFIG_H264CHROMA) += x86/h264_chromamc.o \ x86/h264_chromamc_10bit.o diff --git a/libavcodec/x86/flac_dsp_gpl.asm b/libavcodec/x86/flac_dsp_gpl.asm new file mode 100644 index 0000000..cc07194 --- /dev/null +++ b/libavcodec/x86/flac_dsp_gpl.asm @@ -0,0 +1,79 @@ +;***************************************************************************** +;* FLAC DSP functions +;* +;* Copyright (c) 2014 James Darnley <james.darnley(a)gmail.com> +;* +;* This file is part of FFmpeg. +;* +;* FFmpeg is free software; you can redistribute it and/or modify +;* it under the terms of the GNU General Public License as published by +;* the Free Software Foundation; either version 2 of the License, or +;* (at your option) any later version. +;* +;* FFmpeg is distributed in the hope that it will be useful, +;* but WITHOUT ANY WARRANTY; without even the implied warranty of +;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;* GNU General Public License for more details. +;* +;* You should have received a copy of the GNU General Public License along +;* with FFmpeg; if not, write to the Free Software Foundation, Inc., +;* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. +;****************************************************************************** + +%include "libavutil/x86/x86util.asm" + +SECTION_TEXT + +INIT_XMM sse4 +%if ARCH_X86_64 + cglobal flac_enc_lpc_16, 5, 7, 4, 0, res, smp, len, order, coefs, posj, negj + %define length r2d + + movsxd orderq, orderd +%else + cglobal flac_enc_lpc_16, 5, 6, 4, 0, res, smp, posj, order, coefs, negj + %define length r2mp +%endif + +; Is it worth looping correctly over the first samples? The most that ever need +; to be copied is 32 so we might as well just unroll the loop and do all 32. +%assign iter 0 +%rep 32/(mmsize/4) + movu m0, [smpq+iter] + movu [resq+iter], m0 + %assign iter iter+mmsize +%endrep + +lea resq, [resq+orderq*4] +lea smpq, [smpq+orderq*4] +lea coefsq, [coefsq+orderq*4] +sub length, orderd +movd m3, r5m +neg orderq + +.looplen: + pxor m0, m0 + mov posjq, orderq + xor negjq, negjq + + .looporder: + movd m2, [coefsq+posjq*4] ; c = coefs[j] + SPLATD m2 + movu m1, [smpq+negjq*4-4] ; s = smp[i-j-1] + pmulld m1, m2 + paddd m0, m1 ; p += c * s + + dec negjq + inc posjq + jnz .looporder + + psrad m0, m3 ; p >>= shift + movu m1, [smpq] + psubd m1, m0 ; smp[i] - p + movu [resq], m1 ; res[i] = smp[i] - (p >> shift) + + add resq, mmsize + add smpq, mmsize + sub length, mmsize/4 +jg .looplen +RET diff --git a/libavcodec/x86/flacdsp_init.c b/libavcodec/x86/flacdsp_init.c index a071b3d..ad88e5b 100644 --- a/libavcodec/x86/flacdsp_init.c +++ b/libavcodec/x86/flacdsp_init.c @@ -27,6 +27,8 @@ void ff_flac_lpc_32_sse4(int32_t *samples, const int coeffs[32], int order, void ff_flac_lpc_32_xop(int32_t *samples, const int coeffs[32], int order, int qlevel, int len); +void ff_flac_enc_lpc_16_sse4(int32_t *, const int32_t *, int, int, const int32_t *,int); + av_cold void ff_flacdsp_init_x86(FLACDSPContext *c, enum AVSampleFormat fmt, int bps) { @@ -36,6 +38,8 @@ av_cold void ff_flacdsp_init_x86(FLACDSPContext *c, enum AVSampleFormat fmt, if (EXTERNAL_SSE4(cpu_flags)) { if (bps > 16 && CONFIG_FLAC_DECODER) c->lpc = ff_flac_lpc_32_sse4; + if (bps == 16 && CONFIG_FLAC_ENCODER && CONFIG_GPL) + c->lpc_encode = ff_flac_enc_lpc_16_sse4; } if (EXTERNAL_XOP(cpu_flags)) { if (bps > 16 && CONFIG_FLAC_DECODER) -- 1.7.9

3 28

One pass volume normalization (ebur128)
by Jan Ehrhardt 15 Jul '14

15 Jul '14

I am once again proposing a patch for one pass volume normalization base on ebur128, as I see this still did not make it into FFMpeg 2.0. My patch is heaveily based on Clément Bsch's proposal in http://permalink.gmane.org/gmane.comp.video.ffmpeg.devel/159978 We have been using this patch now for more than 4 months and 1800+ videos of approximately 1 hour have been transcoded with it. Part of our FFMpeg commandline reads as -filter_complex \ "[0:v]setpts=PTS-STARTPTS[v0];[0:a]asetpts=PTS-STARTPTS,ebur128=metadata=1,volume=metadata=lavfi.r128.I,ebur128[a0]" \ -map [v0] -map [a0] It uses the already present ebur128 meta injection to adjust the volume on the fly. What would be the objection to move this into the FFMpeg core, so I do not have to patch my FFMpeg every time I compile a new one? I applied the patch below to FFMpeg Release/v.2.0. Jan diff --git a/libavfilter/af_volume.c b/libavfilter/af_volume.c index a2ac1e2..6372bb2 100644 --- a/libavfilter/af_volume.c +++ b/libavfilter/af_volume.c @@ -51,18 +51,24 @@ static const AVOption volume_options[] = { { "fixed", "select 8-bit fixed-point", 0, AV_OPT_TYPE_CONST, { .i64 = PRECISION_FIXED }, INT_MIN, INT_MAX, A|F, "precision" }, { "float", "select 32-bit floating-point", 0, AV_OPT_TYPE_CONST, { .i64 = PRECISION_FLOAT }, INT_MIN, INT_MAX, A|F, "precision" }, { "double", "select 64-bit floating-point", 0, AV_OPT_TYPE_CONST, { .i64 = PRECISION_DOUBLE }, INT_MIN, INT_MAX, A|F, "precision" }, + { "metadata", "set the metadata key for loudness normalization", OFFSET(metadata), AV_OPT_TYPE_STRING, { .str = NULL }, .flags = A|F }, { NULL }, }; AVFILTER_DEFINE_CLASS(volume); +static void set_fixed_volume(VolumeContext *vol, double volume) +{ + vol->volume_i = (int)(volume * 256 + 0.5); + vol->volume = vol->volume_i / 256.0; +} + static av_cold int init(AVFilterContext *ctx) { VolumeContext *vol = ctx->priv; if (vol->precision == PRECISION_FIXED) { - vol->volume_i = (int)(vol->volume * 256 + 0.5); - vol->volume = vol->volume_i / 256.0; + set_fixed_volume(vol, vol->volume); av_log(ctx, AV_LOG_VERBOSE, "volume:(%d/256)(%f)(%1.2fdB) precision:fixed\n", vol->volume_i, vol->volume, 20.0*log(vol->volume)/M_LN10); } else { @@ -171,13 +177,13 @@ static av_cold void volume_init(VolumeContext *vol) switch (av_get_packed_sample_fmt(vol->sample_fmt)) { case AV_SAMPLE_FMT_U8: - if (vol->volume_i < 0x1000000) + if (vol->volume_i < 0x1000000 && !vol->metadata) vol->scale_samples = scale_samples_u8_small; else vol->scale_samples = scale_samples_u8; break; case AV_SAMPLE_FMT_S16: - if (vol->volume_i < 0x10000) + if (vol->volume_i < 0x10000 && !vol->metadata) vol->scale_samples = scale_samples_s16_small; else vol->scale_samples = scale_samples_s16; @@ -216,11 +222,30 @@ static int config_output(AVFilterLink *outlink) static int filter_frame(AVFilterLink *inlink, AVFrame *buf) { - VolumeContext *vol = inlink->dst->priv; - AVFilterLink *outlink = inlink->dst->outputs[0]; + AVFilterContext *ctx = inlink->dst; + VolumeContext *vol = ctx->priv; + AVFilterLink *outlink = ctx->outputs[0]; int nb_samples = buf->nb_samples; AVFrame *out_buf; + if (vol->metadata) { + double loudness, new_volume, timestamp, mx; + AVDictionaryEntry *e; + mx = 20; + timestamp = (float)(1.0 * buf->pts / outlink->sample_rate); + mx = fmin(mx, timestamp); + e = av_dict_get(buf->metadata, vol->metadata, NULL, 0); + if (e) { + loudness = av_strtod(e->value, NULL); + if (loudness > -69) { + new_volume = fmax(-mx,fmin(mx,(-23 - loudness))); + av_log(NULL, AV_LOG_VERBOSE, "loudness=%f => %f => volume=%f\n", + loudness, new_volume, pow(10, new_volume / 20)); + set_fixed_volume(vol, pow(10, new_volume / 20)); + } + } + } + if (vol->volume == 1.0 || vol->volume_i == 256) return ff_filter_frame(outlink, buf); @@ -269,6 +294,12 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *buf) return ff_filter_frame(outlink, out_buf); } +static av_cold void uninit(AVFilterContext *ctx) +{ + VolumeContext *vol = ctx->priv; + av_opt_free(vol); +} + static const AVFilterPad avfilter_af_volume_inputs[] = { { .name = "default", @@ -294,6 +325,7 @@ AVFilter avfilter_af_volume = { .priv_size = sizeof(VolumeContext), .priv_class = &volume_class, .init = init, + .uninit = uninit, .inputs = avfilter_af_volume_inputs, .outputs = avfilter_af_volume_outputs, .flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC, diff --git a/libavfilter/af_volume.h b/libavfilter/af_volume.h index bd7932e..4deca9c 100644 --- a/libavfilter/af_volume.h +++ b/libavfilter/af_volume.h @@ -48,6 +48,7 @@ typedef struct VolumeContext { void (*scale_samples)(uint8_t *dst, const uint8_t *src, int nb_samples, int volume); int samples_align; + char *metadata; } VolumeContext; void ff_volume_init_x86(VolumeContext *vol); diff --git a/libavfilter/f_ebur128.c b/libavfilter/f_ebur128.c index 88d37e8..f4ce6d9 100644 --- a/libavfilter/f_ebur128.c +++ b/libavfilter/f_ebur128.c @@ -410,7 +410,7 @@ static av_cold int init(AVFilterContext *ctx) if (ebur128->loglevel != AV_LOG_INFO && ebur128->loglevel != AV_LOG_VERBOSE) { - if (ebur128->do_video || ebur128->metadata) + if (ebur128->do_video) ebur128->loglevel = AV_LOG_VERBOSE; else ebur128->loglevel = AV_LOG_INFO; @@ -689,7 +689,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *insamples) SET_META("LRA.high", ebur128->lra_high); } - av_log(ctx, ebur128->loglevel, "t: %-10s " LOG_FMT "\n", + av_log(ctx, ebur128->metadata || !ebur128->do_video ? AV_LOG_VERBOSE : ebur128->loglevel, "t: %-10s " LOG_FMT "\n", av_ts2timestr(pts, &outlink->time_base), loudness_400, loudness_3000, ebur128->integrated_loudness, ebur128->loudness_range); diff --git a/libavfilter/x86/af_volume_init.c b/libavfilter/x86/af_volume_init.c index 81d605f..fab5a03 100644 --- a/libavfilter/x86/af_volume_init.c +++ b/libavfilter/x86/af_volume_init.c @@ -39,7 +39,7 @@ av_cold void ff_volume_init_x86(VolumeContext *vol) enum AVSampleFormat sample_fmt = av_get_packed_sample_fmt(vol->sample_fmt); if (sample_fmt == AV_SAMPLE_FMT_S16) { - if (EXTERNAL_SSE2(mm_flags) && vol->volume_i < 32768) { + if (EXTERNAL_SSE2(mm_flags) && vol->volume_i < 32768 && !vol->metadata) { vol->scale_samples = ff_scale_samples_s16_sse2; vol->samples_align = 8; }

9 46

[RFC] Browser remote content API
by Lukasz Marek 05 Jul '14

05 Jul '14

Hi, I have an idea of new API, hopely useful. FFmpeg allows to use content on remote servers via network protocols (http, ftp, sftp). It can be read, written, played, probed etc, but in case user want to make application with some UI that displays that content (file names, dirs) then they need to implement it aside. It would be nice to extend URLProtocol with some kind of "readdir" callback. Furhtermore, it would be nice to have single function that read media related metadata like resolution, duration, codec,maybe thumbnail etc. (copy paste from ffprobe can be probably done). Maybe this is a project for GSoC? -- Best Regards, Lukasz Marek Insanity: doing the same thing over and over again and expecting different results. - Albert Einstein

5 11

[PATCH] examples/demuxing_decoding: Fix invalidd API usage
by Derek Buitenhuis 19 Jun '14

19 Jun '14

We are not technically allowed to reuse the codec context found inside the AVStream in the format context. Signed-off-by: Derek Buitenhuis <derek.buitenhuis(a)gmail.com> --- doc/examples/demuxing_decoding.c | 37 ++++++++++++++++++++++++++----------- 1 file changed, 26 insertions(+), 11 deletions(-) diff --git a/doc/examples/demuxing_decoding.c b/doc/examples/demuxing_decoding.c index 0ef5df7..c69e07b 100644 --- a/doc/examples/demuxing_decoding.c +++ b/doc/examples/demuxing_decoding.c @@ -135,12 +135,11 @@ static int decode_packet(int *got_frame, int cached) return decoded; } -static int open_codec_context(int *stream_idx, - AVFormatContext *fmt_ctx, enum AVMediaType type) +static int open_codec_context(int *stream_idx, AVFormatContext *fmt_ctx, + enum AVMediaType type, AVCodecContext **dec_ctx) { int ret; AVStream *st; - AVCodecContext *dec_ctx = NULL; AVCodec *dec = NULL; AVDictionary *opts = NULL; @@ -153,19 +152,37 @@ static int open_codec_context(int *stream_idx, *stream_idx = ret; st = fmt_ctx->streams[*stream_idx]; - /* find decoder for the stream */ - dec_ctx = st->codec; - dec = avcodec_find_decoder(dec_ctx->codec_id); + /* find decoder for the stream as a sanity check */ + dec = avcodec_find_decoder(st->codec->codec_id); if (!dec) { fprintf(stderr, "Failed to find %s codec\n", av_get_media_type_string(type)); return AVERROR(EINVAL); } + /* + * Allocate our decoding context, since we are not allowed + * to reuse the one from the AVStream + */ + *dec_ctx = avcodec_alloc_context3(NULL); + if (!*dec_ctx) { + fprintf(stderr, "Failed to allocate %s codec context.\n", + av_get_media_type_string(type)); + return AVERROR(ENOMEM); + } + + /* Copy the properties over */ + ret = avcodec_copy_context(*dec_ctx, st->codec); + if (ret < 0) { + fprintf(stderr, "Failed to copy %s codec context.\n", + av_get_media_type_string(type)); + return AVERROR(EINVAL); + } + /* Init the decoders, with or without reference counting */ if (api_mode == API_MODE_NEW_API_REF_COUNT) av_dict_set(&opts, "refcounted_frames", "1", 0); - if ((ret = avcodec_open2(dec_ctx, dec, &opts)) < 0) { + if ((ret = avcodec_open2(*dec_ctx, dec, &opts)) < 0) { fprintf(stderr, "Failed to open %s codec\n", av_get_media_type_string(type)); return ret; @@ -252,9 +269,8 @@ int main (int argc, char **argv) exit(1); } - if (open_codec_context(&video_stream_idx, fmt_ctx, AVMEDIA_TYPE_VIDEO) >= 0) { + if (open_codec_context(&video_stream_idx, fmt_ctx, AVMEDIA_TYPE_VIDEO, &video_dec_ctx) >= 0) { video_stream = fmt_ctx->streams[video_stream_idx]; - video_dec_ctx = video_stream->codec; video_dst_file = fopen(video_dst_filename, "wb"); if (!video_dst_file) { @@ -274,9 +290,8 @@ int main (int argc, char **argv) video_dst_bufsize = ret; } - if (open_codec_context(&audio_stream_idx, fmt_ctx, AVMEDIA_TYPE_AUDIO) >= 0) { + if (open_codec_context(&audio_stream_idx, fmt_ctx, AVMEDIA_TYPE_AUDIO, &audio_dec_ctx) >= 0) { audio_stream = fmt_ctx->streams[audio_stream_idx]; - audio_dec_ctx = audio_stream->codec; audio_dst_file = fopen(audio_dst_filename, "wb"); if (!audio_dst_file) { fprintf(stderr, "Could not open destination file %s\n", video_dst_filename); -- 1.9.rc1

3 7

drop entire frame when RTP packets are lost
by Martin Carroll 12 Jun '14

12 Jun '14

[Please forgive me if this issue has already been discussed -- I am new to ffmpeg-devel. I could not, however, find any discussion on this topic on this list in the past year, so here goes.] In the current design of ffplay/ffmpeg, when an RTP packet is lost, the containing -- and incomplete -- encoded frame is still passed along to the decoder. I have implemented a patch that has the read thread keep track of how many packets are missed, and when any are missed, the entire containing frame is dropped (that is, never makes it out of the read thread). I would be happy to send this patch upstream, or discuss it further if anyone likes. (The patch touches six files and about 25 lines of code.) Martin Carroll

4 20