Merge remote-tracking branch 'qatar/master'

* qatar/master:
  doc: Improve references to external URLs.
  h264: move decode_mb_skip() from h264.h to h.264_mvpred.h
  ffplay: skip return value of avcodec_decode_video2 / avcodec_decode_subtitle2
  dnxhdenc: Replace a forward declaration by the proper #include.
  h264: move h264_mvpred.h include.
  pix_fmt: Fix number of bits per component in yuv444p9be
  lavf: deprecate AVFormatContext.timestamp
  ffmpeg: merge input_files_ts_scale into InputStream.
  ffmpeg: don't abuse a global for passing sample format from input to output
  ffmpeg: don't abuse a global for passing channel layout from input to output
  ffmpeg: factor common code from new_a/v/s/d_stream to new_output_stream()
  matroskaenc: make SSA default subtitle codec.
  oggdec: prevent heap corruption.

Conflicts:
	doc/developer.texi
	doc/faq.texi
	doc/general.texi
	ffmpeg.c
	ffplay.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

19 files changed, 187 insertions, 152 deletions

diff --git a/doc/developer.texi b/doc/developer.texi
index 7b22abf02f..f7fc6d652c 100644
--- a/doc/developer.texi
+++ b/doc/developer.texi
@@ -34,6 +34,7 @@ You can use libavcodec or libavformat in your commercial program, but
 @emph{any patch you make must be published}. The best way to proceed is
 to send your patches to the FFmpeg mailing list.
 
+
 @anchor{Coding Rules}
 @section Coding Rules
 
diff --git a/doc/faq.texi b/doc/faq.texi
index 225f139b3c..535ca31446 100644
--- a/doc/faq.texi
+++ b/doc/faq.texi
@@ -47,7 +47,7 @@ Likely reasons
 @item We are busy and haven't had time yet to read your report or
 investigate the issue.
 @item You didn't follow @url{http://ffmpeg.org/bugreports.html}.
-@item You didn't use git HEAD.
+@item You didn't use git master.
 @item You reported a segmentation fault without gdb output.
 @item You describe a problem but not how to reproduce it.
 @item It's unclear if you use ffmpeg as command line tool or use
@@ -123,7 +123,8 @@ problem and an NP-hard problem...
 
 @section ffmpeg does not work; what is wrong?
 
-Try a @code{make distclean} in the ffmpeg source directory before the build. If this does not help see
+Try a @code{make distclean} in the ffmpeg source directory before the build.
+If this does not help see
 (@url{http://ffmpeg.org/bugreports.html}).
 
 @section How do I encode single pictures into movies?
@@ -285,7 +286,8 @@ Just create an "input.avs" text file with this single line ...
   ffmpeg -i input.avs
 @end example
 
-For ANY other help on Avisynth, please visit @url{http://www.avisynth.org/}.
+For ANY other help on Avisynth, please visit the
+@uref{http://www.avisynth.org/, Avisynth homepage}.
 
 @section How can I join video files?
 
@@ -417,7 +419,7 @@ No. These tools are too bloated and they complicate the build.
 FFmpeg is already organized in a highly modular manner and does not need to
 be rewritten in a formal object language. Further, many of the developers
 favor straight C; it works for them. For more arguments on this matter,
-read "Programming Religion" at (@url{http://www.tux.org/lkml/#s15}).
+read @uref{http://www.tux.org/lkml/#s15, "Programming Religion"}.
 
 @section Why are the ffmpeg programs devoid of debugging symbols?
 
diff --git a/doc/ffmpeg.texi b/doc/ffmpeg.texi
index 2b08998b9b..0a6c75805c 100644
--- a/doc/ffmpeg.texi
+++ b/doc/ffmpeg.texi
@@ -888,8 +888,8 @@ ffmpeg -f oss -i /dev/dsp -f video4linux2 -i /dev/video0 /tmp/out.mpg
 @end example
 
 Note that you must activate the right video source and channel before
-launching ffmpeg with any TV viewer such as xawtv
-(@url{http://linux.bytesex.org/xawtv/}) by Gerd Knorr. You also
+launching ffmpeg with any TV viewer such as
+@uref{http://linux.bytesex.org/xawtv/, xawtv} by Gerd Knorr. You also
 have to set the audio recording levels correctly with a
 standard mixer.
 
diff --git a/doc/general.texi b/doc/general.texi
index f44e57359c..a52ce69a13 100644
--- a/doc/general.texi
+++ b/doc/general.texi
@@ -848,7 +848,7 @@ noticeable when running make for a second time (for example in
 @code{make install}).
 
 @item In order to compile FFplay, you must have the MinGW development library
-of SDL. Get it from @url{http://www.libsdl.org}.
+of @uref{http://www.libsdl.org/, SDL}.
 Edit the @file{bin/sdl-config} script so that it points to the correct prefix
 where SDL was installed. Verify that @file{sdl-config} can be launched from
 the MSYS command line.
@@ -1044,8 +1044,7 @@ Then configure FFmpeg with the following options:
 (you can change the cross-prefix according to the prefix chosen for the
 MinGW tools).
 
-Then you can easily test FFmpeg with Wine
-(@url{http://www.winehq.com/}).
+Then you can easily test FFmpeg with @uref{http://www.winehq.com/, Wine}.
 
 @subsection Compilation under Cygwin
 
@@ -1084,8 +1083,8 @@ If you want to build FFmpeg with additional libraries, download Cygwin
 libogg-devel, libvorbis-devel
 @end example
 
-These library packages are only available from Cygwin Ports
-(@url{http://sourceware.org/cygwinports/}) :
+These library packages are only available from
+@uref{http://sourceware.org/cygwinports/, Cygwin Ports}:
 
 @example
 yasm, libSDL-devel, libdirac-devel, libfaac-devel, libgsm-devel,
diff --git a/doc/protocols.texi b/doc/protocols.texi
index a71c262368..991ec888af 100644
--- a/doc/protocols.texi
+++ b/doc/protocols.texi
@@ -242,7 +242,7 @@ data transferred over RDT).
 
 The muxer can be used to send a stream using RTSP ANNOUNCE to a server
 supporting it (currently Darwin Streaming Server and Mischa Spiegelmock's
-RTSP server, @url{http://github.com/revmischa/rtsp-server}).
+@uref{http://github.com/revmischa/rtsp-server, RTSP server}).
 
 The required syntax for a RTSP url is:
 @example
diff --git a/ffmpeg.c b/ffmpeg.c
index 0da0a1f930..c4729df87b 100644
--- a/ffmpeg.c
+++ b/ffmpeg.c
@@ -117,8 +117,8 @@ static const OptionDef options[];
 #define MAX_STREAMS 1024    /* arbitrary sanity check value */
 
 static const char *last_asked_format = NULL;
-static double *input_files_ts_scale[MAX_FILES] = {NULL};
-static int nb_input_files_ts_scale[MAX_FILES] = {0};
+static double *ts_scale;
+static int  nb_ts_scale;
 
 static AVFormatContext *output_files[MAX_FILES];
 static int nb_output_files = 0;
@@ -171,7 +171,6 @@ static char *vfilters = NULL;
 
 static int intra_only = 0;
 static int audio_sample_rate = 0;
-static int64_t channel_layout = 0;
 #define QSCALE_NONE -99999
 static float audio_qscale = QSCALE_NONE;
 static int audio_disable = 0;
@@ -194,7 +193,6 @@ static float mux_max_delay= 0.7;
 
 static int64_t recording_time = INT64_MAX;
 static int64_t start_time = 0;
-static int64_t recording_timestamp = 0;
 static int64_t input_ts_offset = 0;
 static int file_overwrite = 0;
 static AVDictionary *metadata;
@@ -327,6 +325,7 @@ typedef struct InputStream {
     int64_t       next_pts;  /* synthetic pts for cases where pkt.pts
                                 is not defined */
     int64_t       pts;       /* current pts */
+    double ts_scale;
     int is_start;            /* is 1 at the start and after a discontinuity */
     int showed_multi_packet_warning;
     int is_past_recording_time;
@@ -538,7 +537,6 @@ static int ffmpeg_exit(int ret)
     }
     for(i=0;i<nb_input_files;i++) {
         av_close_input_file(input_files[i].ctx);
-        av_free(input_files_ts_scale[i]);
     }
 
     av_free(intra_matrix);
@@ -671,10 +669,16 @@ static void choose_pixel_fmt(AVStream *st, AVCodec *codec)
     }
 }
 
-static OutputStream *new_output_stream(AVFormatContext *oc, int file_idx)
+static OutputStream *new_output_stream(AVFormatContext *oc, int file_idx, AVCodec *codec)
 {
-    int idx = oc->nb_streams - 1;
     OutputStream *ost;
+    AVStream *st = av_new_stream(oc, oc->nb_streams < nb_streamid_map ? streamid_map[oc->nb_streams] : 0);
+    int idx      = oc->nb_streams - 1;
+
+    if (!st) {
+        av_log(NULL, AV_LOG_ERROR, "Could not alloc stream.\n");
+        ffmpeg_exit(1);
+    }
 
     output_streams_for_file[file_idx] =
         grow_array(output_streams_for_file[file_idx],
@@ -689,6 +693,10 @@ static OutputStream *new_output_stream(AVFormatContext *oc, int file_idx)
     }
     ost->file_index = file_idx;
     ost->index = idx;
+    ost->st    = st;
+    ost->enc   = codec;
+
+    avcodec_get_context_defaults3(st->codec, codec);
 
     ost->sws_flags = av_get_int(sws_opts, "sws_flags", NULL);
     return ost;
@@ -704,28 +712,21 @@ static int read_ffserver_streams(AVFormatContext *s, const char *filename)
     if (err < 0)
         return err;
     /* copy stream format */
-    s->nb_streams = 0;
-    s->streams = av_mallocz(sizeof(AVStream *) * ic->nb_streams);
     for(i=0;i<ic->nb_streams;i++) {
         AVStream *st;
+        OutputStream *ost;
         AVCodec *codec;
 
-        s->nb_streams++;
+        codec = avcodec_find_encoder(ic->streams[i]->codec->codec_id);
+        ost   = new_output_stream(s, nb_output_files, codec);
+        st    = ost->st;
 
         // FIXME: a more elegant solution is needed
-        st = av_mallocz(sizeof(AVStream));
         memcpy(st, ic->streams[i], sizeof(AVStream));
         st->info = av_malloc(sizeof(*st->info));
         memcpy(st->info, ic->streams[i]->info, sizeof(*st->info));
-        st->codec = avcodec_alloc_context();
-        if (!st->codec) {
-            print_error(filename, AVERROR(ENOMEM));
-            ffmpeg_exit(1);
-        }
         avcodec_copy_context(st->codec, ic->streams[i]->codec);
-        s->streams[i] = st;
 
-        codec = avcodec_find_encoder(st->codec->codec_id);
         if (st->codec->codec_type == AVMEDIA_TYPE_AUDIO) {
             if (audio_stream_copy) {
                 st->stream_copy = 1;
@@ -740,13 +741,8 @@ static int read_ffserver_streams(AVFormatContext *s, const char *filename)
 
         if(st->codec->flags & CODEC_FLAG_BITEXACT)
             nopts = 1;
-
-        new_output_stream(s, nb_output_files);
     }
 
-    if (!nopts)
-        s->timestamp = av_gettime();
-
     av_close_input_file(ic);
     return 0;
 }
@@ -2102,7 +2098,6 @@ static int transcode(AVFormatContext **output_files,
         for(i=0;i<os->nb_streams;i++,n++) {
             int found;
             ost = ost_table[n] = output_streams_for_file[k][i];
-            ost->st = os->streams[i];
             if (nb_stream_maps > 0) {
                 ost->source_index = input_files[stream_maps[n].file_index].ist_index +
                     stream_maps[n].stream_index;
@@ -2278,6 +2273,9 @@ static int transcode(AVFormatContext **output_files,
                 }
                 choose_sample_rate(ost->st, ost->enc);
                 codec->time_base = (AVRational){1, codec->sample_rate};
+                if (codec->sample_fmt == AV_SAMPLE_FMT_NONE)
+                    codec->sample_fmt = icodec->sample_fmt;
+                choose_sample_fmt(ost->st, ost->enc);
                 if (!codec->channels) {
                     codec->channels = icodec->channels;
                     codec->channel_layout = icodec->channel_layout;
@@ -2738,12 +2736,11 @@ static int transcode(AVFormatContext **output_files,
         if (pkt.pts != AV_NOPTS_VALUE)
             pkt.pts += av_rescale_q(input_files[ist->file_index].ts_offset, AV_TIME_BASE_Q, ist->st->time_base);
 
-        if (pkt.stream_index < nb_input_files_ts_scale[file_index]
-            && input_files_ts_scale[file_index][pkt.stream_index]){
+        if (ist->ts_scale) {
             if(pkt.pts != AV_NOPTS_VALUE)
-                pkt.pts *= input_files_ts_scale[file_index][pkt.stream_index];
+                pkt.pts *= ist->ts_scale;
             if(pkt.dts != AV_NOPTS_VALUE)
-                pkt.dts *= input_files_ts_scale[file_index][pkt.stream_index];
+                pkt.dts *= ist->ts_scale;
         }
 
 //        fprintf(stderr, "next:%"PRId64" dts:%"PRId64" off:%"PRId64" %d\n", ist->next_pts, pkt.dts, input_files[ist->file_index].ts_offset, ist->st->codec->codec_type);
@@ -3203,8 +3200,8 @@ static int opt_input_ts_scale(const char *opt, const char *arg)
     if(stream >= MAX_STREAMS)
         ffmpeg_exit(1);
 
-    input_files_ts_scale[nb_input_files] = grow_array(input_files_ts_scale[nb_input_files], sizeof(*input_files_ts_scale[nb_input_files]), &nb_input_files_ts_scale[nb_input_files], stream + 1);
-    input_files_ts_scale[nb_input_files][stream]= scale;
+    ts_scale = grow_array(ts_scale, sizeof(*ts_scale), &nb_ts_scale, stream + 1);
+    ts_scale[stream] = scale;
     return 0;
 }
 
@@ -3222,7 +3219,14 @@ static int opt_start_time(const char *opt, const char *arg)
 
 static int opt_recording_timestamp(const char *opt, const char *arg)
 {
-    recording_timestamp = parse_time_or_die(opt, arg, 0) / 1000000;
+    char buf[128];
+    int64_t recording_timestamp = parse_time_or_die(opt, arg, 0) / 1E6;
+    struct tm time = *gmtime((time_t*)&recording_timestamp);
+    strftime(buf, sizeof(buf), "creation_time=%FT%T%z", &time);
+    opt_metadata("metadata", buf);
+
+    av_log(NULL, AV_LOG_WARNING, "%s is deprecated, set the 'creation_time' metadata "
+                                 "tag instead.\n", opt);
     return 0;
 }
 
@@ -3401,14 +3405,15 @@ static int opt_input_file(const char *opt, const char *filename)
         ist->file_index = nb_input_files;
         ist->discard = 1;
 
+        if (i < nb_ts_scale)
+            ist->ts_scale = ts_scale[i];
+
         switch (dec->codec_type) {
         case AVMEDIA_TYPE_AUDIO:
             ist->dec = avcodec_find_decoder_by_name(audio_codec_name);
             if(!ist->dec)
                 ist->dec = avcodec_find_decoder(dec->codec_id);
             set_context_opts(dec, avcodec_opts[AVMEDIA_TYPE_AUDIO], AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_DECODING_PARAM, ist->dec);
-            channel_layout    = dec->channel_layout;
-            audio_sample_fmt  = dec->sample_fmt;
             if(audio_disable)
                 st->discard= AVDISCARD_ALL;
             break;
@@ -3475,6 +3480,9 @@ static int opt_input_file(const char *opt, const char *filename)
     frame_width  = 0;
     audio_sample_rate = 0;
     audio_channels    = 0;
+    audio_sample_fmt  = AV_SAMPLE_FMT_NONE;
+    av_freep(&ts_scale);
+    nb_ts_scale = 0;
 
     av_freep(&video_codec_name);
     av_freep(&audio_codec_name);
@@ -3535,23 +3543,20 @@ static void new_video_stream(AVFormatContext *oc, int file_idx)
     enum CodecID codec_id = CODEC_ID_NONE;
     AVCodec *codec= NULL;
 
-    st = av_new_stream(oc, oc->nb_streams < nb_streamid_map ? streamid_map[oc->nb_streams] : 0);
-    if (!st) {
-        fprintf(stderr, "Could not alloc stream\n");
-        ffmpeg_exit(1);
-    }
-    ost = new_output_stream(oc, file_idx);
-
     if(!video_stream_copy){
         if (video_codec_name) {
             codec_id = find_codec_or_die(video_codec_name, AVMEDIA_TYPE_VIDEO, 1,
                                          avcodec_opts[AVMEDIA_TYPE_VIDEO]->strict_std_compliance);
             codec = avcodec_find_encoder_by_name(video_codec_name);
-            ost->enc = codec;
         } else {
             codec_id = av_guess_codec(oc->oformat, NULL, oc->filename, NULL, AVMEDIA_TYPE_VIDEO);
             codec = avcodec_find_encoder(codec_id);
         }
+    }
+
+    ost = new_output_stream(oc, file_idx, codec);
+    st  = ost->st;
+    if (!video_stream_copy) {
         ost->frame_aspect_ratio = frame_aspect_ratio;
         frame_aspect_ratio = 0;
 #if CONFIG_AVFILTER
@@ -3560,7 +3565,6 @@ static void new_video_stream(AVFormatContext *oc, int file_idx)
 #endif
     }
 
-    avcodec_get_context_defaults3(st->codec, codec);
     ost->bitstream_filters = video_bitstream_filters;
     video_bitstream_filters= NULL;
 
@@ -3674,26 +3678,18 @@ static void new_audio_stream(AVFormatContext *oc, int file_idx)
     AVCodecContext *audio_enc;
     enum CodecID codec_id = CODEC_ID_NONE;
 
-    st = av_new_stream(oc, oc->nb_streams < nb_streamid_map ? streamid_map[oc->nb_streams] : 0);
-    if (!st) {
-        fprintf(stderr, "Could not alloc stream\n");
-        ffmpeg_exit(1);
-    }
-    ost = new_output_stream(oc, file_idx);
-
     if(!audio_stream_copy){
         if (audio_codec_name) {
             codec_id = find_codec_or_die(audio_codec_name, AVMEDIA_TYPE_AUDIO, 1,
                                          avcodec_opts[AVMEDIA_TYPE_AUDIO]->strict_std_compliance);
             codec = avcodec_find_encoder_by_name(audio_codec_name);
-            ost->enc = codec;
         } else {
             codec_id = av_guess_codec(oc->oformat, NULL, oc->filename, NULL, AVMEDIA_TYPE_AUDIO);
             codec = avcodec_find_encoder(codec_id);
         }
     }
-
-    avcodec_get_context_defaults3(st->codec, codec);
+    ost = new_output_stream(oc, file_idx, codec);
+    st  = ost->st;
 
     ost->bitstream_filters = audio_bitstream_filters;
     audio_bitstream_filters= NULL;
@@ -3722,11 +3718,10 @@ static void new_audio_stream(AVFormatContext *oc, int file_idx)
         }
         if (audio_channels)
             audio_enc->channels = audio_channels;
-        audio_enc->sample_fmt = audio_sample_fmt;
+        if (audio_sample_fmt != AV_SAMPLE_FMT_NONE)
+            audio_enc->sample_fmt = audio_sample_fmt;
         if (audio_sample_rate)
             audio_enc->sample_rate = audio_sample_rate;
-        audio_enc->channel_layout = channel_layout;
-        choose_sample_fmt(st, codec);
     }
     if (audio_language) {
         av_dict_set(&st->metadata, "language", audio_language, 0);
@@ -3742,21 +3737,16 @@ static void new_audio_stream(AVFormatContext *oc, int file_idx)
 static void new_data_stream(AVFormatContext *oc, int file_idx)
 {
     AVStream *st;
-    AVCodec *codec=NULL;
+    OutputStream *ost;
     AVCodecContext *data_enc;
 
-    st = av_new_stream(oc, oc->nb_streams < nb_streamid_map ? streamid_map[oc->nb_streams] : 0);
-    if (!st) {
-        fprintf(stderr, "Could not alloc stream\n");
-        ffmpeg_exit(1);
-    }
-    new_output_stream(oc, file_idx);
+    ost = new_output_stream(oc, file_idx, NULL);
+    st  = ost->st;
     data_enc = st->codec;
     if (!data_stream_copy) {
         fprintf(stderr, "Data stream encoding not supported yet (only streamcopy)\n");
         ffmpeg_exit(1);
     }
-    avcodec_get_context_defaults3(st->codec, codec);
 
     data_enc->codec_type = AVMEDIA_TYPE_DATA;
 
@@ -3784,25 +3774,19 @@ static void new_subtitle_stream(AVFormatContext *oc, int file_idx)
     AVCodecContext *subtitle_enc;
     enum CodecID codec_id = CODEC_ID_NONE;
 
-    st = av_new_stream(oc, oc->nb_streams < nb_streamid_map ? streamid_map[oc->nb_streams] : 0);
-    if (!st) {
-        fprintf(stderr, "Could not alloc stream\n");
-        ffmpeg_exit(1);
-    }
-    ost = new_output_stream(oc, file_idx);
-    subtitle_enc = st->codec;
     if(!subtitle_stream_copy){
         if (subtitle_codec_name) {
             codec_id = find_codec_or_die(subtitle_codec_name, AVMEDIA_TYPE_SUBTITLE, 1,
                                          avcodec_opts[AVMEDIA_TYPE_SUBTITLE]->strict_std_compliance);
             codec = avcodec_find_encoder_by_name(subtitle_codec_name);
-            ost->enc = codec;
         } else {
             codec_id = av_guess_codec(oc->oformat, NULL, oc->filename, NULL, AVMEDIA_TYPE_SUBTITLE);
             codec = avcodec_find_encoder(codec_id);
         }
     }
-    avcodec_get_context_defaults3(st->codec, codec);
+    ost = new_output_stream(oc, file_idx, codec);
+    st  = ost->st;
+    subtitle_enc = st->codec;
 
     ost->bitstream_filters = subtitle_bitstream_filters;
     subtitle_bitstream_filters= NULL;
@@ -3938,8 +3922,6 @@ static int opt_output_file(const char *opt, const char *filename)
         if (use_subtitle) new_subtitle_stream(oc, nb_output_files);
         if (use_data)     new_data_stream(oc, nb_output_files);
 
-        oc->timestamp = recording_timestamp;
-
         av_dict_copy(&oc->metadata, metadata, 0);
         av_dict_free(&metadata);
     }
@@ -4005,6 +3987,7 @@ static int opt_output_file(const char *opt, const char *filename)
     frame_height  = 0;
     audio_sample_rate = 0;
     audio_channels    = 0;
+    audio_sample_fmt  = AV_SAMPLE_FMT_NONE;
 
     av_freep(&forced_key_frames);
     uninit_opts();
diff --git a/ffplay.c b/ffplay.c
index 40e0cc1ef4..b11642ae8e 100644
--- a/ffplay.c
+++ b/ffplay.c
@@ -1431,7 +1431,7 @@ static int queue_picture(VideoState *is, AVFrame *src_frame, double pts1, int64_
 
 static int get_video_frame(VideoState *is, AVFrame *frame, int64_t *pts, AVPacket *pkt)
 {
-    int len1 av_unused, got_picture, i;
+    int got_picture, i;
 
     if (packet_queue_get(&is->videoq, pkt, 1) < 0)
         return -1;
@@ -1458,9 +1458,7 @@ static int get_video_frame(VideoState *is, AVFrame *frame, int64_t *pts, AVPacke
         return 0;
     }
 
-    len1 = avcodec_decode_video2(is->video_st->codec,
-                                 frame, &got_picture,
-                                 pkt);
+    avcodec_decode_video2(is->video_st->codec, frame, &got_picture, pkt);
 
     if (got_picture) {
         if (decoder_reorder_pts == -1) {
@@ -1807,7 +1805,7 @@ static int subtitle_thread(void *arg)
     VideoState *is = arg;
     SubPicture *sp;
     AVPacket pkt1, *pkt = &pkt1;
-    int len1 av_unused, got_subtitle;
+    int got_subtitle;
     double pts;
     int i, j;
     int r, g, b, y, u, v, a;
@@ -1841,9 +1839,9 @@ static int subtitle_thread(void *arg)
         if (pkt->pts != AV_NOPTS_VALUE)
             pts = av_q2d(is->subtitle_st->time_base)*pkt->pts;
 
-        len1 = avcodec_decode_subtitle2(is->subtitle_st->codec,
-                                    &sp->sub, &got_subtitle,
-                                    pkt);
+        avcodec_decode_subtitle2(is->subtitle_st->codec, &sp->sub,
+                                 &got_subtitle, pkt);
+
         if (got_subtitle && sp->sub.format == 0) {
             sp->pts = pts;
 
diff --git a/libavcodec/dnxhdenc.c b/libavcodec/dnxhdenc.c
index 62bc9f0dd5..8b63bbb19b 100644
--- a/libavcodec/dnxhdenc.c
+++ b/libavcodec/dnxhdenc.c
@@ -28,6 +28,7 @@
 #include "avcodec.h"
 #include "dsputil.h"
 #include "mpegvideo.h"
+#include "mpegvideo_common.h"
 #include "dnxhdenc.h"
 
 #define VE AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
@@ -38,8 +39,6 @@ static const AVOption options[]={
 };
 static const AVClass class = { "dnxhd", av_default_item_name, options, LIBAVUTIL_VERSION_INT };
 
-int dct_quantize_c(MpegEncContext *s, DCTELEM *block, int n, int qscale, int *overflow);
-
 #define LAMBDA_FRAC_BITS 10
 
 static av_always_inline void dnxhd_get_pixels_8x4(DCTELEM *restrict block, const uint8_t *pixels, int line_size)
diff --git a/libavcodec/h264.h b/libavcodec/h264.h
index 6190db6458..ee52dee16d 100644
--- a/libavcodec/h264.h
+++ b/libavcodec/h264.h
@@ -770,8 +770,6 @@ static av_always_inline int get_chroma_qp(H264Context *h, int t, int qscale){
     return h->pps.chroma_qp_table[t][qscale];
 }
 
-static av_always_inline void pred_pskip_motion(H264Context * const h);
-
 static void fill_decode_neighbors(H264Context *h, int mb_type){
     MpegEncContext * const s = &h->s;
     const int mb_xy= h->mb_xy;
@@ -1302,45 +1300,4 @@ static av_always_inline int get_dct8x8_allowed(H264Context *h){
         return !(AV_RN64A(h->sub_mb_type) & ((MB_TYPE_16x8|MB_TYPE_8x16|MB_TYPE_8x8|MB_TYPE_DIRECT2)*0x0001000100010001ULL));
 }
 
-/**
- * decodes a P_SKIP or B_SKIP macroblock
- */
-static void av_unused decode_mb_skip(H264Context *h){
-    MpegEncContext * const s = &h->s;
-    const int mb_xy= h->mb_xy;
-    int mb_type=0;
-
-    memset(h->non_zero_count[mb_xy], 0, 48);
-
-    if(MB_FIELD)
-        mb_type|= MB_TYPE_INTERLACED;
-
-    if( h->slice_type_nos == AV_PICTURE_TYPE_B )
-    {
-        // just for fill_caches. pred_direct_motion will set the real mb_type
-        mb_type|= MB_TYPE_L0L1|MB_TYPE_DIRECT2|MB_TYPE_SKIP;
-        if(h->direct_spatial_mv_pred){
-            fill_decode_neighbors(h, mb_type);
-        fill_decode_caches(h, mb_type); //FIXME check what is needed and what not ...
-        }
-        ff_h264_pred_direct_motion(h, &mb_type);
-        mb_type|= MB_TYPE_SKIP;
-    }
-    else
-    {
-        mb_type|= MB_TYPE_16x16|MB_TYPE_P0L0|MB_TYPE_P1L0|MB_TYPE_SKIP;
-
-        fill_decode_neighbors(h, mb_type);
-        pred_pskip_motion(h);
-    }
-
-    write_back_motion(h, mb_type);
-    s->current_picture.f.mb_type[mb_xy]      = mb_type;
-    s->current_picture.f.qscale_table[mb_xy] = s->qscale;
-    h->slice_table[ mb_xy ]= h->slice_num;
-    h->prev_mb_skipped= 1;
-}
-
-#include "h264_mvpred.h" //For pred_pskip_motion()
-
 #endif /* AVCODEC_H264_H */
diff --git a/libavcodec/h264_mvpred.h b/libavcodec/h264_mvpred.h
index 9d91c205c7..227dc4a662 100644
--- a/libavcodec/h264_mvpred.h
+++ b/libavcodec/h264_mvpred.h
@@ -327,4 +327,43 @@ zeromv:
     return;
 }
 
+/**
+ * decodes a P_SKIP or B_SKIP macroblock
+ */
+static void av_unused decode_mb_skip(H264Context *h){
+    MpegEncContext * const s = &h->s;
+    const int mb_xy= h->mb_xy;
+    int mb_type=0;
+
+    memset(h->non_zero_count[mb_xy], 0, 48);
+
+    if(MB_FIELD)
+        mb_type|= MB_TYPE_INTERLACED;
+
+    if( h->slice_type_nos == AV_PICTURE_TYPE_B )
+    {
+        // just for fill_caches. pred_direct_motion will set the real mb_type
+        mb_type|= MB_TYPE_L0L1|MB_TYPE_DIRECT2|MB_TYPE_SKIP;
+        if(h->direct_spatial_mv_pred){
+            fill_decode_neighbors(h, mb_type);
+        fill_decode_caches(h, mb_type); //FIXME check what is needed and what not ...
+        }
+        ff_h264_pred_direct_motion(h, &mb_type);
+        mb_type|= MB_TYPE_SKIP;
+    }
+    else
+    {
+        mb_type|= MB_TYPE_16x16|MB_TYPE_P0L0|MB_TYPE_P1L0|MB_TYPE_SKIP;
+
+        fill_decode_neighbors(h, mb_type);
+        pred_pskip_motion(h);
+    }
+
+    write_back_motion(h, mb_type);
+    s->current_picture.f.mb_type[mb_xy]      = mb_type;
+    s->current_picture.f.qscale_table[mb_xy] = s->qscale;
+    h->slice_table[ mb_xy ]= h->slice_num;
+    h->prev_mb_skipped= 1;
+}
+
 #endif /* AVCODEC_H264_MVPRED_H */
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index a0c2047e61..7c041f6e80 100644
--- a/libavformat/avformat.h
+++ b/libavformat/avformat.h
@@ -698,7 +698,12 @@ typedef struct AVFormatContext {
     AVStream **streams;
     char filename[1024]; /**< input or output filename */
     /* stream info */
-    int64_t timestamp;
+#if FF_API_TIMESTAMP
+    /**
+     * @deprecated use 'creation_time' metadata tag instead
+     */
+    attribute_deprecated int64_t timestamp;
+#endif
 
     int ctx_flags; /**< Format-specific flags, see AVFMTCTX_xx */
     /* private data for pts handling (do not modify directly). */
diff --git a/libavformat/dvenc.c b/libavformat/dvenc.c
index 0cbe7a5484..3b2a8eb711 100644
--- a/libavformat/dvenc.c
+++ b/libavformat/dvenc.c
@@ -43,7 +43,7 @@ struct DVMuxContext {
     AVStream         *ast[2];        /* stereo audio streams */
     AVFifoBuffer     *audio_data[2]; /* FIFO for storing excessive amounts of PCM */
     int               frames;        /* current frame number */
-    time_t            start_time;    /* recording start time */
+    int64_t           start_time;    /* recording start time */
     int               has_audio;     /* frame under contruction has audio */
     int               has_video;     /* frame under contruction has video */
     uint8_t           frame_buf[DV_MAX_FRAME_SIZE]; /* frame under contruction */
@@ -290,6 +290,7 @@ static DVMuxContext* dv_init_mux(AVFormatContext* s)
 {
     DVMuxContext *c = s->priv_data;
     AVStream *vst = NULL;
+    AVDictionaryEntry *t;
     int i;
 
     /* we support at most 1 video and 2 audio streams */
@@ -337,7 +338,16 @@ static DVMuxContext* dv_init_mux(AVFormatContext* s)
     c->frames     = 0;
     c->has_audio  = 0;
     c->has_video  = 0;
-    c->start_time = (time_t)s->timestamp;
+#if FF_API_TIMESTAMP
+    if (s->timestamp)
+        c->start_time = s->timestamp;
+    else
+#endif
+    if (t = av_dict_get(s->metadata, "creation_time", NULL, 0)) {
+        struct tm time = {0};
+        strptime(t->value, "%Y - %m - %dT%T", &time);
+        c->start_time = mktime(&time);
+    }
 
     for (i=0; i < c->n_ast; i++) {
         if (c->ast[i] && !(c->audio_data[i]=av_fifo_alloc(100*AVCODEC_MAX_AUDIO_FRAME_SIZE))) {
diff --git a/libavformat/gxfenc.c b/libavformat/gxfenc.c
index da25ddcc2a..62ee4eb751 100644
--- a/libavformat/gxfenc.c
+++ b/libavformat/gxfenc.c
@@ -394,6 +394,20 @@ static int gxf_write_umf_material_description(AVFormatContext *s)
     GXFContext *gxf = s->priv_data;
     AVIOContext *pb = s->pb;
     int timecode_base = gxf->time_base.den == 60000 ? 60 : 50;
+    int64_t timestamp = 0;
+    AVDictionaryEntry *t;
+
+#if FF_API_TIMESTAMP
+    if (s->timestamp)
+        timestamp = s->timestamp;
+    else
+#endif
+    if (t = av_dict_get(s->metadata, "creation_time", NULL, 0)) {
+        struct tm time = {0};
+        strptime(t->value, "%Y - %m - %dT%T", &time);
+        timestamp = mktime(&time);
+    }
+
 
     // XXX drop frame
     uint32_t timecode =
@@ -409,8 +423,8 @@ static int gxf_write_umf_material_description(AVFormatContext *s)
     avio_wl32(pb, gxf->nb_fields); /* mark out */
     avio_wl32(pb, 0); /* timecode mark in */
     avio_wl32(pb, timecode); /* timecode mark out */
-    avio_wl64(pb, s->timestamp); /* modification time */
-    avio_wl64(pb, s->timestamp); /* creation time */
+    avio_wl64(pb, timestamp); /* modification time */
+    avio_wl64(pb, timestamp); /* creation time */
     avio_wl16(pb, 0); /* reserved */
     avio_wl16(pb, 0); /* reserved */
     avio_wl16(pb, gxf->audio_tracks);
diff --git a/libavformat/matroskaenc.c b/libavformat/matroskaenc.c
index 4bfd197c52..063e8320db 100644
--- a/libavformat/matroskaenc.c
+++ b/libavformat/matroskaenc.c
@@ -1212,7 +1212,7 @@ AVOutputFormat ff_matroska_muxer = {
     mkv_write_trailer,
     .flags = AVFMT_GLOBALHEADER | AVFMT_VARIABLE_FPS,
     .codec_tag = (const AVCodecTag* const []){ff_codec_bmp_tags, ff_codec_wav_tags, 0},
-    .subtitle_codec = CODEC_ID_TEXT,
+    .subtitle_codec = CODEC_ID_SSA,
 };
 #endif
 
diff --git a/libavformat/movenc.c b/libavformat/movenc.c
index b0cdcbc87c..41c59d49c1 100644
--- a/libavformat/movenc.c
+++ b/libavformat/movenc.c
@@ -2142,6 +2142,7 @@ static int mov_write_header(AVFormatContext *s)
 {
     AVIOContext *pb = s->pb;
     MOVMuxContext *mov = s->priv_data;
+    AVDictionaryEntry *t;
     int i, hint_track = 0;
 
     if (!s->pb->seekable) {
@@ -2272,7 +2273,18 @@ static int mov_write_header(AVFormatContext *s)
     }
 
     mov_write_mdat_tag(pb, mov);
-    mov->time = s->timestamp + 0x7C25B080; //1970 based -> 1904 based
+
+#if FF_API_TIMESTAMP
+    if (s->timestamp)
+        mov->time = s->timestamp;
+    else
+#endif
+    if (t = av_dict_get(s->metadata, "creation_time", NULL, 0)) {
+        struct tm time = {0};
+        strptime(t->value, "%Y - %m - %dT%T", &time);
+        mov->time = mktime(&time);
+    }
+    mov->time += 0x7C25B080; //1970 based -> 1904 based
 
     if (mov->chapter_track)
         mov_create_chapter_track(s, mov->chapter_track);
diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c
index d44e7676d6..ad91c6a2ad 100644
--- a/libavformat/mxfenc.c
+++ b/libavformat/mxfenc.c
@@ -1407,6 +1407,8 @@ static int mxf_write_header(AVFormatContext *s)
     int i;
     uint8_t present[FF_ARRAY_ELEMS(mxf_essence_container_uls)] = {0};
     const int *samples_per_frame = NULL;
+    AVDictionaryEntry *t;
+    int64_t timestamp = 0;
 
     if (!s->nb_streams)
         return -1;
@@ -1512,8 +1514,18 @@ static int mxf_write_header(AVFormatContext *s)
         sc->order = AV_RB32(sc->track_essence_element_key+12);
     }
 
+#if FF_API_TIMESTAMP
     if (s->timestamp)
-        mxf->timestamp = mxf_parse_timestamp(s->timestamp);
+        timestamp = s->timestamp;
+    else
+#endif
+    if (t = av_dict_get(s->metadata, "creation_time", NULL, 0)) {
+        struct tm time = {0};
+        strptime(t->value, "%Y - %m - %dT%T", &time);
+        timestamp = mktime(&time);
+    }
+    if (timestamp)
+        mxf->timestamp = mxf_parse_timestamp(timestamp);
     mxf->duration = -1;
 
     mxf->timecode_track = av_mallocz(sizeof(*mxf->timecode_track));
diff --git a/libavformat/oggdec.c b/libavformat/oggdec.c
index 597ea3bd80..cd5592208a 100644
--- a/libavformat/oggdec.c
+++ b/libavformat/oggdec.c
@@ -605,15 +605,15 @@ static int64_t ogg_read_timestamp(AVFormatContext *s, int stream_index,
                                   int64_t *pos_arg, int64_t pos_limit)
 {
     struct ogg *ogg = s->priv_data;
-    struct ogg_stream *os = ogg->streams + stream_index;
     AVIOContext *bc = s->pb;
     int64_t pts = AV_NOPTS_VALUE;
-    int i;
+    int i = -1;
     avio_seek(bc, *pos_arg, SEEK_SET);
     ogg_reset(ogg);
 
     while (avio_tell(bc) < pos_limit && !ogg_packet(s, &i, NULL, NULL, pos_arg)) {
         if (i == stream_index) {
+            struct ogg_stream *os = ogg->streams + stream_index;
             pts = ogg_calc_pts(s, i, NULL);
             if (os->keyframe_seek && !(os->pflags & AV_PKT_FLAG_KEY))
                 pts = AV_NOPTS_VALUE;
@@ -639,6 +639,7 @@ static int ogg_read_seek(AVFormatContext *s, int stream_index,
         os->keyframe_seek = 1;
 
     ret = av_seek_frame_binary(s, stream_index, timestamp, flags);
+    os = ogg->streams + stream_index;
     if (ret < 0)
         os->keyframe_seek = 0;
     return ret;
diff --git a/libavformat/version.h b/libavformat/version.h
index 96cc6d4a14..7eb30768b5 100644
--- a/libavformat/version.h
+++ b/libavformat/version.h
@@ -86,5 +86,8 @@
 #ifndef FF_API_LOOP_OUTPUT
 #define FF_API_LOOP_OUTPUT             (LIBAVFORMAT_VERSION_MAJOR < 54)
 #endif
+#ifndef FF_API_TIMESTAMP
+#define FF_API_TIMESTAMP               (LIBAVFORMAT_VERSION_MAJOR < 54)
+#endif
 
 #endif /* AVFORMAT_VERSION_H */
diff --git a/libavutil/pixdesc.c b/libavutil/pixdesc.c
index 491955a020..8a19be1702 100644
--- a/libavutil/pixdesc.c
+++ b/libavutil/pixdesc.c
@@ -918,9 +918,9 @@ const AVPixFmtDescriptor av_pix_fmt_descriptors[PIX_FMT_NB] = {
         .log2_chroma_w= 0,
         .log2_chroma_h= 0,
         .comp = {
-            {0,1,1,0,9},        /* Y */
-            {1,1,1,0,9},        /* U */
-            {2,1,1,0,9},        /* V */
+            {0,1,1,0,8},        /* Y */
+            {1,1,1,0,8},        /* U */
+            {2,1,1,0,8},        /* V */
         },
         .flags = PIX_FMT_BE,
     },