Commit graph

773 commits

Author SHA1 Message Date
Jan Schmidt
3a12092ae6 Enable threading in ffmpeg decoders that support it.
Add a max-threads property, which defaults to '0 = auto'
Add a utility function taken from libschroedinger which sets
the ffmpeg worker thread count to match the computer processor
count by default.
2011-06-21 14:33:29 +10:00
Raimo Järvi
8e2404b8df ffdec: Fix calculating frame duration from last timestamp and frame count.
Fixes bug #651714.
2011-06-06 14:45:56 +02:00
Raimo Järvi
4a66ecdf25 ffdec: Check for AV_NOPTS_VALUE in PTS value.
https://bugzilla.gnome.org/show_bug.cgi?id=651625
2011-06-01 16:11:05 +02:00
Raimo Järvi
82463fe8bc ffdec: Add mpegversion 2 to AAC decoder caps
Fixes bug #651447.
2011-05-30 12:31:31 +02:00
Edward Hervey
9db205f9f7 ffmpegdec: Don't discard timestamps if output AND input are in order
Avoids bogus timestamps for AVCHD-lite streams
2011-05-29 20:09:47 +02:00
Sebastian Dröge
d1fac3e779 ffenc_aac: Add profile and level to the caps
Fixes bug #650596.
2011-05-27 10:54:12 +02:00
Sebastian Dröge
b8f7aab1ba ffdec: Always use parser for AAC LATM/LOAS 2011-05-26 15:09:18 +02:00
Rafael Diniz
df40381a83 ffmpeg: Add codec mapping for AAC LATM/LOAS
Also add the stream-format fields to the CODEC_ID_AAC caps.

Fixes bug #650695.
2011-05-25 10:10:48 +02:00
Thadeu Lima de Souza Cascardo
a3c56f60cd ffdec: Do not use invalid input timestamp as next timestamp.
When input buffer timestamps are invalid, next timestamp are used for
audio. Then, the next out timestamp is updated with the used timestamp
and the calculated duration. However, if the used timestamp is invalid,
it should not be used. Otherwise, the next buffer will use a wrong
timestamp that is not in the clipped segment, making the buffer to be
dropped.

This fixes playback with SBTVD MPEG TS streams, using AAC LATM.
2011-05-09 11:02:50 +02:00
Vincent Penquerc'h
920bfb0da2 ffdec: do not try to resync to the next keyframe when late
A keyframe may be quite a while in the future, and the decoder
has no way of knowing this. A poor decision could mean quite some
time with no video output.
This decision should be left to the upstream element: a demuxer
might know about incoming keyframes, or some other element might
be able to request a keyframe.

Fixes bug #649372.
2011-05-06 13:23:45 +02:00
Edward Hervey
775a3a0640 postproc: pp_*_t => pp_* 2011-04-21 20:56:45 +02:00
Edward Hervey
2ce58ad764 audioresample: Use new av_audio_resample API
We might want to expose the various properties in the future
2011-04-21 20:56:45 +02:00
Edward Hervey
a30b434b53 ffmpegdec: Set caps on outgoing audio buffers 2011-04-21 20:56:45 +02:00
Edward Hervey
fe1a971b31 ffmpegdec: Use new AVPacket-based API
Right now it doesn't use any of the extra fields AVPacket provides.
It might be wise to investigate the pts/dts ones to see if we can finally
get rid of the timing-related cruft we have.
2011-04-21 20:56:45 +02:00
Edward Hervey
436a211aff gstffmpegdec: Use non-deprecated av_parser_parse2 2011-04-21 20:56:45 +02:00
Edward Hervey
7c85e2d743 ffmpegdec: Use skip_frame instead of deprecated hurry_up 2011-04-21 20:56:45 +02:00
Edward Hervey
71f48a1d84 ffmpegcfg: mb_qmin/mb_qmax are deprecated/removed
We just use qmin/qmax instead
2011-04-21 20:56:45 +02:00
Edward Hervey
a004f0bc34 ffmpegdemux: Define our own MAX_STREAMS 2011-04-21 20:56:45 +02:00
Edward Hervey
f16291a6cc ffmpeg: Disable code using old metadata API
It is now gone and we need to switch to the new API

See #566605
2011-04-21 20:56:45 +02:00
Edward Hervey
5aa722965c ffmpegdemux: The index_built AVFormatContext hasn't been used in over 2 years 2011-04-21 20:56:45 +02:00
Edward Hervey
fd80f177c9 codecmap: _XVID is gone
It was always a bogus one, since it used the external libxvid for encoding.
2011-04-21 20:56:45 +02:00
Edward Hervey
78ae6a34cb ffmpeg: PKT_FLAG_KEY => AV_PKT_FLAG_KEY 2011-04-21 20:56:44 +02:00
Edward Hervey
dfd26c36f3 ffmpeg: AVERROR_* => AVERROR(*) 2011-04-21 20:56:44 +02:00
Edward Hervey
2f81d0d63b ffmpeg: CodecType => AVMediaType 2011-04-21 20:56:44 +02:00
Edward Hervey
446c50e052 gstffmpeg: register_protocol => av_register_protocol2 2011-04-21 20:56:44 +02:00
Marc Plano-Lesay
9c42b7cae9 Fix unused-but-set-variable warnings with gcc 4.6 2011-04-19 16:21:20 +01:00
Edward Hervey
c35c1d1446 gstffmpeg(de)mux: Disable known broken (de)muxers
Idea provided by David Schleef
Support provided by Michael Smith and Tim Mueller
Food and Drinks provided by Toronado
2011-04-09 01:17:24 +02:00
Stefan Kost
dd7b219c42 ffmpeg: don't divide by 0 when checking the framerates
Check for denom==0 first and set it to on in that case.
2011-04-08 11:42:33 +03:00
Wim Taymans
6444bd2557 dec: avoid making the buffer writable
Set the caps right after allocation of the buffer because we know the buffer is
writable then and we are correctly negotiated. Since ffmpeg keeps around
references to frames, making the buffer metadata writable where it was done
before pushing will always end up with a copy and that makes the sink do a slow
memcpy all the time.
2011-04-07 12:34:51 +02:00
Miguel Angel Cabrera Moya
f63d36ade8 ffmpegdec: do buffer padding before parsing and before decoding
FFMpeg parsing and decoding calls require to additionally allocate bytes
at the end of the input bitstream and this padding must be initialized
to zero.

https://bugzilla.gnome.org/show_bug.cgi?id=595590
2011-04-05 14:14:38 +02:00
Julien Isorce
32ca9c111b ffvideoscale: add support for UYVY
Fixes #642015
2011-02-14 16:58:49 +01:00
Tim-Philipp Müller
8d40c6357d ffmpegdec: improve error message when set_caps is called but we have no mapping
This may happen e.g. if gst-ffmpeg is compiled against an external
libavcodec and the external lib is upgraded.

See e.g.
https://bugzilla.gnome.org/show_bug.cgi?id=640825
2011-01-31 23:28:33 +00:00
Edward Hervey
a2a4ee93c4 codecmap: Add mapping for g722 2011-01-25 16:34:04 +01:00
Edward Hervey
5429138b89 codecmap: Add proper caps name for Camstudio 2011-01-24 18:29:24 +01:00
Edward Hervey
7879105579 gstffmpegdemux: Ignore more pseudo demuxers 2011-01-24 18:29:00 +01:00
Edward Hervey
9a1d1fef32 gstffmpegmux: Blacklist some muxer formats 2011-01-24 17:43:44 +01:00
Edward Hervey
cc4a11831d gstffmpegcodecmap: Add some new codec mappings
VP8
LAGARITH
GSM
GSM_MS
2011-01-24 17:26:20 +01:00
Edward Hervey
ed1b1eefaf ffmpegprotocol: Slight API change 2011-01-24 17:26:20 +01:00
Edward Hervey
2de5aaf22d ffmpegdec: Set the mp3 decoder to a rank of NONE
It's just causing too many headaches. This will force people to use
a working mp3 decoder, like mad.
2011-01-07 18:08:15 +01:00
Edward Hervey
051e2de3d3 ffmpegdec: Fix unitialized variables on macosx 2011-01-05 16:59:55 +01:00
Edward Hervey
a90df26e15 ffmpegenc: Disable non-audio/video encoders
Fixes #629648
2011-01-03 11:45:04 +01:00
Wim Taymans
22d44c87b1 ffmpgdec: work around parser timestamp bug
Sometimes the parser loses track of timestamps and starts to reuse old
timestamp. Feed it some dummy data and clear some context variables to work
around the problem.
2010-12-21 12:14:22 +01:00
Sebastian Dröge
dde194dc1a ffvideoscale: Change classification to Filter/Converter/Video/Scaler 2010-12-16 10:12:02 +01:00
Sebastian Dröge
a5910b4ec4 ffdeinterlace: Change classification to Filter/Effect/Video/Deinterlace 2010-12-16 10:05:03 +01:00
Wim Taymans
ec48b24291 dec: scale the estimated duration by number of frames
When estimating the frame duration, the diff between two incomming timestamps
should be scaled by the amount of frames in the interval. Improves duration
estimation and DTS interpolation.
2010-12-12 12:38:55 +01:00
Wim Taymans
957c728b3d dec: use input timestamp diff for duration estimation
Use the diff between input timestamps to estimate the duration when no duration
is set on input buffers. Only do this when there are no reordered input
timestamps. Improves interpolation in DTS mode when no input duration is set.
2010-12-12 11:20:22 +01:00
Mark Nauwelaerts
e837ec505d ffmpegenc: ensure proper clean state prior to (re-)negotiation
Fixes #632691.
2010-10-25 12:16:44 +02:00
Wim Taymans
d82633c4e3 ffdec: don't destroy the parser state
When we flush the parser cache, we only need to clear the bytes of the cache,
not the complete state of the cache. In the case of H264 this doesn't require
the parser to receive a new SPS/PPS after a DISCONT buffer.
2010-10-22 13:05:16 +02:00
Wim Taymans
cc2a6e493b ffdec: be more careful with DTS timestamp interpolation
Don't use -1 and 0 duration buffers to calculate next output timestamp as they
are very likely wrong.

Fixes #632877
2010-10-22 13:05:16 +02:00
Stefan Kost
07ce1d1906 various: add missing G_PARAM_STATIC_STRINGS flags 2010-10-19 17:26:08 +03:00