avformat/mpc: attempt to allocate a packet that is not smaller than the data inside it
avcodec/alsdec: Clear MPEG4AudioConfig so that no use of uninitialized memory is possible
avformat/flvenc: Do not allow creating h263/mpeg4 in flv without unofficial format extensions being enabled
avcodec/mss4: Fix () in MKVAL() macro
avcodec/mss34dsp: fix () in SOP* macros
avcodec/mlpdec: fix () in MSB_MASK() macro
avcodec/golomb-test: fix () in EXTEND() macro
avcodec/aac: fix () in IS_CODEBOOK_UNSIGNED macro
avcodec/libvorbisenc: dont add the duration to AV_NOPTS_VALUE
avutil/cpu: force mmx on selection of higher x86 SIMD features
avcodec/diracdec: move mc buffer allocation to per frame
avfilter/graphdump: Fix pointer to local outside scope
more...
* avcodec/utvideoenc: fix slice_bits size
* avcodec/snow: split block clipping checks
* avcodec/ansi: fix integer overflow
* avcodec/msrle: use av_image_get_linesize() to calculate the linesize
* avformat/mpegtsenc: Check data array size in mpegts_write_pmt()
* avcodec/wmalosslessdec: fix mclms_coeffs* array size
* samplefmt: avoid integer overflow in av_samples_get_buffer_size()
* avcodec/vc1: reset fcm/field_mode in non advanced header parsing
* avcodec/takdec: always check bits_per_raw_sample
* avcodec/vmnc: Check that rectangles are within the picture
* dnxhdenc: fix mb_rc size
* pthread_frame: flush all threads on flush, not just the first one
* rtpdec_asf: Copy the need_parsing field from the chained demuxer
* Fix compilation on ARM with android gcc 4.7
* Allow decoding of slightly broken Nikon avi files.
1.2.5 was released on 2014-01-16. It is the latest stable FFmpeg
release from the 1.2 release branch, which was cut from master on
2013-03-07. Amongst lots of other changes, it includes all changes
from ffmpeg-mt, libav master of 2013-03-07, libav 9 as of 2013-03-07.
to address issues with NetBSD-6(and earlier)'s fontconfig not being
new enough for pango.
While doing that, also bump freetype2 dependency to current pkgsrc
version.
Suggested by tron in PR 47882
version 1.2:
- VDPAU hardware acceleration through normal hwaccel
- SRTP support
- Error diffusion dither in Swscale
- Chained Ogg support
- Theora Midstream reconfiguration support
- EVRC decoder
- audio fade filter
- filtering audio with unknown channel layout
- allpass, bass, bandpass, bandreject, biquad, equalizer, highpass, lowpass
and treble audio filter
- improved showspectrum filter, with multichannel support and sox-like colors
- histogram filter
- tee muxer
- il filter ported from libmpcodecs
- support ID3v2 tags in ASF files
- encrypted TTA stream decoding support
- RF64 support in WAV muxer
- noise filter ported from libmpcodecs
- Subtitles character encoding conversion
- blend filter
- stereo3d filter ported from libmpcodecs
It is the latest stable FFmpeg release from the 1.1 release branch, which was cut from master on 2013-01-06. Amongst lots of other changes, it includes all changes from ffmpeg-mt, libav master of 2013-01-06, libav 9.1 as of 2013-02-23.
1.1.2 was released on 2013-02-06. It is the latest stable FFmpeg
release from the 1.1 release branch, which was cut from master on
2013-01-06. Amongst lots of other changes, it includes all changes
from ffmpeg-mt, libav master of 2013-01-06, libav 9.1 as of
2013-02-05.
- stream disposition information printing in ffprobe
- filter for loudness analysis following EBU R128
- Opus encoder using libopus
- ffprobe -select_streams option
- Pinnacle TARGA CineWave YUV16 decoder
- TAK demuxer, decoder and parser
- DTS-HD demuxer
- remove -same_quant, it hasn't worked for years
- FFM2 support
- X-Face image encoder and decoder
- 24-bit FLAC encoding
- multi-channel ALAC encoding up to 7.1
- metadata (INFO tag) support in WAV muxer
- subtitles raw text decoder
- support for building DLLs using MSVC
- LVF demuxer
- ffescape tool
- metadata (info chunk) support in CAF muxer
- field filter ported from libmpcodecs
- AVR demuxer
- geq filter ported from libmpcodecs
- remove ffserver daemon mode
- AST muxer/demuxer
- new expansion syntax for drawtext
- BRender PIX image decoder
- ffprobe -show_entries option
- ffprobe -sections option
- ADPCM IMA Dialogic decoder
- BRSTM demuxer
- animated GIF decoder and demuxer
- PVF demuxer
- subtitles filter
- IRCAM muxer/demuxer
- Paris Audio File demuxer
- Virtual concatenation demuxer
- VobSub demuxer
- JSON captions for TED talks decoding support
- SOX Resampler support in libswresample
- aselect filter
- SGI RLE 8-bit decoder
- Silicon Graphics Motion Video Compressor 1 & 2 decoder
- Silicon Graphics Movie demuxer
- apad filter
- Resolution & pixel format change support with multithreading for H.264
- documentation split into per-component manuals
- pp (postproc) filter ported from MPlayer
- NIST Sphere demuxer
- MPL2, VPlayer, MPlayer, AQTitle, PJS and SubViewer v1 subtitles demuxers and decoders
- Sony Wave64 muxer
- adobe and limelight publisher authentication in RTMP
- data: URI scheme
- support building on the Plan 9 operating system
- kerndeint filter ported from MPlayer
- histeq filter ported from VirtualDub
- Megalux Frame demuxer
- 012v decoder
- Improved AVC Intra decoding support
from the changelog:
- many bug fixes that where found with Coverity
- The following CVE fixes where backported:
CVE-2012-2796, CVE-2012-2775, CVE-2012-2772, CVE-2012-2776,
CVE-2012-2779, CVE-2012-2787, CVE-2012-2794, CVE-2012-2800,
CVE-2012-2802, CVE-2012-2801, CVE-2012-2786, CVE-2012-2798,
CVE-2012-2793, CVE-2012-2789, CVE-2012-2788, CVE-2012-2790,
CVE-2012-2777, CVE-2012-2784
- hundreads of other bug fixes, some possibly security relevant,
see the git log for details.
changes:
Fixes: CVE-2012-0853, CVE-2012-0858, CVE-2011-3929, CVE-2011-3936,
CVE-2011-3937, CVE-2011-3940, CVE-2011-3945, CVE-2011-3947
Several security issues that dont have CVE numbers.
and about 150 bugfixes
Due to multiple inclusions of stdint.h, DragonFly was failing to build
one of the akode plugins with a typedef redefinition error. It had
previously built, so the UINT64_C fix for the akode mpc plugin broke
the ffmpeg plugin.
It turns out that the first patch effectively just imported
machine/int_const.h, so that header can be directly imported instead
rather than having to include the main contents of <stdint.h> multiple
times. This refinement also works on NetBSD, but OpenBSD and FreeBSD
unfortunately embed UINT64_C directly within <stdint.h>. Therefore
original patch is left in place to cover those systems.
