- imc: use correct position for flcoeffs2 calculation
- hevc: check slice address length
- snow: remove an obsolete av_assert2
- webp: fix infinite loop in webp_decode_frame
- wavpack: limit extra_bits to 32 and use get_bits_long
- ffmpeg: only count got_output/errors in decode_error_stat
- ffmpeg: exit_on_error if decoding a packet failed
- pthread_frame: forward error codes when flushing
- huffyuvdec: validate image size
- wavpack: use get_bits_long to read up to 32 bits
- nutdec: check maxpos in read_sm_data before returning success
- s302m: fix arithmetic exception
- vc1dec: use get_bits_long and limit the read bits to 32
- mpegaudiodec: copy AVFloatDSPContext from first context to all contexts
- avcodec/vp8: Check buffer size in vp8_decode_frame_header()
- avcodec/vp8: Fix null pointer dereference in ff_vp8_decode_free()
- avcodec/diracdec: Check for hpel_base allocation failure
- avcodec/rv34: Clear pointers in ff_rv34_decode_init_thread_copy()
- avfilter/af_aresample: Check ff_all_* for allocation failures
- avcodec/pthread_frame: clear priv_data, avoid stale pointer in error case
- swscale/utils: Clear pix buffers
- avutil/fifo: Fix the case where func() returns less bytes than requested in av_fifo_generic_write()
- ffmpeg: Fix cleanup after failed allocation of output_files
- avformat/mov: Fix deallocation when MOVStreamContext failed to allocate
- ffmpeg: Fix crash with ost->last_frame allocation failure
- ffmpeg: Fix cleanup with ost = NULL
- avcodec/pthread_frame: check avctx on deallocation
- avcodec/sanm: Reset sizes in destroy_buffers()
- avcodec/alac: Clear pointers in allocate_buffers()
- bytestream2: set the reader to the end when reading more than available
- avcodec/utils: use a minimum 32pixel width in avcodec_align_dimensions2() for H.264
- avcodec/mpegvideo: Clear pointers in ff_mpv_common_init()
- oggparsedirac: check return value of init_get_bits
- wmalosslessdec: reset frame->nb_samples on packet loss
- wmalosslessdec: avoid reading 0 bits with get_bits
- Put a space between string literals and macros.
- avcodec/rawenc: Use ff_alloc_packet() instead of ff_alloc_packet2()
- avcodec/aacsbr: check that the element type matches before applying SBR
- avcodec/h264_slice: Use w/h from the AVFrame instead of mb_w/h
- vp9/update_prob: prevent out of bounds table read
- avfilter/vf_transpose: Fix rounding error
- avcodec/h264_refs: discard mismatching references
- avcodec/mjpegdec: Fix small picture upscale
- avcodec/pngdec: Check values before updating context in decode_fctl_chunk()
- avcodec/pngdec: Copy IHDR & plte state from last thread
- avcodec/pngdec: Require a IHDR chunk before fctl
- avcodec/pngdec: Only allow one IHDR chunk
- wmavoice: limit wmavoice_decode_packet return value to packet size
- swscale/swscale_unscaled: Fix rounding difference with RGBA output between little and big endian
- ffmpeg: Do not use the data/size of a bitstream filter after failure
- swscale/x86/rgb2rgb_template: fix signedness of v in shuffle_bytes_2103_{mmx,mmxext}
- vda: unlock the pixel buffer base address.
- swscale/rgb2rgb_template: Fix signedness of v in shuffle_bytes_2103_c()
- swscale/rgb2rgb_template: Implement shuffle_bytes_0321_c and fix shuffle_bytes_2103_c on BE
- swscale/rgb2rgb_template: Disable shuffle_bytes_2103_c on big endian
- swr: Remember previously set int_sample_format from user
- swresample: soxr implementation for swr_get_out_samples()
- avformat/swfdec: Do not error out on pixel format changes
- ffmpeg_opt: Fix forcing fourccs
- configure: Check for x265_api_get
- swscale/x86/rgb2rgb_template: don't call emms on sse2/avx functions
- swscale/x86/rgb2rgb_template: add missing xmm clobbers
- library.mak: Workaround SDL redefining main and breaking fate tests on mingw
- vaapi_h264: fix RefPicList[] field flags.
- avcodec/libtheoraenc: Check for av_malloc failure
- ffmpeg_opt: Fix -timestamp parsing
- hevc: make avcodec_decode_video2() fail if get_format() fails
- avcodec/cavsdec: Use ff_set_dimensions()
- swr: fix alignment issue caused by 8ch sse functions
- avcodec/mjpegdec: fix len computation in ff_mjpeg_decode_dqt()
- avcodec/jpeg2000dec: fix boolean operator
- avcodec/hevc_ps: Explicitly check num_tile_* for negative values
- avformat/matroskadec: Cleanup error handling for bz2 & zlib
- avformat/nutdec: Fix use of uinitialized value
- tools/graph2dot: use larger data types than int for array/string sizes
- avformat/matroskaenc: Check ff_vorbiscomment_length in put_flac_codecpriv()
- avcodec/mpeg12dec: use the correct dimensions for checking SAR
- xcbgrab: Validate the capture area
- xcbgrab: Do not assume the non shm image data is always available
- avfilter/lavfutils: disable frame threads when decoding a single image
- avformat/mov: Do not read ACLR into extradata for H.264
- ffmpeg: remove incorrect network deinit
- OpenCL: Avoid potential buffer overflow in cmdutils_opencl.c
- libvpxenc: only set noise reduction w/vp8
- vp9: remove another optimization branch in iadst16 which causes overflows.
- lavf: Reset global flag on deinit
- network: Do not leave context locked on error
- vp9: remove one optimization branch in iadst16 which causes overflows.
- fate: Include branch information in the payload header
- avformat/utils: Ensure that AVFMT_FLAG_CUSTOM_IO is set before use
- avformat/img2dec: do not rewind custom io buffers
- avcodec/alsdec: Use av_mallocz_array() for chan_data to ensure the arrays never contain random data
- avcodec/atrac3plusdsp: fix on stack alignment
- swresample/swresample-test: Randomly wipe out channel counts
- swresample: Check channel layouts and channels against each other and print human readable error messages
- swresample: Allow reinitialization without ever setting channel layouts (cherry picked from commit 80a28c7509a11114e1aea5b208d56c6646d69c07)
- swresample: Allow reinitialization without ever setting channel counts
- dashenc: replace attribute id with contentType for the AdaptationSet element
- avformat/matroskaenc: Use avoid_negative_ts_use_pts if no stream writes dts
- avformat/mux: Add avoid_negative_ts_use_pts
- tests/fate-run: do not attempt to parse tiny_psnrs output if it failed
- cafdec: free extradata before allocating it
- imgutils: initialize palette padding bytes in av_image_alloc
- aacdec: don't return frames without data
- id3v2: catch avio_read errors in check_tag
- avi: Validate sample_size
- aacsbr: break infinite loop in sbr_hf_calc_npatches
- diracdec: avoid overflow of bytes*8 in decode_lowdelay
- diracdec: prevent overflow in data_unit_size check
- avformat/matroskadec: Use tracks[k]->stream instead of s->streams[k]
- matroskadec: use uint64_t instead of int for index_scale
- pngdec: don't use AV_PIX_FMT_MONOBLACK for apng
- pngdec: return correct error code from decode_frame_common
- nutdec: fix illegal count check in decode_main_header
- nutdec: fix memleaks on error in nut_read_header
- apedec: prevent out of array writes in decode_array_0000
- apedec: set s->samples only when init_frame_decoder succeeded
- swscale/ppc/swscale_altivec.c: POWER LE support in yuv2planeX_8() delete macro GET_VF() it was wrong
- alac: reject rice_limit 0 if compression is used
- alsdec: only adapt order for positive max_order
- bink: check vst->index_entries before using it
- mpeg4videodec: only allow a positive length
- aacpsy: correct calculation of minath in psy_3gpp_init
- alsdec: validate time diff index
- alsdec: ensure channel reordering is reversible
- ac3: validate end in ff_ac3_bit_alloc_calc_mask
- aacpsy: avoid psy_band->threshold becoming NaN
- aasc: return correct buffer size from aasc_decode_frame
- matroskadec: export cover art correctly
- mxfenc: don't try to write footer without header
- mxfenc: fix memleaks in mxf_write_footer
- rtpenc_mpegts: Set chain->rtp_ctx only after avformat_write_header succeeded
- rtpenc_mpegts: Free the right ->pb in the error path in the init function
The most important new features are AVFoundation screen-grabbing support,
animated WebP decoding support, and Animated PNG support. In addition, many
exciting features for video streaming are also implemented, including MPEG-
DASH fragmenting muxer, HEVC RTP payload muxer, and UDP Lite support.
Add -mstackrealign -mpreferred-stack-boundary=4 to CFLAGS on i386 for all
gcc 4.x versions, this appears to still be needed with gcc 4.8.4.
Bump PKGREVISION for ffmpeg2 and ffplay2 (which uses ffmpeg2 patches)
