generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
30,329 Commits 24 Branches 241 Tags 173 MiB
641c7afe3c
Commit Graph

14273 Commits

Author SHA1 Message Date
Mans Rullgard
60f10e0ad3 h264pred: use unsigned types for pixel values, fix signed overflows 
Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-09 12:27:19 +01:00
Mans Rullgard
4d1418cd4f h264: fix signed overflows in x*0x01010101 expressions 
Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-09 12:27:19 +01:00
Mans Rullgard
d66b9dec11 h264pred: remove unused variables 
Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-09 12:27:19 +01:00
Mans Rullgard
bb59156606 vp8: fix signed overflows 
In addition to avoiding undefined behaviour, an unsigned type
makes more sense for packing multiple 8-bit values.

Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-08 20:03:55 +01:00
Mans Rullgard
e708afd3c0 motion_est: fix some signed overflows 
Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-08 20:03:55 +01:00
Mans Rullgard
559c244d42 dca: fix signed overflow in shift 
Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-08 20:03:55 +01:00
Mans Rullgard
d12294304a aacdec: fix undefined shifts 
Since nnz can be zero, this is needed to avoid a shift by 32.

Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-08 20:03:55 +01:00
Laurent Aimar
a00676e48e bink: Check for various out of bound writes 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-08 16:32:01 +02:00
Laurent Aimar
24adf7832b bink: Check for out of bound writes when building tree 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-08 16:25:56 +02:00
Mans Rullgard
ac6eab1496 put_bits: fix invalid shift by 32 in flush_put_bits() 
If flush_put_bits() is called when the 32-bit buffer is empty,
e.g. after writing a multiple of 32 bits, and invalid shift by
32 is performed.  Since flush_put_bits() is called infrequently,
this additional check should have negligible performance impact.

Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-08 02:41:58 +01:00
Laurent Aimar
9bd854b1ff mpc8: Check out of bound bands limit 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 17:15:45 +02:00
Laurent Aimar
7d17a794f0 xan: Prevent NULL dereference with missing palette 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 17:15:31 +02:00
Laurent Aimar
3db3fdf4c6 xan: Check for out of bound reads in xan_huffman_decode() 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:32 +02:00
Laurent Aimar
3e0757c2a8 xan: Fixed out of bound accesses in xan_unpack() 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:32 +02:00
Laurent Aimar
1cd0a55163 motionpixels: Prevent calling init_vlc() with invalid parameters 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:32 +02:00
Laurent Aimar
5f05cf4ea9 shorten: Fix out of bound writes in fix_bitshift() 
The data pointers s->decoded[*] already take into account s->nwrap.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
1720603287 dsicinav: Check for out of bounds writes 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
64263dd526 tiertexseqv: Check for out of bound reads 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
4fd56f842c quickdraw: Check for out of bound reads 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
e3ca9b93d9 dsicinav: Check for out of bounds reads 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
210c80331e motionpixels: Fix the size of workspace buffers 
Some buffers must be mod 4 in width and/or height.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
d337dd3a90 motionpixels: Clear FF_INPUT_BUFFER_PADDING_SIZE bytes at the end of the temporary buffer 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
d99427cb8b wmavoice: Check for corrupted extra data 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
1c1449b548 wmavoice: Check for out of bound writes 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
06be075cda xan: Prevent NULL dereferences with missing reference frame 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
c7e631986b bink: Prevent NULL dereferences with missing reference frame 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
2c6cf13940 wavpack: Reset internal state on corrupted blocks 
wavpack_decode_block() supposes that it is called back with the exact
same buffer unless it has returned with an error. With multi-channels
files, wavpack_decode_frame() was breaking this assumption.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:31 +02:00
Laurent Aimar
2c1ba79941 wmapro: Validate the number of audio channels before using it 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:30 +02:00
Laurent Aimar
95010d18b2 shorten: Prevent block size from increasing 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:30 +02:00
Laurent Aimar
124a16f678 xan: Prevent out of bound accesses 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 16:25:30 +02:00
Michael Niedermayer
14c21c1ff5 H264: Only wait before triggering ff_thread_setup_complete() until the next slice that contains a start-of-field/frame macroblock 
This allows concurrent decoding of the last field/frame, rather than
only the last slice, of data packets with multiple NAL units packed
together.

This will fix the slowdown reported in e.g. bug 52.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
 2011-10-07 14:23:26 +02:00
Laurent Aimar
a72cad0a6c vp6: Reset the internal state when aborting key frames header parsing 
It prevents leaving the state only half initialized.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 00:37:32 +02:00
Laurent Aimar
3d09d0017d vp56: Release old pictures after a resolution changes 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 00:27:24 +02:00
Laurent Aimar
066fff755a vp6: Check for huffman tree build errors 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 00:27:03 +02:00
Laurent Aimar
0ec6d6e9b6 vp56: Check for missing reference frame data 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-07 00:25:43 +02:00
Laurent Aimar
d239d4b447 cinepak: Fix invalid read access on extra data 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-06 23:35:29 +02:00
Laurent Aimar
c0cbe36b18 vmd: fix segfaults on corruped streams 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-06 23:33:09 +02:00
Laurent Aimar
3a742470a8 cook: Fix js_vlc_bits value validation for joint stereo 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-06 23:31:06 +02:00
Laurent Aimar
69a0bce753 Fixed deference of NULL pointer in motionpixels decoder. 
Some of the arguments given to init_vlc() come from the stream
and can be corrupted.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-10-06 22:29:12 +02:00
Anton Khirnov
d97efd7f87 libx264: support 9- and 10-bit output. 2011-10-06 09:16:06 +02:00
Ronald S. Bultje
4418aa9cb3 h264: correct implicit_weight for field-interlaced pictures. 2011-10-05 04:01:23 -07:00
Ronald S. Bultje
330deb7592 mpegvideo: set correct offset for edge emulation buffer. 
Using the old code, half of it was unused and the other half was too
small for e.g. >8bpp interlaced data, causing random buffer overruns.
 2011-10-05 04:01:23 -07:00
Ronald S. Bultje
0884dd5a1b mpegvideo: fix position of bottom edge. 
It was wrong in colorspaces where horizontal and vertical chroma
subsampling are not the same, e.g. 422.
 2011-10-05 04:01:23 -07:00
Diego Biurrun
e83c2ddebf Fix 'heigth' vs. 'height' typos. 2011-10-05 11:12:01 +02:00
Anton Khirnov
a4ea00d021 lavc/lavf: use unique private classes. 
This is needed by the new AVOptions API.
 2011-10-05 07:52:30 +02:00
Anton Khirnov
0ba1e1978d lavc: use designated initializers for av_codec_context_class 2011-10-05 07:52:05 +02:00
Justin Ruggles
dd376b1a12 qcelpdec: cosmetics: do not add line break before opening bracket in 'for', 
'while', 'if/else', and 'switch' statements.

also fixes some spacing, but only if already changing a line
 2011-10-04 17:10:11 -04:00
Justin Ruggles
e43dd3d2a8 qcelp: check output buffer size before decoding 2011-10-04 17:10:11 -04:00
Chris Rankin
bde2570013 qcelpdec: fix the return value of qcelp_decode_frame(). 
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
 2011-10-04 17:09:53 -04:00
Justin Ruggles
1b5a189f06 sipr: fix the output data size check and only calculate it once. 2011-10-04 16:13:20 -04:00
Diego Biurrun
12bd8532cf Synchronize various 4CCs and codec tags from FFmpeg. 2011-10-04 11:58:53 +02:00
Justin Ruggles
7d49f79f1c qdm2: check output buffer size before decoding 2011-10-03 21:34:19 -04:00
Laurent Aimar
5a19acb17c Fix out of bound reads in the QDM2 decoder. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
 2011-10-03 21:34:19 -04:00
Laurent Aimar
291d74a46d Check for out of bound writes in the QDM2 decoder. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
 2011-10-03 21:34:19 -04:00
Anton Khirnov
24ec9ac475 libvpx: fix build with older libvpx versions. 
VPX_ERROR_RESILIENT_DEFAULT and VPX_ERROR_RESILIENT_PARTITIONS weren't
defined before 4cb0ebe5b27d35ccc2a78c1d16f2622ddef21f74 (CommitDate: Tue
Jun 28 11:10:17 2011)
 2011-10-03 13:06:24 +02:00
Mans Rullgard
6308729e68 ARM: check for inline asm 'y' operand modifier support 
The inline asm added in bf5d46d uses the 'y' modifier which
is only supported from gcc 4.5.  This check allows building
with older compilers.

Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-10-03 08:56:24 +01:00
Justin Ruggles
5674d4b0a3 mpc8: check output buffer size before decoding 2011-10-02 10:34:39 -04:00
Justin Ruggles
8290d1f38b mpc7: return error if packet is too small. 2011-10-02 10:34:39 -04:00
Justin Ruggles
c8b5c4d274 mpc7: check output buffer size before decoding 2011-10-02 10:34:39 -04:00
Justin Ruggles
fac6b7f9f2 nellymoserdec: allocate float_buf only when decoding to int16 2011-10-02 10:19:04 -04:00
Justin Ruggles
32b484464c nellymoserdec: use NELLY_BUF_LEN instead of 128 2011-10-02 10:19:04 -04:00
Justin Ruggles
ded0800d44 nellymoserdec: use NELLY_BLOCK_LEN instead of 64 when appropriate. 2011-10-02 10:19:04 -04:00
Justin Ruggles
84464b2e7f nellymoserdec: allow user to request SAMPLE_FMT_FLT for output samples. 2011-10-02 10:19:04 -04:00
Justin Ruggles
8b31c086b6 nellymoser: check output buffer size before decoding 2011-10-02 10:19:04 -04:00
Luca Barbato
0e5cc174bb win32: improve threading algorithm warning 
If no threading is requested do not issue warning about unsupported
threading algorithm.
 2011-10-02 08:58:08 +02:00
Anton Khirnov
b036a7180c lavc: add video/audio/encoding flags to global_quality option 2011-10-01 07:16:56 +02:00
Luca Barbato
6450f26c9a libvpxenc: use libvpx's own defaults for some parameters 
Specifically, qmin/qmax, gop_size and keyint_min.

Fixes bug 47.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
 2011-10-01 07:16:56 +02:00
Luca Barbato
e3eb015ba4 vpxenc: add private options 
Make libvpx support close to the libx264 one. Thanks to
Jan Gerber <j@v2v.cc> for the support.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
 2011-10-01 07:16:56 +02:00
Justin Ruggles
08bd22a61b dpcm: return error if packet is too small 2011-09-30 18:15:53 -04:00
Justin Ruggles
0354fb7ebe dpcm: use smaller data types for static tables 2011-09-30 18:15:53 -04:00
Justin Ruggles
5a54d51017 dpcm: use sol_table_16 directly instead of through the DPCMContext. 2011-09-30 18:15:53 -04:00
Justin Ruggles
f47f7efd14 dpcm: replace short with int16_t 2011-09-30 18:15:53 -04:00
Justin Ruggles
8d77d12a2b dpcm: check to make sure channels is 1 or 2. 2011-09-30 18:15:53 -04:00
Justin Ruggles
4bad464e7f dpcm: misc pretty-printing 2011-09-30 18:15:53 -04:00
Justin Ruggles
1de8401c56 dpcm: remove unnecessary variable by using bytestream functions. 
Uses 'buf' directly instead of a separate iterator variable 'in'.
 2011-09-30 18:15:53 -04:00
Justin Ruggles
fc6faee0d1 dpcm: move codec-specific variable declarations to their corresponding decoding blocks. 2011-09-30 18:15:53 -04:00
Justin Ruggles
989bb7bd04 dpcm: consistently use the variable name 'n' for the next input byte. 2011-09-30 18:15:52 -04:00
Justin Ruggles
04b24cf94b dpcm: output AV_SAMPLE_FMT_U8 for Sol DPCM subcodecs 1 and 2. 
Uses the native sample format for the codec instead of left-shifting all
samples by 8.
 2011-09-30 18:15:52 -04:00
Justin Ruggles
76db17dc7d dpcm: calculate and check actual output data size prior to decoding. 2011-09-30 18:15:52 -04:00
Justin Ruggles
3db8db406f dpcm: factor out the stereo flag calculation 2011-09-30 18:15:52 -04:00
Justin Ruggles
b09c7eefb7 dpcm: cosmetics: rename channel_number to ch 
Make the code easier to read.
 2011-09-30 18:15:52 -04:00
Carl Eugen Hoyos
bcb1555489 dxva2_h264: pass the correct 8x8 scaling lists 
Copy the Inter 8x8 scaling list as second 8x8 matrix into DXVA2's
quantization matrix data structure instead of a potentially unset
Intra chroma scaling matrix.

Fix dxva2 decoding for some H264 samples.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-09-30 20:15:17 +02:00
Mans Rullgard
bf5d46d8e6 dca: NEON optimised high freq VQ decoding 
Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-09-30 19:01:23 +01:00
Justin Ruggles
6326afd5e9 avcodec: reject audio packets with NULL data and non-zero size 
There is no valid reason the user should ever send such packets in the
first place, but the documentation for CODEC_CAP_DELAY states that the
codec is guaranteed not to get a NULL packet unless that capability is
set. That isn't true without preventing this case.
 2011-09-30 11:02:50 -04:00
Joakim Plate
dcce09d64b dxva: Add ability to enable workaround for older ATI cards 
The workaround needs to be enabled per PCI ID which cannot be detected inside
libavcodec. So add a flag to manually enable the alternate behavior.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2011-09-30 11:17:53 +02:00
Justin Ruggles
e562fbd003 adpcm: move codec-specific variable declarations to the sections for the corresponding codecs. 2011-09-29 16:54:01 -04:00
Justin Ruggles
9662539c10 adpcm: check buffer size in Funcom ISS decoder before reading header. 
Also use the post-header data size to control termination of the main
decoding loop.
 2011-09-29 16:54:01 -04:00
Justin Ruggles
ba5d2890d7 adpcm: simplify reading of Funcom ISS frame header. 2011-09-29 16:54:01 -04:00
Justin Ruggles
5c9eb4fabb adpcm: check buffer size in IMA DK4 decoder before reading header. 
Also use the post-header data size to control termination of the main
decoding loop.
 2011-09-29 16:54:01 -04:00
Justin Ruggles
a57ea1a87e adpcm: simplify reading of IMA DK4 frame header. 2011-09-29 16:54:01 -04:00
Justin Ruggles
8114f94ac9 adpcm_ms: clean up reading of predictor coefficients 2011-09-29 16:54:01 -04:00
Justin Ruggles
943f4db552 adpcm_4xm: process planar packets sequentially rather than simultaneously. 
Also properly clip the right channel step_index.
 2011-09-29 16:54:00 -04:00
Justin Ruggles
119974b164 adpcm_ima_wav: process channel-interleaved blocks sequentially rather than simultaneously. 
Speeds up the ADPCM IMA WAV decoder by 15-20% overall.
 2011-09-29 16:54:00 -04:00
Justin Ruggles
7c287b18a0 adpcm: update reference links 
Add Multimedia Wiki link.
Mark dead links with [dead]. Some can still be accessed through archive.org.
Update URLs for pages which have moved.
Replace duplicated links in adpcmenc.c with a note to see the ADPCM decoder
reference documents.
 2011-09-29 16:54:00 -04:00
Justin Ruggles
ac94b8bcc6 adpcm: simplify packet size bounds checking in the ADPCM IMA QT decoder. 
This is easier to understand. It also avoids returning existing samples mixed
with new samples when the packet is too small.
 2011-09-29 16:54:00 -04:00
Justin Ruggles
de0b586a87 adpcm: simplify and speed up several ADPCM decoders. 2011-09-29 16:54:00 -04:00
Justin Ruggles
5a9ed7c110 adpcm: pretty-print tables 2011-09-29 16:54:00 -04:00
Alex Converse
fcbe421cee prores: Handle 0 or fewer bits left 
show_bits() is undefined when the number of bits is less than or equal to
zero.
 2011-09-29 09:45:57 -07:00
Mans Rullgard
bf00a73ace dca: clear inactive subbands only once in qmf_32_subbands() 
Writing zeros to the high entries in the array need only be
done once as the cutoff position is constant throughout the
loop.

Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-09-29 12:34:17 +01:00
Mans Rullgard
baf6b738f2 ARM: NEON optimised vector_fmac_scalar() 
Signed-off-by: Mans Rullgard <mans@mansr.com>
 2011-09-28 15:56:09 +01:00