Fixes integer overflow and out of array accesses.
Found-by: Mateusz j00ru Jurczyk and Gynvael Coldwind
(cherry picked from commit 4d33873c2990b8d6096f60fef384f0efc4482b55)
Fixes use of uninitialized memory and out of stack array read.
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
(cherry picked from commit 6cc94e971933cd38c452172bb048bf760e65cc3e)
Was missed in aeaf268e52fc11c1f64914a319e0edddf1346d6a when integrating
clear_blocks into the idct.
(cherry picked from commit 4506a854a4d846692ba71daeeff661dc214c8fa2)
The overread avoidance fix in cbddee1cca0ebd01e8c5aa694d31228eb4de4b41
broke the computation for the last row since it prevented the safe
reading from the height+1-th row.
Once the frame has been given to ff_filter_frame(), it can
no longer be used, even on error.
Fix trac ticket #3430.
(cherry picked from commit bc6901c94944e4a81be49a6b11183cd0c55d2738)
The s390 architecture requires shared libraries to be built in PIC mode.
Otherwise applications will get wrong relocations at run-time, leading
to confusing segmentation faults.
CC: libav-stable@libav.org
(cherry picked from commit 5ddc9f5052316608799b932c604f9e7561f8ce24)
* qatar/release/10:
ituh263: reject b-frame with pp_time = 0
af_compand: replace strtok_r() with av_get_token()
latm: Always reconfigure if no extradata was set previously
af_compand: add a dependency on strtok_r
lavfi: add compand audio filter
Conflicts:
Changelog
doc/filters.texi
libavfilter/Makefile
libavfilter/af_compand.c
libavfilter/allfilters.c
libavfilter/version.h
All changes are already in our 2.2 branch, this is just for metadata
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '596d3e20ae69a278d562eea08f5e8c0ae5a5bfc4':
parser: cosmetics: Drop some unnecessary parentheses
parser: K&R formatting cosmetics
parser: Remove commented-out cruft
doc: name correct header
af_volume: preserve frame properties
Conflicts:
doc/APIchanges
libavcodec/parser.c
libavfilter/af_volume.c
All changes are already in our 2.2 branch, this is just for metadata
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '7933039ade01b39638ec3d9e638b6ae06ee84984':
af_resample: preserve frame properties
avconv: Do not divide by zero
dca: replace some memcpy by AV_COPY128
h264: avoid undefined behavior in chroma motion compensation
x86: dsputil: Use correct file name as multiple inclusion guard
Conflicts:
ffmpeg.c
All changes are already in our 2.2 branch, this is just for metadata
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '4015829accc2382393d42d62654eb96d896d1326':
bit_depth_template: Use file name as multiple inclusion guard
svq3: Adjust #endif comment
hevc: Mention the missing SPS in the error message
doc: Name the MOV muxer as it should be called
doc: Sort the muxer documentation
Conflicts:
doc/muxers.texi
All changes are already in our 2.2 branch, this is just for metadata
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '39dc4a6bb34baf833ce1e5eabad7d0dbf933237d':
x86: dca: Add missing multiple inclusion guards
gitignore: Add all examples below doc/examples
arm: Mark the stack as non-executable
doxygen: Replace @parblock syntax with manual linebreaks
doxygen: Add a number of missing function parameter descriptions
Conflicts:
.gitignore
libavformat/avformat.h
All changes are already in our 2.2 branch, this is just for metadata
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'a6a2d8eb8f125a2edb512a7a47df33dbd70d6b35':
qt-faststart: Add a note about the -movflags +faststart feature
qt-faststart: Avoid unintentionally sign extending BE_32
qt-faststart: Check offset_count before reading from the moov_atom buffer
qt-faststart: Check the ftello() return codes
qt-faststart: Fix the signedness of variables keeping the ftello return values
qt-faststart: Check fseeko() return codes
qt-faststart: Simplify code by using a MIN() macro
qt-faststart: Increase the copy buffer size to 64 KB
Conflicts:
tools/qt-faststart.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '1d1df82093fdacb2cbc443c70c80f8f801002d28':
pthread_frame: flush all threads on flush, not just the first one
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'de187e3e9ec4803575deb1c293ccad84d2a88da8':
h264: Fix a typo from the previous commit
h264: Lower bound check for slice offsets
hevc: Always consider VLC NALU type mismatch fatal
Prepare for 10_beta2 Release
build: The MPEG-4 video parser depends on h263dsp
Conflicts:
RELEASE
configure
All changes are already in our 2.2 branch, this is just for metadata
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Avoid a division by 0 in ff_mpeg4_set_one_direct_mv.
Sample-Id: 00000168-google
Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
(cherry picked from commit 9514440337875e0c63b409abcd616b68c518283f)
AAC LOAS can have new audio config objects in the stream itself.
Make sure the decoder reconfigures itself when the first one arrives
midstream.
Bug-Id: 644
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
(cherry picked from commit 3aca10bf762a94d7de555cedf1ff0e4f6792bf41)
Without this cast, the BE_32() expression is sign extended when
assigned to an uint64_t, since the uint8_t|uint8_t expression
is promoted to an int.
Also avoid undefined behaviour when left shifting an uint8_t
by 24 by casting it to an uint32_t explicitly before shifting.
Based on a patch by Michael Niedermayer.
Signed-off-by: Martin Storsjö <martin@martin.st>
(cherry picked from commit ea7f79f93796d68559a495be824b6bbd94dfe5f6)
If linking in an object file without this attribute set, the
linker will assume that an executable stack might be needed.
Signed-off-by: Martin Storsjö <martin@martin.st>
(cherry picked from commit 543156d7518f5e5d731123da066d86278f9fa492)
Makes fate-h264 pass under valgrind --undef-value-errors=yes with
-cpuflags none. {avg,put}_h264_chroma_mc8_8 approximately 5% faster,
{avg,put}_h264_chroma_mc4_8 2% faster both on x86 and arm.
(cherry picked from commit 982b596ea6640bfe218a31f6c3fc542d9fe61c31)
