Fix timestamp calculation for code 3 Ogg Opus packets with less than
2 bytes in the last segment (e.g. packet length 255 or 256).
A sample that would seek incorrectly in ffplay can be created with:
ffmpeg -i in.wav -b:a 34k -vbr off -frame_duration 60 out.opus
and libopus 1.1
Also do not read past the end of the buffer when a packet has length 0.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This only updates the case where there is no B frame reordering delay.
Tested-by: Anssi Hannula <anssi.hannula@iki.fi>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
According to http://doc.iridas.com/index.php?title=LUT_Formats
for 3D textures, the data is also RGB, and ordered in such a way
that the red coordinate changes fastest, then the green coordinate,
and finally, the blue coordinate changes slowest:
Fixes Ticket #3236.
Its usefull to know immediatly if extradata is wrong and not just when later
the decoder tries to use it.
This check was removed by: a6a2282c25
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'aec25b1c4650944d32706bfd40eb02bbd5587303':
mpegvideo: split the encoding-only parts of frame_start() into a separate function
Conflicts:
libavcodec/mpegvideo.c
libavcodec/mpegvideo_enc.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'b318106fae65149356934fc72feafef3272fd4ea':
FATE: add a test for the lavr mixing case fixed in fc6a3ef40d
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'a8cc88b1a23dc1515f27cfa98af16a273c539091':
tests/Makefile: allow FILTER* to be called with lists of filter names
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'cc976a75dffa148d655b52604331679ff669e8a2':
audio_mix: print (SKIP) instead of 0.0 for matrix columns removed along with output zeroing
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'fc6a3ef40d34ce8443ae57c2452f3f273d7d4891':
audio_mix: fix zeroing output channels in certain cases
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '58312b2472d3a44d7458865c459d59ef2e02bf1a':
h264: reset data_partitioning if decoding the slice header for NAL_DPA fails
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '1b5d065ca722eb8028c7a08e054b6da3419faf5d':
pmpdec: check that there is at least one audio packet.
Conflicts:
libavformat/pmpdec.c
See: 8b1cd25ca7
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '94a417acc05cc5151b473abc0bf51fad26f8c5a0':
mathematics: remove asserts from av_rescale_rnd()
Conflicts:
libavutil/mathematics.c
The asserts are left in place for now as no code checks the return
value, but we sure can change this if application developers
prefer
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '5569146d48f06564e8fa393424782cceed510916':
adx: check that the offset is not negative
See: 30bce34b67
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'feded990e3ef9af4a0b827d5b6d8fe86f0b94942':
mpegvideo: set reference/pict_type on generated reference frames
Conflicts:
libavcodec/mpegvideo.c
The picture type is left at P type as it was explicitly set before
Merged-by: Michael Niedermayer <michaelni@gmx.at>
With interlaced vc1 it was possible that accesses could happen outside these
tables before this.
Regression since 017e234c20
Reproduced with a sample from Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Fixes (again) : 480i30__codec_WVC1__mode_2__framerate_29.970__type_2__preproc_17.SIGFPE.bfa.390.wmv
No releases are affected by this
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
The error was fixed in commit ce9bd30379
more than one year ago, and does not affect all the currently supported
versions.
In particular, fix trac issue #1535.
Specifically, when the corresponding input channel exists and its matrix
column is all-zero (which is necessary for zeroing the output), the
matrix column must be removed from the matrix.
This is not done currently, so the mixing code would end up using
uninitialized pointers from stack.
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
This should make it easier to catch problems where some of those
pointers are used uninitialized, since reading from NULL should always
crash, while random numbers from stack can turn out to be valid
pointers, so random memory may be silently overwritten.
If it was set before then we can end up trying to decode a slice without
a valid slice header, which can lead to invalid memory access.
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC:libav-stable@libav.org
The code cannot handle there being none, but that should not happen for
valid files.
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC:libav-stable@libav.org
