This is probably unneeded and normal int would be fine, but its
safer to use LL and this isnt speed relevant
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Pass instead the two variables from the struct needed in the function.
This simplifies writing asm optimized versions of the function
Signed-off-by: James Almer <jamrial@gmail.com>
Since the decoder has been fixed to output something similar to
other implementations, the encoder weight table needed this fix.
Reference PSNR values on a progressive sequence (from tiny_psnr)
from a chained encoding and decoding:
Full progressive: stddev: 0.74 PSNR: 50.69 MAXDIFF: 19
Before: stddev: 1.43 PSNR: 44.97 MAXDIFF: 17
This patch: stddev: 0.76 PSNR: 50.44 MAXDIFF: 14
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* cehoyos/master:
lavc/arm: Use the neon vertical chroma loop filter also for H.264 4:2:2.
lavc/aarch64: Do not use the neon horizontal chroma loop filter for H.264 4:2:2.
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '1a07df31128da3a0020b66502399989b91770d44':
DNxHD: Add support for id 1258 (DNx100 960x720@8)
Conflicts:
Changelog
libavcodec/dnxhddata.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '6996fd204a7f28b46a8c3c97bcf223998218c743':
libopenh264: Log debug messages to a non-null context
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '61928b68dc28e080b8c8191afe5541123c682bbd':
h264: Do not share rbsp_buffer across threads
Conflicts:
libavcodec/h264.c
See: ecbf838c7d
Merged-by: Michael Niedermayer <michaelni@gmx.at>
This avoids a potential conflict with the equally named function for closing files
It also could reduce confusion in debugger backtraces
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Quite a few programs missed that detail of the spec
(including old versions of FFmpeg I believe) and when
we would otherwise fail anyway it seems worth a try
to use a simple byte-aligned stride instead.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
* commit 'a536a4e4bc52d05f59869761337452fb1f1977f6':
lavc: support extracting audio service type from side data
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '728685f37ab333ca35980bd01766c78d197f784a':
Add a side data type for audio service type.
Conflicts:
doc/APIchanges
libavcodec/avcodec.h
libavcodec/version.h
libavutil/frame.h
libavutil/version.h
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'f771b3ab5d3c0b763ee356152be550f4121babd0':
avidec: do not export stream_codec_tag
Conflicts:
libavformat/avidec.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '167e004e1aca7765686ed95d7cd8ea5064d4f6f6':
h264: drop any pretense of support for data partitioning
Conflicts:
libavcodec/h264.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'ecab21ac47d0d4ca604bebf494017ae5090853a8':
h264: do not reset the ref lists in flush_change()
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '9404a47a2d1df418946a338938eb6cdb3afed474':
h264: move parser-only variables to their own context
Conflicts:
libavcodec/h264_parser.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'cf1e0786ed64e69614760bfb4ecd7adbde8e6094':
error_resilience: move the MECmpContext initialization into ER code
Conflicts:
libavcodec/error_resilience.c
libavcodec/h264.c
libavcodec/h264.h
libavcodec/h264_slice.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '443b71928b2f36362e805c037751e6c3c79ea4e8':
hevc: unref the current frame if frame_start() fails
Conflicts:
libavcodec/hevc.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
It does not work correctly and apparently never did. There is no
indication that this (mis)feature is ever used in the wild or even that
any software other than the reference supports it.
Since the code that attempts to support it adds some nontrivial
complexity and has resulted in several bugs in the past, it is better to
just drop it.
Currently, it needs to be initialized by the ER caller (which is
currently either a mpegvideo decoder or h264dec). However, since none of
those decoders use MECmpContext for anything except ER, it makes more
sense to handle it purely inside ER.
This should behave similar to x264 and other encoders, as it handles a
gop_size of 0 as Intra-Only, while it's still possible to control how
many B-Frames it inserts.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Before
2843 decicycles in ff_sbr_autocorrelate_sse3, 262086 runs, 58 skips
After
2693 decicycles in ff_sbr_autocorrelate_sse3, 262117 runs, 27 skips
Signed-off-by: James Almer <jamrial@gmail.com>
if the openjpeg parameter tcp_rates is not 0 ( using the ffmpeg compression_level option )
every 2nd image per thread is badly encoded. By moving the opj_setup_encoder function from
libopenjpeg_encode_init to libopenjpeg_encode_frame this can be prevented.
This fixes ticket #3754.
Signed-off-by: Jean First <jeanfirst@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This improves motion estimation and avoids using uninitialized data
for resolutions that aren't a multiple of 16.
Prior to d2a25c40, the edges used to be initialized so that encoding
was deterministic, but after that commit it started using uninitialized
data (for non multiple of 16 resolutions).
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
This like the previous attempt does not fully correctly decode this
type of non standard H.264, but it now works fully automatic
requiring no manual filters or flags to be used
See Ticket2254
Reviewed-by: Kieran Kunhya <kierank@obe.tv>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
The vec_ste calls were mistakenly changed to vec_vsx_st in c5ca76a, which
caused stack smashing.
Changing them back fixes crashes on ppc64el, when configured with
--toolchain=hardened.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* commit '440119b18836887d98c9e337c5911563bb43588c':
libopenh264enc: Move a declaration of a variable into an ifdef
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'bba02479260d0e7dec8c530a7e75a1c7aa53c06e':
libopenh264enc: Remove a workaround for silencing warnings about unused variables in the OpenH264 header
Merged-by: Michael Niedermayer <michaelni@gmx.at>
The 1.3 release branch of OpenH264 (as well as the master branch)
have been updated so that GCC no longer warns about this variable
as being unused.
Signed-off-by: Martin Storsjö <martin@martin.st>
This leaked a frame on each avcodec_flush_buffers() call, if frame
threading was enabled. It caused severe memory usage in player if you
were seeking a lot.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This fixes out of array reads and/or infinite loops.
30 is the maximum number of bits that can be read into
coeff_abs below.
CC: libav-stable@libav.org
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Martin Storsjö <martin@martin.st>
Move the lavc/imgconvert functions and rename them as follows:
avpicture_get_size -> av_image_get_buffer_size()
avpicture_fill -> av_image_fill_arrays()
avpicture_layout -> av_image_copy_to_buffer()
The new functions have an align parameter, which allows to define the
linesize alignment assumed in the buffer (which is set or read).
The names of the functions are consistent with the lavu/samples API
(av_samples_get_buffer_size(), av_samples_fill_arrays()).
A redundant check has been dropped from av_image_fill_arrays().
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
If refdata was NULL, the memcpy() ended up copying the same memory
block onto itself, which is not only pointless, but also undefined
behavior.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Anshul Maheshwari <anshul.ffmpeg@gmail.com>
To test Closed caption use following command
ffmpeg -f lavfi -i "movie=input.ts[out0+subcc]" -map s output.srt
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This and the next commit improve error concealment for
green-block-artifacts-from-canon-100-hs.MOV
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Default libvpx msvc buils used module definition to export symbols.
These symbols are exported as pointers to data which dont link when
referenced directly.
Reviewed-by: Timothy Gu <timothygu99@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
With a certain fuzzed file, the parser will always return 0 consumed
bytes, which makes calling code call the parser infinitely. Return the
full packet size on error instead. (Here it would be nice if parsers
could return errors at all.)
Additionally, _if_ there's some data left, return that too, which might
help with somewhat broken but still somehow playable files.
Fixes ticket #4242.
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
The parser must always set the out_size and out_data pointers. The API
seems to require it, and the common code in parser.c also relies on it.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
dvdsub_decode() can call append_to_cached_buf() 2 times, the second time
with ctx->buf as argument. If the second append_to_cached_buf() reallocs
ctx->buf, the argument will be a pointer to the previous, freed block.
This can cause invalid reads at least with some fuzzed files - and
possibly with valid files.
Since packets can apparently not be larger than 64K (even if packets are
combined), just use a fixed size buffer. It will be allocated as part of
the DVDSubContext, and although some memory is "wasted", it's relatively
minimal by modern standards and should be acceptable.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* commit '2dbd35b00c6433e587d5f44d5dbc8972ebbaa88e':
libfdk-aacdec: Make sure decoding doesn't add any extra delay in the latest version of fdk-aac
Merged-by: Michael Niedermayer <michaelni@gmx.at>
The latest version added support for a new option for enabling
a signal level limiter, which adds some extra delay. In fdk-aac, this
is enabled by default, but disable it by default here since we'd rather
have zero-delay decoding.
Signed-off-by: Martin Storsjö <martin@martin.st>
These have a DXSA tag and contain alpha in addition to
color values for palette.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
* commit 'ed97963bdbf3bb17fca4f9ea0aa1a97722dec907':
ulti: invert the order of parameters of ulti_decode_frame()
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '2cef68da69a17ed09c313ba3c3850ec1cc0a80e0':
vda: error out if decoded CVPixelBuffer is empty
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'ca09effb01e126b0ac74ff3de70a475423ddee82':
ffv1: Drop unnecessary casts and const qualifiers to match function signatures
Conflicts:
libavcodec/ffv1dec.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Compared to existing, common opensource H264 encoders, this can be
useful since it has got a different license (BSD instead of GPL).
Performance- and qualitywise it is comparable to x264 in ultrafast
mode.
Hooking it up as an encoder in libavcodec also simplifies comparing
it against other common encoders.
This requires OpenH264 1.3 or newer. Since the OpenH264 API and ABI
changes frequently, only releases are supported.
To take advantage of the OpenH264 patent offer, the OpenH264 library
must not be redistributed, but downloaded at runtime at the end-user's
system.
Signed-off-by: Martin Storsjö <martin@martin.st>
This is the order that the caller uses in the rest of the file.
Variables are modified to reflect the order above too and their
initialization is merged with their declarationt. No behavioral
change.
Bug-Id: CID 732286
x86inc can translate r*m into a register or stack on its own
Reviewed-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: James Almer <jamrial@gmail.com>
On some video samples, VDA silently fails to decode frames and returns
kVDADecoderNoErr. Error out in these cases to avoid producing AVFrames with
empty planes.
Signed-off-by: Stefano Pigozzi <stefano.pigozzi@gmail.com>
The code blindly trusted buffer offsets read from the file in the RLE
decoder. Explicitly check the offset. Also error out on other RLE
decoding errors.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* rbultje/vp9-32bit-lpf:
vp9/x86: add myself to copyright holders for loopfilter assembly.
vp9/x86: make filter_16_h work on 32-bit.
vp9/x86: make filter_48/84/88_h work on 32-bit.
vp9/x86: make filter_44_h work on 32-bit.
vp9/x86: make filter_16_v work on 32-bit.
vp9/x86: make filter_48/84_v work on 32-bit.
vp9/x86: make filter_88_v work on 32-bit.
vp9/x86: make filter_44_v work on 32-bit.
vp8/x86: save one register in SIGN_ADD/SUB.
vp9/x86: store unpacked intermediates for filter6/14 on stack.
vp8/x86: move variable assigned inside macro branch.
vp9/x86: simplify ABSSUM_CMP by inverting the comparison meaning.
vp8/x86: remove unused register from ABSSUB_CMP macro.
vp9/x86: slightly simplify 44/48/84/88 h stores.
vp9/x86: make cglobal statement more conservative in register allocation.
vp9/x86: save one register in loopfilter surface coverage.
Merged-by: Michael Niedermayer <michaelni@gmx.at>
The r10k and avrp decoders would previously store 12 bit precision
for the blue channel, which is inconsistent and probably not a
desirable behaviour.
Now the 2 unused extra bits are set to 0.
This is possibly not ideal either as RGBA1010102 format has the same
layout but stores alpha in these bits, thus explicitly setting them
to 1 might be preferable.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
* commit '737d35e33408263c04d7730f5487eed0d04938ba':
vdpau: add support for the H.264 High 4:4:4 Predictive profile
Conflicts:
libavcodec/vdpau_internal.h
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'ebd5320afd42d4315851f3e0ca7f5d4a6300eb68':
vdpau: add support for 4:2:2 and 4:4:4 chroma sampling
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'c220a60f92dde9c7c118fc4deddff5c1f617cda9':
vdpau: add helper for surface chroma type and size
Conflicts:
libavcodec/vdpau.c
libavcodec/version.h
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Since the VDPAU pixel format does not distinguish between different
VDPAU video surface chroma types, we need another way to pass this
data to the application.
Originally VDPAU in libavcodec only supported decoding to 8-bits YUV
with 4:2:0 chroma sampling. Correspondingly, applications assumed that
libavcodec expected VDP_CHROMA_TYPE_420 video surfaces for output.
However some of the new HEVC profiles proposed for addition to VDPAU
would require different depth and/or sampling:
http://lists.freedesktop.org/archives/vdpau/2014-July/000167.html
...as would lossless AVC profiles:
http://lists.freedesktop.org/archives/vdpau/2014-November/000241.html
To preserve backward binary compatibility with existing applications,
a new av_vdpau_bind_context() flag is introduced in a further change.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
This can be used by the application to signal its ability to cope with
video surface of types other than 8-bits YUV 4:2:0.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
This carries the pixel format that would be used if it were not for
hardware acceleration. This is equal to AVCodecContext.pix_fmt if
hardware acceleration is not in use.
Signed-off-by: Anton Khirnov <anton@khirnov.net>
This is to avoid proliferation of similar tables in following changes.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
* commit '9cfa68c560bdec82d2d5ec079f9c5b0f9ca37af0':
mpegts: add support for Opus
Conflicts:
libavcodec/opus_parser.c
libavformat/mpegts.c
See: 74141f693d
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'cee4490b521fd0d02476d46aa2598af24fb8d686':
on2avc: check number of channels
See: 550f3e9df3
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Fixes invalid writes when there are more blocks in a run than total
remaining blocks.
CC: libav-stable@libav.org
Bug-ID: CVE-2014-8548
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Anton Khirnov <anton@khirnov.net>
Fixes invalid writes with very small image heights.
CC: libav-stable@libav.org
Bug-ID: CVE-2014-8547
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Anton Khirnov <anton@khirnov.net>
The frame size must be set by the caller and each dimension must be a
multiple of 2.
CC: libav-stable@libav.org
Bug-ID: CVE-2014-8543
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
The frame size must be set by the caller and each dimension must be a
multiple of 8.
CC: libav-stable@libav.org
Bug-ID: CVE-2014-8542
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Fixes possible invalid memory access.
Based on code by Michael Niedermayer <michaelni@gmx.at>
CC: libav-stable@libav.org
Bug-ID: CVE-2014-8541
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
* commit '67fc8a15e4182ea111cfcd05897709f09d99a33a':
theora: support different visible and coded frame size
Conflicts:
libavcodec/vp3.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Also adds a lot of infrastructure necessary for it.
Some of it is a bit ugly though.
Increases binary size for hardcoded tables by about 12 kB,
which is about 15 kB from qdm2_table minus data and code
saved that was only used for creating it.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
This is the order that the caller uses in the rest of the file. The
same operation is applied to both parameters, so this change is only
done for consistency, it doesn't change the actual behaviour.
Bug-Id: CID 732285 / CID 732286
ff_mpv_common_init sets s->context_initialized.
This fixes decoding of h261 in the cases where the demuxer
hasn't already set the frame size.
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
This avoids trying to do sliced encoding, even if a slice/packet
size is requested (via the -ps option or the rtp_payload_size
field), since the encoder currently doesn't support it (or at least
our decoder can't decode it, even if the h261_encode_gob_header
function is hooked up to be called from the slicing part in
mpegvideo_enc.c).
Signed-off-by: Martin Storsjö <martin@martin.st>
Fixes use of freed memory
Fixes: case5_av_frame_copy_props.mp4
Found-by: Michal Zalewski <lcamtuf@coredump.cx>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This avoids use of uninitialized memory
Fixes: asan_static-oob_17aa046_582_cov_212287884_DBLK_G_VIXS_1.bit
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Fixes out of array read
Fixes: signal_sigsegv_1b0a4da_1865_cov_2167818389_computer_anger.avi
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Should improve cache usage and reduces stack usage.
Also reduces number of copies in case many levels
have the same number of bits.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
Fixes use of freed memory
Fixes: asan_heap-uaf_3660f67_757_cov_1257014655_Hi422FR1_SONY_A.jsv
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Fixes out of array accesses
Fixes: asan_heap-oob_25bcd7e_3783_cov_3553517262_utvideo_rgba_median.avi
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Fixes out of array access
Fixes: asan_stack-oob_32c12e5_2536_cov_2442316831_lena.xface
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Fixes out of array access
Fixes: asan_heap-oob_4d23ba_91_cov_3853393937_128.vmd
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
If allocation fails earlier on, and the next frame is processed,
the slice buffer could be left in a state where line and data_stack
have already been freed, or are otherwise null pointers.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
* commit '10d4c5e55e5b23e165aa4c7723073ebe2c2e7da0':
tiff: set the correct return value when check_size() fails
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '3c27275c1309190f2d6ed69140b67d014215b6c9':
tiff: Check the check_size() return value and forward it
Conflicts:
libavcodec/tiffenc.c
See: d50aa006fb
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Also use the same type for add_entry and check_size.
Bug-Id: CID 700699
CC: libav-stable@libav.org
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Signed-off-by: Vittorio Giovara <vittorio.giovarao@gmail.com>
* commit 'bcbdeb318e0828cf00b458885ffa1e1596c85d06':
asvenc: Forward the error instead of asserting on it
Conflicts:
libavcodec/asvenc.c
Not merged, conditions which are impossible should be checked by assert*()
adding impossible to be true error checks has the potential to confuse
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Also a slight change to the ssse3 code, which prevents a theoretical
overflow in the sharp filter.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Fix linking when only a subset of vaapi decoders is enabled.
Bug-Id: 760
CC: libav-stable@libav.org
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
The input packets are always assumed to be padded and
the av_fast_ family of function takes a pointer to a pointer.
Thanks to Nicolas Dufresne <nicolas.dufresne@collabora.com> for
a similar patch.
Introduced in 7b588bb691.
Bug-Id: 766
CC: libav-stable@libav.org
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Old VDPAU drivers do not support this newly defined profile, so falling
back to Main profile is necessary for backward binary compatibility.
Signed-off-by: Anton Khirnov <anton@khirnov.net>
These fix failures of --enable-xmm-clobber-test
It would be better to change the code to use fewer registers, but until
someone does the used register count must not be too small
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This avoids potential out of bounds writes, with potential future
versions of the library.
Bug-Id: CID 1254945
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
* commit '581c7f0e12b1fa39f73d683e54d6ecda0772c5a9':
arm: make ff_mlp_filter_channel_arm and ff_mlp_rematrix_channel_arm position independent
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '780cd20b00a69e26bbfffbb8eec16fbe999ea793':
aarch64: Use .data.rel.ro for const data with relocations
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'f963f80399deb1a2b44c1bac3af7123e8a0c9e46':
arm: Use .data.rel.ro for const data with relocations
Conflicts:
configure
Merged-by: Michael Niedermayer <michaelni@gmx.at>
This test doesn't cover every possible issue with this function.
It covers options management only.
Signed-off-by: Lukasz Marek <lukasz.m.luki2@gmail.com>
* commit '6fd91fa11909f27902498648680dbb3d13f1f175':
h264: increase MAX_SLICES to 32
The available sample decodes correctly before, but the reporter of the bug
claims that this change reduces artifacts. This is thus merged
If someone has samples that decode differently depending in the MAX_SLICES
value, please open a ticket on trac.
Also this change should be reverted if it turns out that the artifacts
that where seen had a different cause
Merged-by: Michael Niedermayer <michaelni@gmx.at>
This fixes artifacts in the last pixel of rows with some widths and pixel formats
Found-by: Dominique Leroux <Dominique.Leroux@autodesk.com>
Tested-by: Dominique Leroux <Dominique.Leroux@autodesk.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
According to the DPX file format description found at
http://www.fileformat.info/format/dpx/egff.htm the ImageElement part of
the GenericImageHeader also contains an an offset to the real image data
beside the same member that can be found in the GenericFileHeader.
Libav keeps this member empty (=0) while some applications expects it to
be filled properly. FATE test updated accordingly.
Bug-Id: 742
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
* commit 'b280c6202b28b371a8d96850194fd69d7ad5dcc0':
arm: fft_vfp: Unify the behaviour in ff_fft_calc_vfp between arm/thumb
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'ae81576414f2d2083d3118fb4abe1ebc5a7a4c54':
arm: fft_vfp: Add a missing "endconst" when building in thumb mode
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Don't include the function pointer table in the code segment
in arm mode.
This shouldn't have any significant performance effect. It does
end up as a few more instructions than before, for ARM, but
only at the entry to this function, not within the fft functions
themselves.
Signed-off-by: Martin Storsjö <martin@martin.st>
We dont fail hard if its not set as the old API allowed this and our examples
did in the distant past not set it, these examples still work with the
current code and some encoders.
Based on suggestion by: funman
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This code only segfaults and fixing the segfault, the resulting
files are unplayable, so disable to avoid the segfault.
Better solution is welcome
See: [FFmpeg-devel] [PATCH] avcodec/libxavs: remove global header code
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'b72727a5248f1ef02db99b378dce1eb48a46357a':
lavc: mention that the parser callback never returns an error
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '3a56bcee7cb7549b2813e39ce3bee3b7c522aecb':
mpeg12dec: Use more specific error codes
Conflicts:
libavcodec/mpeg12dec.c
See: 1852b2a0f4
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'e0bb74a1403ed77ef369b9d62866f8a4afaf3f1d':
exr: Add a gamma flag to exr loader to avoid banding
Conflicts:
libavcodec/exr.c
See: cd3daad77e
Merged-by: Michael Niedermayer <michaelni@gmx.at>
libxvidcore calculate number of threads basing on video height.
If height is small enough it allocates 0 bytes long memory and
writes to it.
Setting thread_count to 0 uses 1 thread and skips bugged code.
Signed-off-by: Lukasz Marek <lukasz.m.luki2@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
The "faulty" samples actually sound fine when ignoring this issue.
For ticket #3886, more samples are decoded.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Rely on the way memcpy is optimized for one's system instead of looping
on a byte buffer for buffer copies to handle P frames.
Tested-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
For test images manually generated to contain only up prediction,
timing results:
8380x3032 255x185
before: 138635 1992
after: 139232 1996
Actually jumping to the proper version depending on the alignment:
8380x3032: 138767
A 0.5% speed improvement for gigantic images is not worth the code
duplication.
Fixes ticket #4148
Signed-off-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Tested-by: Benoit Fouet <benoit.fouet@free.fr>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
