PSEUDOPAL pixel formats are not paletted, but carried a palette with the
intention of allowing code to treat unpaletted formats as paletted. The
palette simply mapped the byte values to the resulting RGB values,
making it some sort of LUT for RGB conversion.
It was used for 1 byte formats only: RGB4_BYTE, BGR4_BYTE, RGB8, BGR8,
GRAY8. The first 4 are awfully obscure, used only by some ancient bitmap
formats. The last one, GRAY8, is more common, but its treatment is
grossly incorrect. It considers full range GRAY8 only, so GRAY8 coming
from typical Y video planes was not mapped to the correct RGB values.
This cannot be fixed, because AVFrame.color_range can be freely changed
at runtime, and there is nothing to ensure the pseudo palette is
updated.
Also, nothing actually used the PSEUDOPAL palette data, except xwdenc
(trivially changed in the previous commit). All other code had to treat
it as a special case, just to ignore or to propagate palette data.
In conclusion, this was just a very strange old mechnaism that has no
real justification to exist anymore (although it may have been nice and
useful in the past). Now it's an artifact that makes the API harder to
use: API users who allocate their own pixel data have to be aware that
they need to allocate the palette, or FFmpeg will crash on them in
_some_ situations. On top of this, there was no API to allocate the
pseuo palette outside of av_frame_get_buffer().
This patch not only deprecates AV_PIX_FMT_FLAG_PSEUDOPAL, but also makes
the pseudo palette optional. Nothing accesses it anymore, though if it's
set, it's propagated. It's still allocated and initialized for
compatibility with API users that rely on this feature. But new API
users do not need to allocate it. This was an explicit goal of this
patch.
Most changes replace AV_PIX_FMT_FLAG_PSEUDOPAL with FF_PSEUDOPAL. I
first tried #ifdefing all code, but it was a mess. The FF_PSEUDOPAL
macro reduces the mess, and still allows defining FF_API_PSEUDOPAL to 0.
Passes FATE with FF_API_PSEUDOPAL enabled and disabled. In addition,
FATE passes with FF_API_PSEUDOPAL set to 1, but with allocation
functions manually changed to not allocating a palette.
This is the same combination of .section directives as used in
aarch64/asm.S.
Since Xcode 9.3, the bundled clang supports altmacro and doesn't
require using gas-preprocessor any longer.
Signed-off-by: Martin Storsjö <martin@martin.st>
This crash was introduced by 8bbf2dacbf,
which could incorrectly overwrite the failure result from creating the
device.
Fixes ticket #7108.
Reviewed-by: wm4 <nfxjfg@googlemail.com>
This new side-data will contain info on how a packet is encrypted.
This allows the app to handle packet decryption.
Signed-off-by: Jacob Trimble <modmaker@google.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The old vaAcquireBufferHandle() API works in fewer cases and provides
less information than the current vaExportSurfaceHandle(), but it exists
on older versions and is already used by the OpenCL code.
Per MediaSDK documentation, it requires width/height to 16 alignment.
Without this patch, hwupload pipeline may fail if 16 alignment is
not met. Although this patch also apply 16 alignment to qsv encoder/decoder,
it will not bring any side-effect to them as they are already aligned.
Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Fixes building with VAAPI but not libdrm, which was broken by
389f4c3e0d. Just unconditionally include
the header, since it doesn't depend on libdrm being present.
This adds a way for an API user to transfer QP data and metadata without
having to keep the reference to AVFrame, and without having to
explicitly care about QP APIs. It might also provide a way to finally
remove the deprecated QP related fields. In the end, the QP table should
be handled in a very similar way to e.g. AV_FRAME_DATA_MOTION_VECTORS.
There are two side data types, because I didn't care about having to
repack the QP data so the table and the metadata are in a single
AVBufferRef. Otherwise it would have either required a copy on decoding
(extra slowdown for something as obscure as the QP data), or would have
required making intrusive changes to the codecs which support export of
this data.
The new side data types are added under deprecation guards, because I
don't intend to change the status of the QP export as being deprecated
(as it was before this patch too).
Everything related to the QP data is deprecated, with qp_table_buf being
an inconsistent exception. Some parts were under the deprecation guards,
some not. It probably didn't even compile.
Adds the level prefix to all log messages, except those with level <=
AV_LOG_QUIET as they seem to be used for flushing the log buffer.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Tobias Rapp <t.rapp@noa-archive.com>
enable dump bit stream filter and update opt fate test ref.
Signed-off-by: Jun Zhao <mypopydev@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
add AV_OPT_FLAG_BSF_PARAM for bit stream filter options.
Signed-off-by: Jun Zhao <mypopydev@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Accepting 'u' suffix for a time specification is neither intuitive nor
consistent (now that we don't accept m). Also there was a bug in the code
accepting an extra 's' even after 'ms'.
Signed-off-by: Marton Balint <cus@passwd.hu>
The commit which added those was pushed prematurely before anyone could object
to illogical suffixes like just m for milliseconds. Without this, we'd be locked
into never being able to implement the "m" suffix for minutes.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
The PicStruct is required by MediaSDK, so give a default value.
hwupload does not work without this.
Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Fixes: 5567/clusterfuzz-testcase-minimized-5769966247739392
Fixes: runtime error: negation of -2147483648 cannot be represented in type 'int'; cast to an unsigned type to negate this value to itself
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
There are 32 pseudo-instructions for each floating-point comparison
instruction, but only 8 of them are actually valid in legacy-encoded mode.
The remaining 24 requires the use of VEX-encoded (v-prefixed) instructions
and can therefore be disregarded for this purpose.
This attribute is supported for this architecture in MSVC as well
(but produces errors if used for 32 bit x86).
Signed-off-by: Martin Storsjö <martin@martin.st>
D3D11 has rather fine grained per format capabilities for different uses
that can be queried at runtime. Since we don't know what the user wants
to do with the formats when av_hwdevice_get_hwframe_constraints() is
called, we simply return all formats that have the most basic support.
In addition, this does not allow creating frames contexts with sw_format
for which no known transfer formats exist. In theory, we should check
whether the chroma format (i.e. the sw_format) is supported at all by
the vdpau driver, but checking for transfer formats has the same effect.
Note that the pre-existing code adds 1 to priv->nb_pix_fmts[i] for
unknown reason, and some checks need to account for that to check for
empty lists. They are not off-by-one errors.
w32threads and os2threads both support static mutex initialization now,
so don't limit it to pthreads only.
Reviewed-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
wm4
Martin Storsjö
Mark Thompson
Jacob Trimble
Ruiling Song
James Almer
Tobias Rapp
Jun Zhao
Marton Balint
Aurelien Jacobs
Rostislav Pehlivanov
Gyan Doshi
Michael Niedermayer
Diego Biurrun
Jun Zhao
Jacob Trimble
Henrik Gramner
Anton Khirnov