The user-provided value is overwritten in ff_mpv_encode_init()
without having ever been read.
(This has been broken when making these options mpegvideo-specific
in commits 910247f172 and
cf7d2f2d21. No one has ever complained,
so this commit removes these fields.)
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Currently, ff_frame_thread_free() uses the last worker thread
to updates the first worker thread via update_context_from_thread()
immediately before freeing all these worker threads. This is
a remnant of the time in which the first worker was special.
(E.g. the first worker shared its AVCodecInternal with the public
AVCodecContext.)
But these times are over (none of the uses of is_copy matter
for ff_frame_thread_free()); nowadays the only thing that
update_context_from_thread() does is referencing a few
buffers/frames and replacing them with other references instead.
These new references will then be freed immediately thereafter
when the first worker thread is freed. Ensuring that the code is
free of double-frees is achieved by using reference-counted structures
(or in case of AVChannelLayouts: by giving each worker its own copy).
Some archaeology:
a) Updating the first worker thread from the last one used
has been done since frame-threading was added in
37b00b47cb.
b) The precursor to ff_mpv_common_end() checked for is_copy
before freeing pictures (i.e. it only freed them for the first
worker thread).
c) Commits c2dfb1e37c and
e33811bd26 modified the
update_thread_context function of the H.264 decoder
so that it could fail before calling ff_mpeg_update_thread_context().
d) This led to a double free/an assert violation with a H.264
sample for which ff_mpeg_update_thread_context() is not reached
for the final update_context_from_thread(). Commit
a6e4796fbf added code to fix this
sample.
e) This issue was fixed (even with the last mentioned commit reverted)
when the H.264 decoder was deMpegEncContextized in commit
b7fe35c9e5 (merging commit
2c54155407).
f) mpegvideo.c stopped using is_copy when it was switched to refcounted
frames in 759001c534.
g) 1f4cf92cfb removed the init_thread_copy
callbacks; now no FFCodec.close callback checks for is_copy at all
any more.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Currently, $(CPPFLAGS) and $(CFLAGS) are prepended to CXXFLAGS
(the flags for compiling C++) like this:
CXXFLAGS := $(CPPFLAGS) $(CFLAGS) $(CXXFLAGS)
Using ":=" creates a simply expanded variable, i.e. the values
of the variable at the time of assignment are used and later
modifications to them are ignored (using a recursively expanding
variable (i.e. "=" instead of ":=") is not really possible here,
as there would be an infinite loop when evaluating CXXFLAGS).
Yet we perform later additions to CPPFLAGS: HAVE_AV_CONFIG_H and
BUILDING_libfoo are defined. These do not reach C++ compilations.
To fix this a trick is employed to prepend to a recursively
expanded variable while keeping it recursively expanded.
There are two practical consequences of this: C++ files now no longer
include the version.h header, but only the version_major.h header
of their library, saving some recompilations. Furthermore, they
now get some optimized math functions (namely the ones from
lavu/intmath.h instead of the ones from lavu/common.h).
(av_parity() is the only one for which it makes a difference.)
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
This is a more appropriate place for this code, since the values we read
from AV_PKT_DATA_QUALITY_STATS side data are primarily written into
video stats. This ensures that the values written into stats actually
apply to the right packet.
Rename the function to update_video_stats() to better reflect its new
purpose.
It retrieves libavformat's internal dts value (contrary to the
function's name), which is not only incorrect in general, but also
unnecessary because we can access the packet directly.
Its use for muxing is not documented, in practice it is incremented per
each packet successfully passed to the muxer's write_packet(). Since
there is a lot of indirection between ffmpeg receiving a packet from the
encoder and it actually being written (e.g. bitstream filters, the
interleaving queue), using nb_frames here is incorrect.
Add a new counter for packets received from encoder instead.
The top/bottom of the barrel are each coded as two semicircles inside a
square block in the frame. Mask out the parts of the square that lie
outside of these semicircles, so they are made transparent when
alpha_mask=1.
Fixes the other part of #9725.
enc_dec() performs two ffmpeg runs - the first one encoding a source
file into a specified output format, the second one decoding previously
encoded file.
The arguments to this function currently have confusing names - e.g.
dec_opt contains _output_ (i.e. encoding) options for the second
(decoding) ffmpeg invocation. It is also possible to supply _input_
(i.e. decoding) options for the second ffmpeg run, but the argument
is currently unnamed and referred to by number.
Add an _in/_out suffix to argument names to make it clear what they are
used for. Give a name to input options for the decoding ffmpeg run.
has_b_frames should be output_reorder_delay field in AVS3 sequence
header and larger than 1. The parser implementation doesn't parse
that field. Decoder can set has_b_frames properly, so use FFMAX
here to avoid resetting has_b_frames from output_reorder_delay to 1.
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
Unify file access operations by replacing usages of direct calls
to posix fopen() to prepare for long filename support on Windows.
Signed-off-by: softworkz <softworkz@hotmail.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
Unify file access operations by replacing usages of direct calls
to posix fopen() to prepare for long filename support on Windows.
Signed-off-by: softworkz <softworkz@hotmail.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
Since every DLL can use an individual CRT on Windows, having
an exported function that opens a FILE* won't work if that
FILE* is going to be used from a different DLL (or from user
application code).
Internally within the libraries, the issue can be worked around
by duplicating the function in all libraries (this already happened
implicitly because the function resided in file_open.c) and renaming
the function to ff_fopen_utf8 (so that it doesn't end up exported from
the DLLs) and duplicating it in all libraries that use it.
This makes the avpriv_fopen_utf8 / ff_fopen_utf8 function work in
the exact same way as the existing avpriv_open / ff_open, with the
same setup as introduced in e743e7ae6e.
That mechanism doesn't work for external users, thus deprecate the
existing function.
Signed-off-by: Martin Storsjö <martin@martin.st>
Provide a header based inline reimplementation of it.
Using av_fopen_utf8 doesn't work outside of the libraries when built
with MSVC as shared libraries (in the default configuration, where
each DLL gets a separate statically linked CRT).
Signed-off-by: Martin Storsjö <martin@martin.st>
In d3d11va_create_staging_texture(), during the hwmap process, the
ctx->internal->priv is not initialized, resulting in the
texDesc.Format not initialized. Now pass the format value from
d3d11va_transfer_data() to fix it.
$ ffmpeg.exe -y -hwaccel qsv -init_hw_device d3d11va=d3d11 \
-init_hw_device qsv=qsv@d3d11 -c:v h264_qsv \
-i input.h264 -vf "hwmap=derive_device=d3d11va,format=d3d11,hwdownload,format=nv12" \
-f null -
Reviewed-by: Soft Works <softworkz@hotmail.com>
Signed-off-by: Tong Wu <tong1.wu@intel.com>
Signed-off-by: Haihao Xiang <haihao.xiang@intel.com>
When the SLOW_GATHER flag was added to the AVX2 version, this
made FMA3-features not enabled on Zen CPUs.
As FMA3 adds 6-7% across all platforms that support it, in
the interest of saving space, this commit removes the AVX
version and replaces it with an FMA3 version.
The only CPUs affected are Sandy Bridge and Bulldozer, which
have AVX support, but no FMA3 support.
In the future, if there's a demand for it, a version of the
function duplicated for AVX can be added.
Instead of having a fixed -64 prio penalty, make the penalties
more granular.
As the prio is based on the register size in bits, decrementing
it by 129 makes AVX SLOW functions be avoided in favor of any
SSE versions.
This reverts commit 82a68a8771.
Smarter slow ISA penalties makes gathers still useful.
The intention is to use gathers with the final stage of non-ptwo iMDCTs,
where they give benefit.
Do this by making this test a transcode test.
Also fix the test requirements and don't add this test to FATE_AFILTER;
instead use a new variable and a new target for flvenc-tests.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Also add a fate-filter-overlays target containing all these tests
and fix the requirements of the tests; furthermore, remove
unnecessary scale filters from filter-overlay-rgba?_rgba.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Also fix the requirements of these tests: Only the anaglyph
tests need a scale filter, yet it has been inserted for all tests
without any check for its presence.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Lots of tests use the framecrc command together with some filters,
so adding a special function for it seems worthwhile. This commit
adds one new one and modifies an already existing one:
All users of FILTERDEMDEC already use framecrc and the more general
FILTERDEMDECENCMUX can be used in scenarios where more control over
the used encoders/muxers is needed, so use this in cases where
an actual input file is involved.
Furthermore, add FILTERFRAMECRC for the cases where no demuxing/decoding
occurs, because the input is generated via lavfi.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It is unused and given that one needs an encoder to produce
packets from AVFrames (as output by filters) this is likely
to remain so, because FILTERDEMDECENCMUX is better for these
scenarios.
The only case where one can use filters without encoders is
with the lavfi input device: It outputs AVPackets which could
be copied without another conversion to AVFrames. Yet the variable
to check for this is CONFIG_LAVFI_INDEV, but FILTERDEMDECMUX
is designed to work with demuxers (i.e. CONFIG_*_DEMUXER).
So there is no usecase for FILTERDEMDECMUX.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Its performance loss ranges from either being just as fast as individual loads
(Skylake), a few percent slower (Alderlake), 8% slower (Zen 3), to completely
disasterous (older/other CPUs).
Sadly, gathers never panned out fast on x86, even with the benefit of time and
implementation experience.
This also saves a register, as there's no need to fill out an additional
register mask.
Zen 3 (16384-point transform):
Before: 1561050 decicycles in av_tx (fft), 131072 runs, 0 skips
After: 1449621 decicycles in av_tx (fft), 131072 runs, 0 skips
Alderlake:
2% slower on big transforms (65536), to 1% (131072), to a few percent for smaller
sizes.
