With the necessary pixel formats defined, we can now expose support for
the remaining 10/12bit combinations that VAAPI can handle.
Specifically, we are adding support for:
* HEVC
** 12bit 420
** 10bit 422
** 12bit 422
** 10bit 444
** 12bit 444
* VP9
** 10bit 444
** 12bit 444
These obviously require actual hardware support to be usable, but where
that exists, it is now enabled.
Note that unlike YUVA/YUVX, the Intel driver does not formally expose
support for the alphaless formats XV30 and XV360, and so we are
implicitly discarding the alpha from the decoder and passing undefined
values for the alpha to the encoder. If a future encoder iteration was
to actually do something with the alpha bits, we would need to use a
formal alpha capable format or the encoder would need to explicitly
accept the alphaless format.
These are the formats we want/need to use when dealing with the Intel
VAAPI decoder for 12bit 4:2:0, 12bit 4:2:2, 10bit 4:4:4 and 12bit 4:4:4
respectively.
As with the already supported Y210 and YUVX (XVUY) formats, they are
based on formats Microsoft picked as their preferred 4:2:2 and 4:4:4
video formats, and Intel ran with it.
P12 and Y212 are simply an extension of 10 bit formats to say 12 bits
will be used, with 4 unused bits instead of 6.
XV30, and XV36, as exotic as they sound, are variants of Y410 and Y412
where the alpha channel is left formally undefined. We prefer these
over the alpha versions because the hardware cannot actually do
anything with the alpha channel and respecting it is just overhead.
Y412/XV46 is a normal looking packed 4 channel format where each
channel is 16bits wide but only the 12msb are used (like P012).
Y410/XV30 packs three 10bit channels in 32bits with 2bits of alpha,
like A/X2RGB10 style formats. This annoying layout forced me to define
the BE version as a bitstream format. It seems like our pixdesc
infrastructure can handle the LE version being byte-defined, but not
when it's reversed. If there's a better way to handle this, please
let me know. Our existing X2 formats all have the 2 bits at the MSB
end, but this format places them at the LSB end and that seems to be
the root of the problem.
There are no particular reasons to force the compiler to use the same
register as output and input operand. This forces an extra MOV
instruction if the input value needs to be reused after the swap.
In most cases, this makes no differences, as the compiler will seleect
the same register for both operands either way.
Signed-off-by: Martin Storsjö <martin@martin.st>
There are no particular reasons to force the compiler to use the same
register as output and input operand. This forces an extra MOV
instruction if the input value needs to be reused after the swap.
In most cases, this makes no differences, as the compiler will seleect
the same register for both operands either way.
Signed-off-by: Martin Storsjö <martin@martin.st>
It reduces typing: Before this patch, there were 11 callbacks
that exceeded the 80 char line length limit; now there are zero.
It also allows to remove ONLY_IF_THREADS_ENABLED() in
libavutil/internal.h.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It reduces typing: Before this patch, there were 105 codecs
whose long_name-definition exceeded the 80 char line length
limit. Now there are only nine of them.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It has been deprecated in b4f59beeb4,
but the attribute_deprecated was not set and there was no entry
in APIchanges. This commit adds these and schedules it for removal.
Given that the reason behind the deprecation is exactly the same
as in av_fopen_utf8(), reuse its FF_API_AV_FOPEN_UTF8.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
They are unused since d63443b968.
Furthermore, they were always in the wrong header:
libavutil/internal.h is auto-included almost everywhere, but
FF_SYMVER would only ever be used at a few places, so a proper
header of its own would be appropriate for it.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The AArch64 assembly accesses those symbols directly, without
indirection via e.g. the GOT on ELF. In order for this not to
require text relocations, those symbols need to be resolved fully
at link time, i.e. those symbols can't be interposable.
Normally, so far, this is achieved when linking shared libraries
in two ways; we have a version script (libavcodec/libavcodec.v) which
marks all symbols that don't start with av* as local. Additionally,
we try to add -Wl,-Bsymbolic to the linker options if supported,
making sure that such symbol references are resolved fully at link
time, instead of making them interposable.
When the libavcodec static library is linked into another shared
library, there's no guarantee that it uses similar options (even though
that would be favourable), which would end up requiring text relocations
in the AArch64 assembly.
Explicitly mark the symbols that are accessed from AArch64 assembly
as hidden, so that they are resolved fully at link time even without
the version script and -Wl,-Bsymbolic.
Signed-off-by: Martin Storsjö <martin@martin.st>
As vaapi doesn't actually do anything useful with the alpha channel,
and we have an alphaless format available, let's use that instead.
The changes here are mostly 1:1 switching, but do note the explicit
change in the number of declared channels from 4 to 3 to reflect that
the alpha is being ignored.
This is the alphaless version of VUYA that I introduced recently. After
further discussion and noting that the Intel vaapi driver explicitly
lists XYUV as a support format for encoding and decoding 8bit 444
content, we decided to switch our usage and avoid the overhead of
having a declared alpha channel around.
Note that I am not removing VUYA, as this turned out to have another
use, which was to replace the need for v408enc/dec when dealing with
the format.
The vaapi switching will happen in the next change
The fastest fast Fourier transform in not just the west, but the world,
now for the most popular toy ISA.
On a high level, it follows the design of the AVX2 version closely,
with the exception that the input is slightly less permuted as we don't have
to do lane switching with the input on double 4pt and 8pt.
On a low level, the lack of subadd/addsub instructions REALLY penalizes
any attempt at writing an FFT. That single register matters a lot,
and reloading it simply takes unacceptably long.
In x86 land, vendors would've noticed developers need this.
In ARM land, you get a badly designed complex multiplication instruction
we cannot use, that's not present on 95% of devices. Because only
compilers matter, right?
Future optimization options are very few, perhaps better register
management to use more ld1/st1s.
All timings below are in cycles:
A53:
Length | C | New (lavu) | Old (lavc) | FFTW
------ |-------------|-------------|-------------|-----
4 | 842 | 420 | 1210 | 1460
8 | 1538 | 1020 | 1850 | 2520
16 | 3717 | 1900 | 3700 | 3990
32 | 9156 | 4070 | 8289 | 8860
64 | 21160 | 9931 | 18600 | 19625
128 | 49180 | 23278 | 41922 | 41922
256 | 112073 | 53876 | 93202 | 101092
512 | 252864 | 122884 | 205897 | 207868
1024 | 560512 | 278322 | 458071 | 453053
2048 | 1295402 | 775835 | 1038205 | 1020265
4096 | 3281263 | 2021221 | 2409718 | 2577554
8192 | 8577845 | 4780526 | 5673041 | 6802722
Apple M1
New - Total for len 512 reps 2097152 = 1.459141 s
Old - Total for len 512 reps 2097152 = 2.251344 s
FFTW - Total for len 512 reps 2097152 = 1.868429 s
New - Total for len 1024 reps 4194304 = 6.490080 s
Old - Total for len 1024 reps 4194304 = 9.604949 s
FFTW - Total for len 1024 reps 4194304 = 7.889281 s
New - Total for len 16384 reps 262144 = 10.374001 s
Old - Total for len 16384 reps 262144 = 15.266713 s
FFTW - Total for len 16384 reps 262144 = 12.341745 s
New - Total for len 65536 reps 8192 = 1.769812 s
Old - Total for len 65536 reps 8192 = 4.209413 s
FFTW - Total for len 65536 reps 8192 = 3.012365 s
New - Total for len 131072 reps 4096 = 1.942836 s
Old - Segfaults
FFTW - Total for len 131072 reps 4096 = 3.713713 s
Thanks to wbs for some simplifications, assembler fixes and a review
and to jannau for giving it a look.
_Float16 support was available on arm/aarch64 for a while, and with gcc
12 was enabled on x86 as long as SSE2 is supported.
If the target arch supports f16c, gcc emits fairly efficient assembly,
taking advantage of it. This is the case on x86-64-v3 or higher.
Same goes on arm, which has native float16 support.
On x86, without f16c, it emulates it in software using sse2 instructions.
This has shown to perform rather poorly:
_Float16 full SSE2 emulation:
frame=50074 fps=848 q=-0.0 size=N/A time=00:33:22.96 bitrate=N/A speed=33.9x
_Float16 f16c accelerated (Zen2, --cpu=znver2):
frame=50636 fps=1965 q=-0.0 Lsize=N/A time=00:33:45.40 bitrate=N/A speed=78.6x
classic half2float full software implementation:
frame=49926 fps=1605 q=-0.0 Lsize=N/A time=00:33:17.00 bitrate=N/A speed=64.2x
Hence an additional check was introduced, that only enables use of
_Float16 on x86 if f16c is being utilized.
On aarch64, a similar uplift in performance is seen:
RPi4 half2float full software implementation:
frame= 6088 fps=126 q=-0.0 Lsize=N/A time=00:04:03.48 bitrate=N/A speed=5.06x
RPi4 _Float16:
frame= 6103 fps=158 q=-0.0 Lsize=N/A time=00:04:04.08 bitrate=N/A speed=6.32x
Since arm/aarch64 always natively support 16 bit floats, it can always
be considered fast there.
I'm not aware of any additional platforms that currently support
_Float16. And if there are, they should be considered non-fast until
proven fast.
IEEE-754 differentiates two different kind of NaNs.
Quiet and Signaling ones. They are differentiated by the MSB of the
mantissa.
For whatever reason, actual hardware conversion of half to single always
sets the signaling bit to 1 if the mantissa is != 0, and to 0 if it's 0.
So our code has to follow suite or fate-testing hardware float16 will be
impossible.
Convert the input from a scatter to a gather instead,
which is faster and better for SIMD.
Also, add a pre-shuffled exptab version to avoid
gathering there at all. This doubles the exptab size,
but the speedup makes it worth it. In SIMD, the
exptab will likely be purged to a higher cache
anyway because of the FFT in the middle, and
the amount of loads stays identical.
For a 960-point inverse MDCT, the speedup is 10%.
This makes it possible to write sane and fast SIMD
versions of inverse MDCTs.
In oneVPL, MFXLoad() and MFXCreateSession() are required to create a
workable mfx session[1]
Add config filters for D3D9/D3D11 session (galinart)
The default device is changed to d3d11va for oneVPL when both d3d11va
and dxva2 are enabled on Microsoft Windows
This is in preparation for oneVPL support
[1] https://spec.oneapi.io/versions/latest/elements/oneVPL/source/programming_guide/VPL_prg_session.html#onevpl-dispatcher
Co-authored-by: galinart <artem.galin@intel.com>
Signed-off-by: galinart <artem.galin@intel.com>
Signed-off-by: Haihao Xiang <haihao.xiang@intel.com>
The following Cflags has been added to libmfx.pc, so mfx/ prefix is no
longer needed when including mfx headers in FFmpeg.
Cflags: -I${includedir} -I${includedir}/mfx
Some old versions of libmfx have the following Cflags in libmfx.pc
Cflags: -I${includedir}
We may add -I${includedir}/mfx to CFLAGS when running 'configure
--enable-libmfx' for old versions of libmfx, if so, mfx headers without
mfx/ prefix can be included too.
If libmfx comes without pkg-config support, we may do a small change to
the settings of the environment(e.g. set -I/opt/intel/mediasdk/include/mfx
instead of -I/opt/intel/mediasdk/include to CFLAGS), then the build can
find the mfx headers without mfx/ prefix
After applying this change, we won't need to change #include for mfx
headers when mfx headers are installed under a new directory.
This is in preparation for oneVPL support (mfx headers in oneVPL are
installed under vpl directory)
Poisoning returned buffers is based around the implicit assumption
that the contents of said buffers are transient. Yet this is not true
for the buffer pools used by the various hardware contexts which store
important state in there that needs to be preserved.
Furthermore, the current code is also based on the assumption
that the complete buffer pointed to by AVBuffer->data coincides with
AVBufferRef->data; yet an implementation might store some data of its
own before the actual user-visible data (accessible via AVBufferRef)
which would be broken by the current code.
(This is of course yet more proof that the AVBuffer API is not the right
tool for the hardware contexts.)
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Directly branch into the special 64-point deinterleave
subroutine rather than going through the general deinterleave.
64-point transform timings on Zen 3:
Before:
1974 decicycles in av_tx (fft),16776864 runs, 352 skips
After:
1956 decicycles in av_tx (fft),16775378 runs, 1838 skips
__lasx_xvldx does not accept a pointer to const (in fact,
no function in lasxintrin.h does so), although it is not allowed
to modify the pointed-to buffer. Therefore this commit adds a wrapper
for it in order to constify the H264Chroma API in a later commit.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
__lsx_vldx does not accept a pointer to const (in fact,
no function in lsxintrin.h does so), although it is not allowed
to modify the pointed-to buffer. Therefore this commit adds a wrapper
for it in order to constify the HEVC DSP functions in a later commit.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The "AYUV" format is defined by Microsoft as their preferred format for
4:4:4 content, and so it is the format used by Intel VAAPI and QSV.
As Microsoft like to define their byte ordering in little-endian
fashion, the memory order is reversed, and so our pix_fmt, which
follows memory order, has a reversed name (VUYA).
The only duration field currently present in AVFrame is pkt_duration,
which is semantically restricted to those frames that are output by
decoders.
Add a new field that stores the frame's duration without regard for how
that frame was produced. Deprecate pkt_duration.
av_fast_realloc and av_fast_mallocz? store the size of
the objects they allocate in an unsigned. Yet they overallocate
and currently they can allocate more than UINT_MAX bytes
in case a user has requested a size of about UINT_MAX * 16 / 17
or more if SIZE_MAX > UINT_MAX (and if the user increased
max_alloc_size via av_max_alloc). In this case it is impossible
to store the true size of the buffer via the unsigned*;
future requests are likely to use the (re)allocation codepath
even if the buffer is actually large enough because of
the incorrect size.
Fix this by ensuring that the actually allocated size
always fits into an unsigned. (This entails erroring out
in case the user requested more than UINT_MAX.)
Reviewed-by: Tomas Härdin <tjoppen@acc.umu.se>
Reviewed-by: Anton Khirnov <anton@khirnov.net>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
