SSSE3 is already quite old (introduced 2006 for Intel, 2011 for AMD),
so that the overwhelming majority of our users (particularly those
that actually update their FFmpeg) will be using the SSSE3 version
of filter_line.
This commit therefore removes the overridden MMXEXT version
(which didn't abide by the ABI) which allows us to remove
an emms_c() from vf_gradfun.c, so that users with SSSE3 no longer
pay a price for the mere existence of an MMXEXT version.
Reviewed-by: Lynne <dev@lynne.ee>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
To avoid pulling in the entire libavfilter when using the DSP functions
from checkasm.
The rest of the struct is not needed outside vf_idet.c and was moved there.
This wrapping logic still considered any nonzero return from the ASM function
to be the overall result, but this is not true since the addition of
FF_ALPHA_TRANSPARENT.
Fix it by only early returning if FF_ALPHA_STRAIGHT is detected.
Fixes: 9b8b78a815
See-Also: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20301#issuecomment-4802
It can be useful to know if the alpha plane consists of fully opaque
pixels or not, in which case it can e.g. safely be stripped.
This only requires a very minor modification to the AVX2 routines, adding
an extra AND on the read alpha value with the reference alpha value, and a
single extra cheap test per line.
detect_alpha_8_full_c: 2849.1 ( 1.00x)
detect_alpha_8_full_avx2: 260.3 (10.95x)
detect_alpha_8_full_avx512icl: 130.2 (21.87x)
detect_alpha_8_limited_c: 8349.2 ( 1.00x)
detect_alpha_8_limited_avx2: 756.6 (11.04x)
detect_alpha_8_limited_avx512icl: 364.2 (22.93x)
detect_alpha_16_full_c: 1652.8 ( 1.00x)
detect_alpha_16_full_avx2: 236.5 ( 6.99x)
detect_alpha_16_full_avx512icl: 134.6 (12.28x)
detect_alpha_16_limited_c: 5263.1 ( 1.00x)
detect_alpha_16_limited_avx2: 797.4 ( 6.60x)
detect_alpha_16_limited_avx512icl: 400.3 (13.15x)
I also tried replacing some of the instructions by more elaborate ones
using masks, but I found no performance gain significant enough to be worth
maintaining two code paths, so this implementation merely replaces the AVX2
implementation by drop-in AVX512 equivalents.
bwdif8_c: 6362.2 ( 1.00x)
bwdif8_sse2: 1004.9 ( 6.33x)
bwdif8_ssse3: 946.0 ( 6.73x)
bwdif8_avx2: 477.9 (13.31x)
bwdif8_avx512: 273.3 (23.28x)
bwdif10_c: 6341.5 ( 1.00x)
bwdif10_sse2: 872.4 ( 7.27x)
bwdif10_ssse3: 803.4 ( 7.89x)
bwdif10_avx2: 416.7 (15.22x)
bwdif10_avx512: 224.3 (28.27x)
Realtime test at 3840x2160 yuv420p:
avx2: frame=20000 fps=3370 q=-0.0 Lsize=N/A time=00:06:40.00 bitrate=N/A speed=67.4x elapsed=0:00:05.93
avx512: frame=20000 fps=5077 q=-0.0 Lsize=N/A time=00:06:40.00 bitrate=N/A speed= 102x elapsed=0:00:03.93
The use of this function is gated behind avx512icl so that it doesn't
downclock on Skylake.
For detect_range, the usage of vpbroadcast{b,w} requires the AVX512BW extension, and for
detect_alpha we don't want ZMM instructions downclocking old CPUs.
Signed-off-by: James Almer <jamrial@gmail.com>
Requested by a user. Even with autovectorization enabled, the compiler
performs a quite poor job of optimizing this function, due to not being
able to take advantage of the pmaxub + pcmpeqb trick for counting the number
of pixels less than or equal-to a threshold.
blackdetect8_c: 4625.0 ( 1.00x)
blackdetect8_avx2: 155.1 (29.83x)
blackdetect16_c: 2529.4 ( 1.00x)
blackdetect16_avx2: 163.6 (15.46x)
Since psadbw only exists for 8-bits, we have to emulate it for 16-bit
inputs. The simplest sequence is to use a normal subtraction, which is safe
as long as the inputs do not exceed 32767 - so limit this implementation
to 15-bit inputs and below.
For 16-bit inputs, we could in theory instead use a pminw / pmaxw to ensure
the resulting difference does not overflow, but this is slower, and also
breaks the subsequent use of pmaddwd, so I opted to skip 16-bit SIMD for
now.
scene_sad10_c: 114175.6 ( 1.00x)
scene_sad10_avx2: 9617.7 (11.87x)
scene_sad10_avx512: 5208.8 (21.92x)
scene_sad12_c: 114537.8 ( 1.00x)
scene_sad12_avx2: 9614.0 (11.91x)
scene_sad12_avx512: 5186.3 (22.08x)
scene_sad14_c: 114113.9 ( 1.00x)
scene_sad14_avx2: 9612.9 (11.87x)
scene_sad14_avx512: 5186.0 (22.00x)
scene_sad15_c: 114108.9 ( 1.00x)
scene_sad15_avx2: 9612.3 (11.87x)
scene_sad15_avx512: 5186.4 (22.00x)
scene_sad16_c: 114136.0 ( 1.00x)
Trivial to add, but a lot faster (on my machine).
scene_sad8_c: 114476.4 ( 1.00x)
scene_sad8_sse2: 8644.3 (13.24x)
scene_sad8_avx2: 4520.1 (25.33x)
scene_sad8_avx512: 3153.0 (36.31x)
Processes two channels in parallel, using 128-bit XMM registers.
In theory, we could go up to YMM registers to process 4 channels, but this is
not a gain except for relatively high channel counts (e.g. 7.1), and also
complicates the sample load/store operations considerably.
I decided to only add an AVX variant, since the C code is not substantially
slower enough to justify a separate function just for ancient CPUs.
The MMX requantize functions have the MMX permutation
(i.e. FF_IDCT_PERM_SIMPLE) hardcoded and therefore
check for the used permutation (namely via a CRC).
Yet this is very ugly and could even lead to misdetection;
furthermore, since d7246ea9f2
the permutation used here is de-facto and since
bfb28b5ce8 definitely
impossible on x64, making this code dead on x64.
So remove it.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
There are lots of files that don't need it: The number of object
files that actually need it went down from 2011 to 884 here.
Keep it for external users in order to not cause breakages.
Also improve the other headers a bit while just at it.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
This already avoids unnecessary indirectly included headers
in the arch-specific vf_bwdif_init.c files; it is also in
preparation for splitting the actual functions out of vf_bwdif.c.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
This commit fixes bug #10495
The code had several bugs related to post-loop compensation code:
- test assembly instruction performs bitwise AND operation and
generate flags used by jz branch instruction. Wrong test condition
leads to incorrect branching
- Incorrect compensation code for some branches
Signed-off-by: Evgeny Pavlov <lucenticus@gmail.com>
From x86inc:
> On AMD cpus <=K10, an ordinary ret is slow if it immediately follows either
> a branch or a branch target. So switch to a 2-byte form of ret in that case.
> We can automatically detect "follows a branch", but not a branch target.
> (SSSE3 is a sufficient condition to know that your cpu doesn't have this problem.)
x86inc can automatically determine whether to use REP_RET rather than
REP in most of these cases, so impact is minimal. Additionally, a few
REP_RETs were used unnecessary, despite the return being nowhere near a
branch.
The only CPUs affected were AMD K10s, made between 2007 and 2011, 16
years ago and 12 years ago, respectively.
In the future, everyone involved with x86inc should consider dropping
REP_RETs altogether.
This commit enabled assembly code with intel AVX512 VNNI and added unit test for sobel filter
sobel_c: 4537
sobel_avx512icl 2136
Signed-off-by: bwang30 <bin.wang@intel.com>
Signed-off-by: Haihao Xiang <haihao.xiang@intel.com>
