checkasm benchmarks on 1.5 GHz Cortex-A72 are as follows. Note that the C
version can still outperform the NEON version in specific cases. The balance
between different code paths is stream-dependent, but in practice the best
case happens about 5% of the time, the worst case happens about 40% of the
time, and the complexity of the remaining cases fall somewhere in between.
Therefore, taking the average of the best and worst case timings is
probably a conservative estimate of the degree by which the NEON code
improves performance.
vc1dsp.vc1_h_loop_filter4_bestcase_c: 19.0
vc1dsp.vc1_h_loop_filter4_bestcase_neon: 48.5
vc1dsp.vc1_h_loop_filter4_worstcase_c: 144.7
vc1dsp.vc1_h_loop_filter4_worstcase_neon: 76.2
vc1dsp.vc1_h_loop_filter8_bestcase_c: 41.0
vc1dsp.vc1_h_loop_filter8_bestcase_neon: 75.0
vc1dsp.vc1_h_loop_filter8_worstcase_c: 294.0
vc1dsp.vc1_h_loop_filter8_worstcase_neon: 102.7
vc1dsp.vc1_h_loop_filter16_bestcase_c: 54.7
vc1dsp.vc1_h_loop_filter16_bestcase_neon: 130.0
vc1dsp.vc1_h_loop_filter16_worstcase_c: 569.7
vc1dsp.vc1_h_loop_filter16_worstcase_neon: 186.7
vc1dsp.vc1_v_loop_filter4_bestcase_c: 20.2
vc1dsp.vc1_v_loop_filter4_bestcase_neon: 47.2
vc1dsp.vc1_v_loop_filter4_worstcase_c: 164.2
vc1dsp.vc1_v_loop_filter4_worstcase_neon: 68.5
vc1dsp.vc1_v_loop_filter8_bestcase_c: 43.5
vc1dsp.vc1_v_loop_filter8_bestcase_neon: 55.2
vc1dsp.vc1_v_loop_filter8_worstcase_c: 316.2
vc1dsp.vc1_v_loop_filter8_worstcase_neon: 72.7
vc1dsp.vc1_v_loop_filter16_bestcase_c: 62.2
vc1dsp.vc1_v_loop_filter16_bestcase_neon: 103.7
vc1dsp.vc1_v_loop_filter16_worstcase_c: 646.5
vc1dsp.vc1_v_loop_filter16_worstcase_neon: 110.7
Signed-off-by: Ben Avison <bavison@riscosopen.org>
Signed-off-by: Martin Storsjö <martin@martin.st>
This avoids unnecessary rebuilds of most source files if only the
list of enabled components has changed, but not the other properties
of the build, set in config.h.
Signed-off-by: Martin Storsjö <martin@martin.st>
This reverts commit 2589060b92 which was
originally to fix the FATE test. The real cause of the test breakage was
fixed in 22b7c37275.
Signed-off-by: J. Dekker <jdek@itanimul.li>
The assembly is written assuming that the width is a multiple of 8.
However the real issue is the functions were errorneously assigned to
the 2, 4, 6 & 12 widths. This behaviour never broke the decoder as
samples which trigger the functions for these widths have not been found
in the wild. This relies on the mappings in ff_hevc_pel_weight[].
Signed-off-by: J. Dekker <jdek@itanimul.li>
Don't use the loaded registers directly, avoiding stalls on in
order cores. Use vrhadd.u8 with q registers where easily possible.
Signed-off-by: Martin Storsjö <martin@martin.st>
This unbreaks the fate-checkasm-hevc_pel test on arm targets.
The assembly assumed that the width passed to the DSP functions is
a multiple of 8, while the checkasm test used other widths too.
This wasn't noticed before, because the hevc_pel checkasm tests
(that were added in 9c513edb79 in
January) weren't run as part of fate until in
b492cacffd in August.
As this hasn't been an issue in practice with actual full decoding
tests, it seems like the actual decoder doesn't call these functions
with such widths. Therefore, we could alternatively fix the test
to only test things that the real decoder does, and this modification
could be reverted.
Signed-off-by: Martin Storsjö <martin@martin.st>
Deprecated in commits 7fc329e2dd
and 31f6a4b4b8.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Signed-off-by: James Almer <jamrial@gmail.com>
Some files currently rely on libavutil/cpu.h to include it for them;
yet said file won't use include it any more after the currently
deprecated functions are removed, so include attributes.h directly.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
No longer used by anything.
Unfortunately the old FFT_FLOAT/FFT_FIXED_32 is left as-is. It's
simply too much work for code meant to be all removed anyway.
Fix an occasional crash for hevc decoder in ARM 32 platform, the
root cause is the memory over read(read cross the memory boundary)
in SAO NENO functions ff_hevc_sao_band_filter_neon_8 and
ff_hevc_sao_edge_filter_neon_8.
After this fix, the crash disapper in the massive Android phone
test.
Signed-off-by: qoroliang <qoroliang@tencent.com>
This makes it similar to put_epel16_v6, and gives a 10-25%
speedup of this function.
Before: Cortex A7 A8 A9 A53 A72
vp8_put_epel16_h6v6_neon: 3058.0 2218.5 2459.8 2183.0 1572.2
After:
vp8_put_epel16_h6v6_neon: 2670.8 1934.2 2244.4 1729.4 1503.9
Signed-off-by: Martin Storsjö <martin@martin.st>
When targeting darwin, clang requires commas between arguments,
while the no-comma form is allowed for other targets.
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>
Clang supports the macro expansion counter (used for making unique
labels within macro expansions), but not when targeting darwin.
Convert uses of the counter into normal local labels, as used
elsewhere.
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>
When targeting darwin, clang requires commas between arguments,
while the no-comma form is allowed for other targets.
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>
Compilation error "out of range" fixed for armeabi-v7a. Compilation failed
trying to build libvlc.aar for ARM7 android on ubuntu 16.04 host. Error
messages is "Offset out of range". The reason of the error is assembler LDR
directives in function "ff_hevc_transform_luma_4x4_neon_8" need local storage
in range <1k, but no such storage provided.
Based on a patch by Ihor Bobalo <bob@eleks.com>
Suggested-by: wbs
Signed-off-by: James Almer <jamrial@gmail.com>
In binutils 2.29, the behavior of the ADR instruction changed so that 1 is
added to the address of a Thumb function (previously nothing was added). This
allows the loaded address to be passed to a BLX instruction and the correct
mode change will occur.
See: https://sourceware.org/bugzilla/show_bug.cgi?id=21458
By using adr with a label that isn't annotated as a thumb function,
we avoid the new behaviour in binutils 2.29 and get the same behaviour
as in prior releases, and as in other assemblers (ms armasm.exe,
clang's built in assembler) - an idea that Janne Grunau came up with.
Signed-off-by: Martin Storsjö <martin@martin.st>
It is redundant with costable. The first half of sintable is
identical with the second half of costable. The second half
of sintable is negative value of the first half of sintable.
The computation is changed to handle sign of sin values, in
C code and ARM assembly code.
Signed-off-by: Muhammad Faiz <mfcc64@gmail.com>
The code originally pre-multiply by 2 the steps, causing the running sum
of the h factors to drift away due to the lack of precision. It quickly
causes an inaccuracy > 0.01.
I tried diverse approaches such as multiply by 2.0 (instead of adding
the value itself) without success.
I'm unable to bench the impact of this change, feel free to compare.
This commit fixes the incoming aacpsdsp tests.
Following is an alternative simplified function (matching the incoming
AArch64 code) that may be used:
function ff_ps_stereo_interpolate_neon, export=1
vld1.32 {q0}, [r2]
vld1.32 {q1}, [r3]
ldr r12, [sp]
vmov.f32 q8, q0
vmov.f32 q9, q1
vzip.32 q8, q0
vzip.32 q9, q1
1:
vld1.32 {d4}, [r0,:64]
vld1.32 {d6}, [r1,:64]
vadd.f32 q8, q8, q9
vadd.f32 q0, q0, q1
vmov.f32 d5, d4
vmov.f32 d7, d6
vmul.f32 q2, q2, q8
vmla.f32 q2, q3, q0
vst1.32 {d4}, [r0,:64]!
vst1.32 {d5}, [r1,:64]!
subs r12, r12, #1
bgt 1b
bx lr
endfunc
clang now (in the upcoming 5.0 version) is capable of building our
arm assembly without relying on gas-preprocessor, although clang/LLVM
doesn't support .dn register aliases.
The VC1 MC assembly was only built and used if the chosen assembler
supported the .dn directives though. This was supported as long as
gas-preprocessor was used.
This means that VC1 decoding got a speed regression on clang 5.0,
unless the user manually chose using gas-preprocessor again.
By avoiding using the .dn register aliases, we can build the VC1 MC
assembly with the latest clang version.
Support for the .dn/.qn directives in clang/LLVM isn't actively planned,
see https://bugs.llvm.org/show_bug.cgi?id=18199.
This partially reverts 896a5bff64.
Signed-off-by: Martin Storsjö <martin@martin.st>
Instead, hardcode the use of the _arm implementation of add_pixels,
and use the C version for put_pixels (as no arm-optimized version
exists). Since there's separate implementations of idct{,_put,_add}
for neon, this has no practical impact on performance.
The main hevcdsp.c file calls this init function if HAVE_ARM is set,
regardless of whether neon support is available or not.
This fixes builds where neon isn't supported by the build tools at all.
Signed-off-by: Martin Storsjö <martin@martin.st>
Ben Avison
Martin Storsjö
J. Dekker
Andreas Rheinhardt
James Almer
Lynne
Anton Khirnov
qoroliang
Aman Gupta
Meng Wang
Aurelien Jacobs
Michael Niedermayer
Alexandra Hájková
Muhammad Faiz
Clément Bœsch
Seppo Tomperi
Ronald S. Bultje
Clément Bœsch