1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-12 19:18:44 +02:00
Commit Graph

6 Commits

Author SHA1 Message Date
Martin Storsjö
4136405c86 aarch64: me_cmp: Don't do uaddlv once per iteration
The max height is currently documented as 16; the max difference per
pixel is 255, and a .8h element can easily contain 16*255, thus keep
accumulating in two .8h vectors, and just do the final accumulationat the
end. This should work for heights up to 256.

This requires a minor register renumbering in ff_pix_abs16_xy2_neon.

Before:       Cortex A53    A72    A73   Graviton 3
pix_abs_0_0_neon:   97.7   47.0   37.5   22.7
pix_abs_0_1_neon:  154.0   59.0   52.0   25.0
pix_abs_0_3_neon:  179.7   96.7   87.5   41.2
After:
pix_abs_0_0_neon:   96.0   39.2   31.2   22.0
pix_abs_0_1_neon:  150.7   59.7   46.2   23.7
pix_abs_0_3_neon:  175.7   83.7   81.7   38.2

Signed-off-by: Martin Storsjö <martin@martin.st>
2022-07-16 17:26:17 +03:00
Martin Storsjö
68a03f6424 aarch64: me_cmp: Switch from uabd to uabal in ff_pix_abs16_xy2_neon
Using absolute-difference-accumulate does use twice the amount of
absolute-difference instructions, but avoids the need for the
uaddl and add instructions, reducing the total number of instructions
by 3.

These can be interleaved in the rest of the calculation, to avoid
tight dependencies at the end. Unfortunately, this is marginally
slower on Cortex A53, but faster on A72 and A73.

Before:       Cortex A53    A72    A73   Graviton 3
pix_abs_0_3_neon:  175.7  109.2   92.0   41.2
After:
pix_abs_0_3_neon:  179.7   96.7   87.5   41.2

Signed-off-by: Martin Storsjö <martin@martin.st>
2022-07-16 17:25:54 +03:00
Martin Storsjö
b46de9aba4 aarch64: me_cmp: Interleave some of the loads in ff_pix_abs16_xy2_neon
Before:       Cortex A53    A72    A73   Graviton 3
pix_abs_0_3_neon:  183.7  112.7   97.5   41.2
After:
pix_abs_0_3_neon:  175.7  109.2   92.0   41.2

Signed-off-by: Martin Storsjö <martin@martin.st>
2022-07-16 17:25:44 +03:00
Martin Storsjö
02e7853fd9 libavcodec: aarch64: Don't clobber v8 in the h%4 case in ff_pix_abs16_xy2_neon
Checkasm doesn't currently test this codepath.

Signed-off-by: Martin Storsjö <martin@martin.st>
2022-07-16 17:25:11 +03:00
Hubert Mazur
01e190dc99 lavc/aarch64: Add pix_abs16_x2 neon implementation
Provide neon implementation for pix_abs16_x2 function.

Performance tests of implementation are below.
 - pix_abs_0_1_c: 283.5
 - pix_abs_0_1_neon: 39.0

Benchmarks and tests run with checkasm tool on AWS Graviton 3.

Signed-off-by: Hubert Mazur <hum@semihalf.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
2022-07-13 23:25:22 +03:00
Swinney, Jonathan
c471cc7474 lavc/aarch64: motion estimation functions in neon
- ff_pix_abs16_neon
 - ff_pix_abs16_xy2_neon

In direct micro benchmarks of these ff functions verses their C implementations,
these functions performed as follows on AWS Graviton 3.

ff_pix_abs16_neon:
pix_abs_0_0_c: 141.1
pix_abs_0_0_neon: 19.6

ff_pix_abs16_xy2_neon:
pix_abs_0_3_c: 269.1
pix_abs_0_3_neon: 39.3

Tested with:
./tests/checkasm/checkasm --test=motion --bench --disable-linux-perf

Signed-off-by: Jonathan Swinney <jswinney@amazon.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
2022-06-28 00:51:39 +03:00