diff options
| author | Kyle Siefring <kylesiefring@gmail.com> | 2017-10-22 19:34:19 -0400 |
|---|---|---|
| committer | Kyle Siefring <kylesiefring@gmail.com> | 2017-10-24 10:39:48 -0400 |
| commit | ae35425ae64a3d9573f85a4a92c5638a58044057 (patch) | |
| tree | d92525876e018c38873e281dd5045f72af1f11be /vp9/encoder/vp9_encoder.c | |
| parent | b3a36f7946f930caa0e96448648db60d7330c98d (diff) | |
| download | libvpx-ae35425ae64a3d9573f85a4a92c5638a58044057.tar libvpx-ae35425ae64a3d9573f85a4a92c5638a58044057.tar.gz libvpx-ae35425ae64a3d9573f85a4a92c5638a58044057.tar.bz2 libvpx-ae35425ae64a3d9573f85a4a92c5638a58044057.zip | |
Optimize convolve8 SSSE3 and AVX2 intrinsics
Changed the intrinsics to perform summation similiar to the way the assembly does.
The new code diverges from the assembly by preferring unsaturated additions.
Results for haswell
SSSE3
Horiz/Vert Size Speedup
Horiz x4 ~32%
Horiz x8 ~6%
Vert x8 ~4%
AVX2
Horiz/Vert Size Speedup
Horiz x16 ~16%
Vert x16 ~14%
BUG=webm:1471
Change-Id: I7ad98ea688c904b1ba324adf8eb977873c8b8668
Diffstat (limited to 'vp9/encoder/vp9_encoder.c')
0 files changed, 0 insertions, 0 deletions