diff options
| author | levytamar82 <levytamar82@gmail.com> | 2014-02-23 23:58:56 -0700 |
|---|---|---|
| committer | levytamar82 <levytamar82@gmail.com> | 2014-02-28 22:51:04 -0700 |
| commit | ea1490968762b96f41548d792e84c7a195bb881f (patch) | |
| tree | 25beaf337a864f27fc8169fe37889964d9eaffac /test/webm_video_source.h | |
| parent | 5a7ac03b9ea8e75f84982e5bc6d3145c8a43d41c (diff) | |
| download | libvpx-ea1490968762b96f41548d792e84c7a195bb881f.tar libvpx-ea1490968762b96f41548d792e84c7a195bb881f.tar.gz libvpx-ea1490968762b96f41548d792e84c7a195bb881f.tar.bz2 libvpx-ea1490968762b96f41548d792e84c7a195bb881f.zip | |
AVX2 SubPixel AVG Variance Optimization
Optimizing 2 functions to process 32 elements in parallel instead of 16:
1. vp9_sub_pixel_avg_variance64x64
2. vp9_sub_pixel_avg_variance32x32
both of those function were calling vp9_sub_pixel_avg_variance16xh_ssse3
instead of calling that function, it calls vp9_sub_pixel_avg_variance32xh_avx2
that is written in avx2 and process 32 elements in parallel.
This Optimization gave 80% function level gain and 2% user level gain
Change-Id: Iea694654e1b7612dc6ed11e2626208c2179502c8
Diffstat (limited to 'test/webm_video_source.h')
0 files changed, 0 insertions, 0 deletions