• levytamar82's avatar
    AVX2 SubPixel AVG Variance Optimization · ea149096
    levytamar82 authored
    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
    ea149096
Name
Last commit
Last update
..
vp9_dct32x32_avx2.c Loading commit data...
vp9_dct32x32_sse2.c Loading commit data...
vp9_dct_avx2.c Loading commit data...
vp9_dct_sse2.c Loading commit data...
vp9_error_sse2.asm Loading commit data...
vp9_mcomp_x86.h Loading commit data...
vp9_quantize_ssse3.asm Loading commit data...
vp9_sad4d_sse2.asm Loading commit data...
vp9_sad_mmx.asm Loading commit data...
vp9_sad_sse2.asm Loading commit data...
vp9_sad_sse3.asm Loading commit data...
vp9_sad_sse4.asm Loading commit data...
vp9_sad_ssse3.asm Loading commit data...
vp9_ssim_opt.asm Loading commit data...
vp9_subpel_variance.asm Loading commit data...
vp9_subpel_variance_impl_intrin_avx2.c Loading commit data...
vp9_subpel_variance_impl_sse2.asm Loading commit data...
vp9_subtract_sse2.asm Loading commit data...
vp9_temporal_filter_apply_sse2.asm Loading commit data...
vp9_variance_avx2.c Loading commit data...
vp9_variance_impl_intrin_avx2.c Loading commit data...
vp9_variance_impl_mmx.asm Loading commit data...
vp9_variance_impl_sse2.asm Loading commit data...
vp9_variance_mmx.c Loading commit data...
vp9_variance_sse2.c Loading commit data...