Add AVX2 support for hbd 4-tap interpolation filter.
authorchiyotsai <chiyotsai@google.com>
Fri, 26 Oct 2018 21:14:28 +0000 (14:14 -0700)
committerChi Yo Tsai <chiyotsai@google.com>
Mon, 29 Oct 2018 22:18:17 +0000 (22:18 +0000)
commit505f2ed7fc3090b2fc33d11cca571acdd4825d4c
tree036b9f833d21c31d831a43ce09b40e0ad382b8ba
parent30ef91ff7d18b42cd55abf0bbba5b1a3ec38dc0f
Add AVX2 support for hbd 4-tap interpolation filter.

Speed gain:

BIT DEPTH | 8TAP FPS | 4TAP FPS | PCT INC |
    10    |   1.69   |   1.85   |  9.46%  |
    12    |   1.64   |   1.78   |  8.54%  |

Speed test is done on jet.y4m on speed 1 profile 2 over 100 frame with
br=500.

Change-Id: I411e122553e2c466be7a26e64b4dd144efb884a9
vpx_dsp/x86/convolve.h
vpx_dsp/x86/convolve_avx2.h
vpx_dsp/x86/highbd_convolve_avx2.c
vpx_dsp/x86/vpx_asm_stubs.c