]> granicus.if.org Git - libvpx/commit
Make vp9 subpixel match vp8
authorJohann <johannkoenig@google.com>
Tue, 26 May 2015 18:30:25 +0000 (11:30 -0700)
committerJohann <johannkoenig@google.com>
Thu, 4 Jun 2015 05:10:51 +0000 (22:10 -0700)
commiteb88b172fe5e5fece6676b24a2b6b787e4901753
tree603cc8dc645403ab20e39702bdf6df2a7c2f327c
parent01853d7ce9511fb528289c44f1b1881288f10090
Make vp9 subpixel match vp8

The only difference between the two was that the vp9 function allowed
for every step in the bilinear filter (16 steps) while vp8 only allowed
for half of those. Since all the call sites in vp9 (<< 1) the input, it
only ever used the same steps as vp8.

This will allow moving the subpel variance to vpx_dsp with the rest of
the variance functions.

Change-Id: I6fa2509350a2dc610c46b3e15bde98a15a084b75
test/variance_test.cc
vp9/common/vp9_filter.h
vp9/encoder/arm/neon/vp9_variance_neon.c
vp9/encoder/vp9_mcomp.c
vp9/encoder/vp9_variance.c
vp9/encoder/x86/vp9_highbd_subpel_variance.asm
vp9/encoder/x86/vp9_subpel_variance.asm
vp9/encoder/x86/vp9_subpel_variance_impl_intrin_avx2.c