]> granicus.if.org Git - libvpx/commit
projects / libvpx / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 853165b) | patch
neon: vpx_quantize_b_32x32
authorJohann <johannkoenig@google.com>
Tue, 8 Aug 2017 21:05:16 +0000 (14:05 -0700)
committerJohann <johannkoenig@google.com>
Tue, 8 Aug 2017 21:05:18 +0000 (14:05 -0700)
commit93166c5e519ecbb8477e80abb4acb1420e5cc2c4
tree18fff58223373f3317d44173d1c58da8924974a5tree | snapshot
parent853165ba3971983cf3b32f909207413a85af39edcommit | diff
neon: vpx_quantize_b_32x32

With skip block the neon is about twice as fast as C.

The neon has no shortcut for coeff < zbin so it always takes the
same amount of time. Even if the C can take the shortcut, it is over
twice as fast in neon. If it can't, that gap increases to over 10x.

BUG=webm:1426

Change-Id: I400722146c1b5a5f6289f67d85fd642463d2bfc6
test/vp9_quantize_test.cc diff | blob | history
vpx_dsp/arm/quantize_neon.c diff | blob | history
vpx_dsp/vpx_dsp_rtcd_defs.pl diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.