projects / platform / upstream / libvpx.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1c66646) | patch
neon: vpx_quantize_b
authorJohann <johannkoenig@google.com>
Thu, 27 Jul 2017 20:25:38 +0000 (13:25 -0700)
committerJohann <johannkoenig@google.com>
Mon, 31 Jul 2017 17:38:46 +0000 (10:38 -0700)
commit2d6b5df657ded665594a7c1d2bdaa092610d1c4d
tree24bb84fc66f9478513cc7b95d788cccfa85974a7tree | snapshot
parent1c666465afc6f9cf470d54bf595a05f9101332facommit | diff
neon: vpx_quantize_b

With skip block or coeff < zbin it is about twice as fast as C.

If most coeff values are > zbin it is about 10-15x as fast as C.

BUG=webm:1426

Change-Id: I5d3c007b014a372d5ef0882b39bb48983b4131c7
test/vp9_quantize_test.cc diff | blob | history
vpx_dsp/arm/quantize_neon.c [new file with mode: 0644] blob
vpx_dsp/vpx_dsp.mk diff | blob | history
vpx_dsp/vpx_dsp_rtcd_defs.pl diff | blob | history
Domain: Multimedia / Codec;