projects / platform / upstream / libvpx.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 9922e43) | patch
SSSE3 assembly implementation of 8x8 Hadamard transform
authorJingning Han <jingning@google.com>
Sat, 4 Apr 2015 16:48:18 +0000 (09:48 -0700)
committerGerrit Code Review <gerrit@gerrit.golo.chromium.org>
Sat, 4 Apr 2015 16:59:37 +0000 (09:59 -0700)
commit7f629dfca4a2ea8c53fe0b42c0a7498edc7a7f72
tree9e1eeffeae256cb0b3ca8afe9b3d3827ff40f9d3tree | snapshot
parent9922e4344a40a147abab06f16c7a9d9403d1a33bcommit | diff
SSSE3 assembly implementation of 8x8 Hadamard transform

It uses about 10% less CPU cycles than the SSE2 intrinsic
implementation.

Change-Id: I91017c0c068679a214b98cdd4cff3a6facfb7499
vp9/common/vp9_rtcd_defs.pl diff | blob | history
vp9/encoder/x86/vp9_dct_ssse3_x86_64.asm diff | blob | history
Domain: Multimedia / Codec;