SSSE3/SSE4/AVX 9-way fully merged i8x8 analysis (sa8d_x9)
authorLoren Merritt <pengvado@akuvian.org>
Mon, 10 Oct 2011 05:42:36 +0000 (05:42 +0000)
committerJason Garrett-Glaser <jason@x264.com>
Sat, 22 Oct 2011 00:22:45 +0000 (17:22 -0700)
commit2f0384dcd68bb85f98fb566b70b863b40082c83e
tree9bad38cedce66212eeefd3c0cd471567f32d9b22
parent46d1f3ab24e8aead7ccb3f89a382e7c92721ba96
SSSE3/SSE4/AVX 9-way fully merged i8x8 analysis (sa8d_x9)
x86_64 only for now, due to register requirements (like sa8d_x3).

i8x8 analysis cycles (per partition):
 penryn sandybridge bulldozer
616->600  482->374  418->356  preset=faster
892->632  725->387  598->373  preset=medium
948->650  789->409  673->383  preset=slower
common/pixel.c
common/x86/pixel-a.asm
common/x86/pixel.h
common/x86/x86util.asm