ARMv6 optimized fdct4x4
authorTero Rintaluoma <teror@google.com>
Mon, 21 Mar 2011 11:33:45 +0000 (13:33 +0200)
committerTero Rintaluoma <teror@google.com>
Mon, 21 Mar 2011 11:33:45 +0000 (13:33 +0200)
commita61785b6a14945cddc45dd7b4ed321df10b08d36
treeeb1a6614d58632f123bd427f773051922fb46840
parent185557344ae79890d4acbbf5972972d20af1dd05
ARMv6 optimized fdct4x4

Optimized fdct4x4 (8x4) for ARMv6 instruction set.
  - No interlocks in Cortex-A8 pipeline
  - One interlock cycle in ARM11 pipeline
  - About 2.16 times faster than current C-code compiled with -O3

Change-Id: I60484ecd144365da45bb68a960d30196b59952b8
vp8/encoder/arm/arm_csystemdependent.c
vp8/encoder/arm/armv6/vp8_fast_fdct4x4_armv6.asm [new file with mode: 0644]
vp8/encoder/arm/dct_arm.c [new file with mode: 0644]
vp8/encoder/arm/dct_arm.h
vp8/vp8cx_arm.mk