arm: Implement x264_pixel_sa8d_satd_16x16_neon
authorMartin Storsjö <martin@martin.st>
Tue, 25 Aug 2015 11:38:17 +0000 (14:38 +0300)
committerHenrik Gramner <henrik@gramner.com>
Sun, 11 Oct 2015 16:44:54 +0000 (18:44 +0200)
commite8b95e92792d9353277995043757430cf3dc3bf7
tree412d6859ff7f18f5d7d659ad96e016c5f249bc05
parent6bbaa2758d53d0d6d645142d7d818c960d137a0e
arm: Implement x264_pixel_sa8d_satd_16x16_neon

This requires spilling some registers to the stack,
contray to the aarch64 version.

checkasm timing        Cortex-A7      A8     A9
sa8d_satd_16x16_neon          12936   6365   7492
sa8d_satd_16x16_separate_neon 14841   6605   8324
common/arm/pixel-a.S
common/arm/pixel.h
common/pixel.c