Yunqing Wang e7cd80718b Improve sad3x16 SSE2 function
Vp9_sad3x16_sse2() is heavily called in decoder, in which the
unaligned reads consume lots of cpu cycles. When CONFIG_SUBPELREFMV
is off, the unaligned offset is 1. In this situation,
we can adjust the src_ptr to be 4-byte aligned, and then do the
aligned reads. This reduced the reading time significantly. Tests
on 1080p clip showed over 2% decoder performance gain with
CONFIG_SUBPELREFM off.

Change-Id: I953afe3ac5406107933ef49d0b695eafba9a6507
2012-11-26 09:53:50 -08:00
..
2012-11-24 19:33:58 -08:00
2012-11-24 19:33:58 -08:00
2012-11-24 19:33:58 -08:00
2012-11-26 09:53:50 -08:00
2012-11-02 08:44:54 -07:00
2012-11-02 08:44:54 -07:00
2012-11-01 16:31:22 -07:00
2012-11-24 19:33:58 -08:00
2012-11-01 16:31:22 -07:00
2012-11-01 16:31:22 -07:00
2012-11-01 16:31:22 -07:00
2012-11-13 08:27:04 -08:00
2012-11-13 08:27:04 -08:00
2012-11-01 16:31:22 -07:00
2012-11-01 16:31:22 -07:00
2012-11-01 16:31:22 -07:00
2012-11-26 09:53:50 -08:00
2012-11-01 16:31:22 -07:00
2012-11-24 19:33:58 -08:00
2012-11-24 19:33:58 -08:00
2012-11-24 19:33:58 -08:00
2012-11-01 16:31:22 -07:00
2012-11-05 16:58:03 -08:00
2012-11-01 16:31:22 -07:00
2012-11-24 19:33:58 -08:00
2012-11-15 15:48:07 -08:00
2012-11-01 16:31:22 -07:00
2012-11-01 16:31:22 -07:00
2012-11-16 06:56:29 -08:00
2012-11-21 09:16:30 -08:00
2012-11-16 06:56:29 -08:00
2012-11-26 09:53:50 -08:00
2012-11-01 16:31:22 -07:00
2012-11-21 09:16:30 -08:00
2012-11-01 16:31:22 -07:00