4 Commits

Author	SHA1	Message	Date
Yunqing Wang	f240782650	Optimize add_constant_residual function ... Optimized adding constant diff to predictor, which gave about 2% decoder performance gain. Change-Id: I47db20c31428e8c4a8f16214a85cbe386a6e9303	2013-03-07 15:49:07 -08:00
Yunqing Wang	b339aea675	Allocate 16-byte aligned diff buffer ... This was done based on John's suggestion. Change-Id: I62516a513c31fe3dbea0d6cd063df79d9e819ec8	2013-03-07 15:29:27 -08:00
Yunqing Wang	3162371544	Fix issue in add_residual intrinsic function ... Yaowu found this function had a compiling issue with MSVC because of using _mm_storel_pi((__m64 *)(dest + 0 * stride), (__m128)p0). To be safe, changed back to use integer store instruction. Also, for some build, diff could not always be 16-byte aligned. Changed that in the code. Change-Id: I9995e5446af15dad18f3c5c0bad1ae68abef6c0d	2013-03-07 09:22:27 -08:00
Yunqing Wang	943c6d7172	Optimize add_residual function ... Optimized adding diff to predictor, which gave 0.8% decoder performance gain. Change-Id: Ic920f0baa8cbd13a73fa77b7f9da83b58749f0f8	2013-03-05 16:27:45 -08:00