Vadim Pisarevsky 43820d89b4 further improvements in split & merge; started using non-temporary store instructions (#12063) ... * 1. changed static const __m128/256 to const __m128/256 to avoid wierd instructions and calls inserted by compiler. 2. added universal intrinsics that wrap MOVNTPS and other such (non-temporary or "no cache" store) instructions. v_store_interleave() and v_store() got respective flags/overloaded variants 3. rewrote split & merge to use the "no cache" store instructions. It resulted in dramatic performance improvement when processing big arrays * hopefully, fixed some test failures where 4-channel v_store_interleave() is used * added missing implementation of the new universal intrinsics (v_store_aligned_nocache() etc.) * fixed silly typo in the new intrinsics in intrin_vsx.hpp * still trying to fix VSX compiler errors * still trying to fix VSX compiler errors * still trying to fix VSX compiler errors * still trying to fix VSX compiler errors		2018-07-26 12:04:28 +03:00
..
opencv2	further improvements in split & merge; started using non-temporary store instructions (#12063)	2018-07-26 12:04:28 +03:00