[PATCH RFC V2] Improve 64bit memcpy/memove for Corei7 with unaligned avx instruction

[Ling Ma]

2013/7/11, Liubov Dmitrieva <liubov.dmitrieva@gmail.com>:
> We need to check performance for core i7 with AVX before install this.
> As far as I understood you checked on Haswell only? But AVX works for
> more architectures than AVX2.
Ling:Sandy Bridge load & store 16bytes per cycle, the code is  for
haswell platform 32bytes per cycle. Our experiment shows haswell also
enhanced non-temporary buffer, it can help us to pre-fetch data but
the same operation will hurt SandyBridge.

> You missed to fix Copyright: s/2010/2013
Ling: Ok

Thanks
Ling