[x265] Fwd: [PATCH] blockcopy_sp_8x2, optimized asm code

[chen]

>+movh       [r0],       m0
>+movhps     [r0 + r1],  m0

>>change movh to movlps is better, movh+movhps is mixed float and integer path
Will movh+movhps cause any problem ? I thought movh will be faster.
In old CPU, the data across float and integer path need extra latency
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20131108/65e37f9e/attachment.html>