[x265] [PATCH] asm code for blockcopy_ps_24x32

[chen]

>+;-----------------------------------------------------------------------------
>+; void blockcopy_ps_%1x%2(int16_t *dest, intptr_t destStride, pixel *src, intptr_t srcStride);
>+;-----------------------------------------------------------------------------
>+%macro BLOCKCOPY_PS_W24_H2 2
>+INIT_XMM sse4
>+cglobal blockcopy_ps_%1x%2, 4, 5, 3, dest, destStride, src, srcStride
>+
>+add        r1,      r1
>+mov        r4d,     %2/2
>+pxor       m0,      m0
>+
>+.loop
>+      movu       m1,             [r2]
>+      pmovzxbw   m2,             m1
>+      movu       [r0],           m2
>+      punpckhbw  m1,             m0
>+      movu       [r0 + 16],      m1
>+
>+      movu       m1,             [r2 + 16]
movh
 
>+      pmovzxbw   m1,             m1
>+      movu       [r0 + 32],      m1
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20131111/aac3a3da/attachment.html>