[x265] [PATCH] asm: assembly code for x265_pixel_satd_32x16

[chen]

>+cglobal pixel_satd_32x16, 4,8,8    ;if WIN64 && notcpuflag(avx)
>+    SATD_START_SSE2 m10, m7
>+    mov r6, r0
>+    mov r7, r2
>+%if vertical
>+    mova m7, [pw_00ff]
>+%endif
>+    call pixel_satd_16x4_internal
>+    call pixel_satd_16x4_internal
>+    call pixel_satd_16x4_internal
>+    call pixel_satd_16x4_internal
>+    lea r0, [r6 + 16]
>+    lea r2, [r7 + 16]
>+    call pixel_satd_16x4_internal
>+    call pixel_satd_16x4_internal
>+    call pixel_satd_16x4_internal
>+    call pixel_satd_16x4_internal
>+    SATD_END_SSE2 m10
>+
> %else
>+cglobal pixel_satd_32x16, 4,8,8
you can't use 8-registers in x86
 
>+%if WIN64                           ;if WIN64 && cpuflag(avx)
>+    SATD_START_SSE2 m6, m7
>+    mov r6, r0
>+    mov r7, r2
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    lea r0, [r6 + 8]
>+    lea r2, [r7 + 8]
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    lea r0, [r6 + 16]
>+    lea r2, [r7 + 16]
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    lea r0, [r6 + 24]
>+    lea r2, [r7 + 24]
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    SATD_END_SSE2 m6
>+%else                               ;if !WIN64
>+    SATD_START_SSE2 m6, m7
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    mov r0, r0mp
>+    mov r2, r2mp
>+    add r0, 8
>+    add r2, 8
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    mov r0, r0mp
>+    mov r2, r2mp
>+    add r0, 16
>+    add r2, 16
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    mov r0, r0mp
>+    mov r2, r2mp
>+    add r0, 24
>+    add r2, 24
>+    call pixel_satd_8x8_internal
>+    call pixel_satd_8x8_internal
>+    SATD_END_SSE2 m6
>+%endif
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20131111/006b1703/attachment.html>