<div dir="ltr"><div>Hello Chen,</div><div><br></div><div>Thanks for your comment. Ok i will try optimizing it. </div><div><br></div><div>Thanks,</div><div>Aasaipriya</div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Dec 5, 2014 at 5:43 AM, chen <span dir="ltr"><<a href="mailto:chenm003@163.com" target="_blank">chenm003@163.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="line-height:1.7;color:#000000;font-size:14px;font-family:arial"><div> </div><pre><br>At 2014-12-04 17:27:18,<a href="mailto:aasaipriya@multicorewareinc.com" target="_blank">aasaipriya@multicorewareinc.com</a> wrote:
># HG changeset patch
># User Aasaipriya Chandran <a href="mailto:aasaipriya@multicorewareinc.com%3E%3E#%C2%A0Date%C2%A01417684329%C2%A0-19800%3E%23%C2%A0%C2%A0%C2%A0%C2%A0%C2%A0%C2%A0Thu%C2%A0Dec%C2%A004%C2%A014:42:09%C2%A02014%C2%A0+0530%3E%23%C2%A0Node%C2%A0ID%C2%A0952f1013ee0ca5aaa25cb597990542fc8dfa8a52%3E%23%C2%A0Parent%C2%A0%C2%A0ce4394109ae51c377593164d956f6e44b8f0d3df%3Echroma_hpp[8x8]%C2%A0for%C2%A0colorspace%C2%A0i420%C2%A0in%C2%A0avx2:%C2%A0improve%C2%A0530c-%3E373c%3E" target="_blank">aasaipriya@multicorewareinc.com>
># Date 1417684329 -19800
>#      Thu Dec 04 14:42:09 2014 +0530
># Node ID 952f1013ee0ca5aaa25cb597990542fc8dfa8a52
># Parent  ce4394109ae51c377593164d956f6e44b8f0d3df
>chroma_hpp[8x8] for colorspace i420 in avx2: improve 530c->373c
>
</a>diff -r ce4394109ae5 -r 952f1013ee0c source/common/x86/ipfilter8.asm
>--- a/source/common/x86/ipfilter8.asm      Wed Nov 26 11:53:40 2014 +0530
>+++ b/source/common/x86/ipfilter8.asm      Thu Dec 04 14:42:09 2014 +0530
>@@ -1195,6 +1195,54 @@
>     jnz               .loop
>     RET
> 
>+INIT_YMM avx2
>+cglobal interp_4tap_horiz_pp_8x8, 4,6,6
</pre><pre>6 mm registers, where is m5?</pre><pre>and you have more register here, so we can unroll here, unless its can't get more perf</pre><pre> </pre><pre><div><div class="h5">>+    mov             r4d, r4m
>+
>+%ifdef PIC
>+    lea             r5, [tab_ChromaCoeff]
>+    vpbroadcastd    m0, [r5 + r4 * 4]
>+%else
>+    vpbroadcastd    m0, [tab_ChromaCoeff + r4 * 4]
>+%endif
>+
>+    movu            m1, [tab_Tm]
>+    vpbroadcastd    m2, [pw_1]
>+
>+    ; register map
>+    ; m0 - interpolate coeff
>+    ; m1 - shuffle order table
>+    ; m2 - constant word 1
>+
>+    mov             r4d,             4
>+    sub             r0, 1
>+    .loop
>+    ; Row 0
>+    vbroadcasti128    m3,           [r0]                        ; [x x x x x A 9 8 7 6 5 4 3 2 1 0]
>+    pshufb            m3,           m1
>+    pmaddubsw         m3,           m0
>+    pmaddwd           m3,           m2
>+
>+    ; Row 1
>+    vbroadcasti128    m4,           [r0 + r1]                        ; [x x x x x A 9 8 7 6 5 4 3 2 1 0]
>+    pshufb            m4,           m1
>+    pmaddubsw         m4,           m0
>+    pmaddwd           m4,           m2
>+
>+    packssdw          m3,           m4
>+    pmulhrsw          m3,           [pw_512]
>+    vextracti128      xm4,          m3,    1
>+    packuswb          xm3,          xm4
>+    pshufd            xm3,          xm3,   11011000b
>+
>+    movq              [r2],         xm3
>+    movhps            [r2 + r3],    xm3
>+    lea               r0,           [r0 + r1 * 2]
>+    lea               r2,           [r2 + r3 * 2]
>+    dec r4d
>+    jnz .loop
>+    RET
>+
> ;--------------------------------------------------------------------------------------------------------------
> ; void interp_8tap_horiz_pp_%1x%2(pixel *src, intptr_t srcStride, pixel *dst, intptr_t dstStride, int coeffIdx)
> ;--------------------------------------------------------------------------------------------------------------
>diff -r ce4394109ae5 -r 952f1013ee0c source/common/x86/ipfilter8.h
>--- a/source/common/x86/ipfilter8.h        Wed Nov 26 11:53:40 2014 +0530
>+++ b/source/common/x86/ipfilter8.h        Thu Dec 04 14:42:09 2014 +0530
>@@ -573,6 +573,7 @@
>     SETUP_CHROMA_SS_FUNC_DEF(16, 64, cpu);
> 
> CHROMA_FILTERS(_sse4);
>+CHROMA_FILTERS(_avx2);
> CHROMA_SP_FILTERS(_sse2);
> CHROMA_SP_FILTERS_SSE4(_</div></div>sse4);
> CHROMA_SS_FILTERS(_sse2);
>_______________________________________________
>x265-devel mailing list
><a href="mailto:x265-devel@videolan.org" target="_blank">x265-devel@videolan.org</a>
><a href="https://mailman.videolan.org/listinfo/x265-devel" target="_blank">https://mailman.videolan.org/listinfo/x265-devel</a>
</pre></div><br>_______________________________________________<br>
x265-devel mailing list<br>
<a href="mailto:x265-devel@videolan.org">x265-devel@videolan.org</a><br>
<a href="https://mailman.videolan.org/listinfo/x265-devel" target="_blank">https://mailman.videolan.org/listinfo/x265-devel</a><br>
<br></blockquote></div><br></div></div>