[x265] [PATCH] asm : chroma_hps[8x8] for i420 avx2 - improved 629c->503c

chen chenm003 at 163.com
Wed Mar 11 01:43:47 CET 2015


 

At 2015-03-10 16:06:41,aasaipriya at multicorewareinc.com wrote:
># HG changeset patch
># User Aasaipriya Chandran <aasaipriya at multicorewareinc.com>
># Date 1425974572 -19800
>#      Tue Mar 10 13:32:52 2015 +0530
># Node ID 4891fef6fa3871f0f4811583bc804f7a0e2889bd
># Parent  42a84ada967bb934ecb68045aadbd039c51ade10
>asm : chroma_hps[8x8] for i420 avx2 - improved 629c->503c
>
>diff -r 42a84ada967b -r 4891fef6fa38 source/common/x86/asm-primitives.cpp
>--- a/source/common/x86/asm-primitives.cpp	Tue Mar 10 13:19:07 2015 +0530
>+++ b/source/common/x86/asm-primitives.cpp	Tue Mar 10 13:32:52 2015 +0530
>@@ -1793,6 +1793,7 @@
> 
>         p.chroma[X265_CSP_I420].pu[CHROMA_420_32x32].filter_hps = x265_interp_4tap_horiz_ps_32x32_avx2;
>         p.chroma[X265_CSP_I420].pu[CHROMA_420_16x16].filter_hps = x265_interp_4tap_horiz_ps_16x16_avx2;
>+        p.chroma[X265_CSP_I420].pu[CHROMA_420_8x8].filter_hps = x265_interp_4tap_horiz_ps_8x8_avx2;
> 
>         p.pu[LUMA_4x4].luma_vps = x265_interp_8tap_vert_ps_4x4_avx2;
> 
>diff -r 42a84ada967b -r 4891fef6fa38 source/common/x86/ipfilter8.asm
>--- a/source/common/x86/ipfilter8.asm	Tue Mar 10 13:19:07 2015 +0530
>+++ b/source/common/x86/ipfilter8.asm	Tue Mar 10 13:32:52 2015 +0530
>@@ -13974,3 +13974,78 @@
>     dec                r6d
>     jnz                .loop
>    RET
>+
>+;-----------------------------------------------------------------------------------------------------------------------------
>+; void interp_4tap_horiz_ps_8x8(pixel *src, intptr_t srcStride, int16_t *dst, intptr_t dstStride, int coeffIdx, int isRowExt)
>+;-----------------------------------------------------------------------------------------------------------------------------;
>+INIT_YMM avx2 
>+cglobal interp_4tap_horiz_ps_8x8, 4,7,6
>+    mov             r4d, r4m
>+    mov             r5d, r5m
>+    add             r3d, r3d
>+
>+%ifdef PIC
>+    lea               r6,           [tab_ChromaCoeff]
>+    vpbroadcastd      m0,           [r6 + r4 * 4]
>+%else
>+    vpbroadcastd      m0,           [tab_ChromaCoeff + r4 * 4]
>+%endif
>+
>+    vbroadcasti128     m2,           [pw_1]
>+    vbroadcasti128     m5,           [pw_2000]
>+    mova               m1,           [tab_Tm]
>+
>+    ; register map
>+    ; m0 - interpolate coeff
>+    ; m1 - shuffle order table
>+    ; m2 - constant word 1
>+
>+    mov                r6d,      4
>+    dec                r0
>+    test                r5d,     r5d
>+    je                 .loop
>+    sub                r0 ,      r1
>+    add                r6d ,     1
>+
>+.loop
>+     dec               r6d
>+    ; Row 0
>+    vbroadcasti128    m3,           [r0]
>+    pshufb            m3,           m1
>+    pmaddubsw         m3,           m0
>+    pmaddwd           m3,           m2
>+
>+    ; Row 1
>+    vbroadcasti128    m4,           [r0 + r1]
>+    pshufb            m4,           m1
>+    pmaddubsw         m4,           m0
>+    pmaddwd           m4,           m2
>+
>+    packssdw          m3,           m4
>+    psubw             m3,           m5
>+
>+    vpermq            m3,           m3,          11011000b
>+    vextracti128      xm4,          m3,     1
>+    movu              [r2],         xm3
>+    movu              [r2 + r3],    xm4
>+
>+    lea               r2,           [r2 + r3 * 2]
>+    lea               r0,           [r0 + r1 * 2]
>+    test              r6d,          r6d
>+    jnz               .loop
>+    test              r5d,         r5d
>+    je                .end
>+
>+    ;Row 11
>+    vbroadcasti128    m3,           [r0]
>+    pshufb            m3,           m1
>+    pmaddubsw         m3,           m0
>+    pmaddwd           m3,           m2
>+
>+    packssdw          m3,           m3
>+    psubw             m3,           m5
>+    vpermq            m3,           m3,          11011000b
>+    vextracti128      xm4,          m3,          1

where use this result?
>+    movu             [r2],         xm3
>+.end
>+   RET
>_______________________________________________
>x265-devel mailing list
>x265-devel at videolan.org
>https://mailman.videolan.org/listinfo/x265-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20150311/5c904f9b/attachment-0001.html>


More information about the x265-devel mailing list