[x265] [PATCH] Luma_hpp[64x64 , 64x48 , 64x32 , 64x16] avx2 asm code : improved 33137c->22606c , 24826c->17202c , 16726c->11560c , 7830c->5534c
Aasaipriya Chandran
aasaipriya at multicorewareinc.com
Fri Nov 21 05:39:51 CET 2014
- Previous message: [x265] [PATCH] Luma_hpp[64x64 , 64x48 , 64x32 , 64x16] avx2 asm code : improved 33137c->22606c , 24826c->17202c , 16726c->11560c , 7830c->5534c
- Next message: [x265] [PATCH] Luma_hpp[64x64 , 64x48 , 64x32 , 64x16] avx2 asm code : improved 33137c->22606c , 24826c->17202c , 16726c->11560c , 7830c->5534c
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
Hello chen,
My apologies for the last mails . I am working on remote connection,
checking it there, and then copying it to my machine and sending the patch.
And i will make changes and will send again for both the patches.
Thanks,
Aasaipriya
On Fri, Nov 21, 2014 at 2:47 AM, chen <chenm003 at 163.com> wrote:
> miss %endmacro, are you verify it?
>
>
> At 2014-11-20 15:35:02,aasaipriya at multicorewareinc.com wrote:
> ># HG changeset patch
> ># User Aasaipriya Chandran <aasaipriya at multicorewareinc.com>
> ># Date 1416391573 -19800
> ># Wed Nov 19 15:36:13 2014 +0530
> ># Node ID 3475a4a465bdbe9e2c1a9993de32e9ff1354ffe8
> ># Parent 6c307cb8d126f9571154ffa0db65bd7f41fab70b
> >Luma_hpp[64x64 , 64x48 , 64x32 , 64x16] avx2 asm code : improved 33137c->22606c , 24826c->17202c , 16726c->11560c , 7830c->5534c
> >
> >diff -r 6c307cb8d126 -r 3475a4a465bd source/common/x86/ipfilter8.asm
> >--- a/source/common/x86/ipfilter8.asm Wed Nov 19 15:25:39 2014 +0530
> >+++ b/source/common/x86/ipfilter8.asm Wed Nov 19 15:36:13 2014 +0530
> >@@ -1035,6 +1035,117 @@
> > RET
> >
> >
> >+%macro IPFILTER_LUMA_64x_avx2 2
> >+INIT_YMM avx2
> >+cglobal interp_8tap_horiz_pp_%1x%2, 4,6,8
> >+ sub r0, 3
> >+ mov r4d, r4m
> >+%ifdef PIC
> >+ lea r5, [tab_LumaCoeff]
> >+ vpbroadcastd m0, [r5 + r4 * 8]
> >+ vpbroadcastd m1, [r5 + r4 * 8 + 4]
> >+%else
> >+ vpbroadcastd m0, [tab_LumaCoeff + r4 * 8]
> >+ vpbroadcastd m1, [tab_LumaCoeff + r4 * 8 + 4]
> >+%endif
> >+ ;mova m2, [tab_Tm]
> >+ movu m3, [tab_Tm + 16]
> >+ vpbroadcastd m7, [pw_1]
> >+
> >+ ; register map
> >+ ; m0 , m1 interpolate coeff
> >+ ; m2 , m2 shuffle order table
> >+ ; m7 - pw_1
> >+
> >+ mov r4d, %2
> >+.loop:
> >+ ; Row 0
> >+ vbroadcasti128 m4, [r0] ; [x E D C B A 9 8 7 6 5 4 3 2 1 0]
> >+ pshufb m5, m4, m3
> >+ pshufb m4, [tab_Tm]
> >+ pmaddubsw m4, m0
> >+ pmaddubsw m5, m1
> >+ paddw m4, m5
> >+ pmaddwd m4, m7
> >+ vbroadcasti128 m5, [r0 + 8]
> >+ pshufb m6, m5, m3
> >+ pshufb m5, [tab_Tm]
> >+ pmaddubsw m5, m0
> >+ pmaddubsw m6, m1
> >+ paddw m5, m6
> >+ pmaddwd m5, m7
> >+ packssdw m4, m5 ; [17 16 15 14 07 06 05 04 13 12 11 10 03 02 01 00]
> >+ pmulhrsw m4, [pw_512]
> >+ vbroadcasti128 m2, [r0 + 16]
> >+ pshufb m5, m2, m3
> >+ pshufb m2, [tab_Tm]
> >+ pmaddubsw m2, m0
> >+ pmaddubsw m5, m1
> >+ paddw m2, m5
> >+ pmaddwd m2, m7
> >+ vbroadcasti128 m5, [r0 + 24]
> >+ pshufb m6, m5, m3
> >+ pshufb m5, [tab_Tm]
> >+ pmaddubsw m5, m0
> >+ pmaddubsw m6, m1
> >+ paddw m5, m6
> >+ pmaddwd m5, m7
> >+ packssdw m2, m5
> >+ pmulhrsw m2, [pw_512]
> >+ packuswb m4, m2
> >+ vpermq m4, m4, 11011000b
> >+ vextracti128 xm5, m4, 1
> >+ pshufd xm4, xm4, 11011000b
> >+ pshufd xm5, xm5, 11011000b
> >+ movu [r2], xm4
> >+ movu [r2 + 16],xm5
> >+
> >+ vbroadcasti128 m4, [r0 + 32]
> >+ pshufb m5, m4, m3
> >+ pshufb m4, [tab_Tm]
> >+ pmaddubsw m4, m0
> >+ pmaddubsw m5, m1
> >+ paddw m4, m5
> >+ pmaddwd m4, m7
> >+ vbroadcasti128 m5, [r0 + 40]
> >+ pshufb m6, m5, m3
> >+ pshufb m5, [tab_Tm]
> >+ pmaddubsw m5, m0
> >+ pmaddubsw m6, m1
> >+ paddw m5, m6
> >+ pmaddwd m5, m7
> >+ packssdw m4, m5
> >+ pmulhrsw m4, [pw_512]
> >+ vbroadcasti128 m2, [r0 + 48]
> >+ pshufb m5, m2, m3
> >+ pshufb m2, [tab_Tm]
> >+ pmaddubsw m2, m0
> >+ pmaddubsw m5, m1
> >+ paddw m2, m5
> >+ pmaddwd m2, m7
> >+ vbroadcasti128 m5, [r0 + 56]
> >+ pshufb m6, m5, m3
> >+ pshufb m5, [tab_Tm]
> >+ pmaddubsw m5, m0
> >+ pmaddubsw m6, m1
> >+ paddw m5, m6
> >+ pmaddwd m5, m7
> >+ packssdw m2, m5
> >+ pmulhrsw m2, [pw_512]
> >+ packuswb m4, m2
> >+ vpermq m4, m4, 11011000b
> >+ vextracti128 xm5, m4, 1
> >+ pshufd xm4, xm4, 11011000b
> >+ pshufd xm5, xm5, 11011000b
> >+ movu [r2 +32], xm4
> >+ movu [r2 + 48],xm5
> >+ lea r0, [r0 + r1]
> >+ lea r2, [r2 + r3]
> >+ dec r4d
> >+ jnz .loop
> >+ RET
> >+
> >+
> > ;--------------------------------------------------------------------------------------------------------------
> > ; void interp_8tap_horiz_pp_%1x%2(pixel *src, intptr_t srcStride, pixel *dst, intptr_t dstStride, int coeffIdx)
> > ;--------------------------------------------------------------------------------------------------------------
> >@@ -1053,7 +1164,10 @@
> > IPFILTER_LUMA_32x_avx2 32 , 16
> > IPFILTER_LUMA_32x_avx2 32 , 64
> >
> >-
> >+ IPFILTER_LUMA_64x_avx2 64 , 64
> >+ IPFILTER_LUMA_64x_avx2 64 , 48
> >+ IPFILTER_LUMA_64x_avx2 64 , 32
> >+ IPFILTER_LUMA_64x_avx2 64 , 16
> >
> > ;--------------------------------------------------------------------------------------------------------------
> > ; void interp_8tap_horiz_pp_%1x%2(pixel *src, intptr_t srcStride, pixel *dst, intptr_t dstStride, int coeffIdx)
> >_______________________________________________
> >x265-devel mailing list
> >x265-devel at videolan.org
> >https://mailman.videolan.org/listinfo/x265-devel
>
>
> _______________________________________________
> 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/20141121/1cbd16d6/attachment-0001.html>
- Previous message: [x265] [PATCH] Luma_hpp[64x64 , 64x48 , 64x32 , 64x16] avx2 asm code : improved 33137c->22606c , 24826c->17202c , 16726c->11560c , 7830c->5534c
- Next message: [x265] [PATCH] Luma_hpp[64x64 , 64x48 , 64x32 , 64x16] avx2 asm code : improved 33137c->22606c , 24826c->17202c , 16726c->11560c , 7830c->5534c
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
More information about the x265-devel
mailing list