[x265] [PATCH] Luma_hpp[64x64 , 64x48 , 64x32 , 64x16] avx2 asm code : improved 33137c->22606c , 24826c->17202c , 16726c->11560c , 7830c->5534c
aasaipriya at multicorewareinc.com
aasaipriya at multicorewareinc.com
Thu Nov 20 08:35:02 CET 2014
# 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)
More information about the x265-devel
mailing list