[x265] [PATCH Review Only] asm code for interp_4tap_vert_pp_4x2

[praveen at multicorewareinc.com]

# HG changeset patch
# User Praveen Tiwari
# Date 1382693966 -19800
# Node ID a20de46f58bd7cd6421b8991343a951d81038451
# Parent  4ca4da7bdd36fbef00b9eefe54c0a56bf11633f3
asm code for interp_4tap_vert_pp_4x2

diff -r 4ca4da7bdd36 -r a20de46f58bd source/common/x86/ipfilter8.asm
--- a/source/common/x86/ipfilter8.asm	Fri Oct 25 12:11:31 2013 +0530
+++ b/source/common/x86/ipfilter8.asm	Fri Oct 25 15:09:26 2013 +0530
@@ -35,6 +35,8 @@
            db 4, 5, 6, 7, 8,  9,  10, 11, 5, 6, 7, 8,  9,  10, 11, 12
            db 6, 7, 8, 9, 10, 11, 12, 13, 7, 8, 9, 10, 11, 12, 13, 14
 
+tab_Cm:    db 0, 2, 1, 3, 0, 2, 1, 3, 0, 2, 1, 3, 0, 2, 1, 3
+
 tab_c_512:  times 8 dw 512
 
 tab_ChromaCoeff: db  0, 64,  0,  0
@@ -623,3 +625,51 @@
     IPFILTER_LUMA 48, 64
     IPFILTER_LUMA 64, 16
     IPFILTER_LUMA 16, 64
+
+;-----------------------------------------------------------------------------
+; void interp_4tap_vert_pp_4x2(pixel *src, intptr_t srcStride, pixel *dst, intptr_t dstStride, int coeffIdx)
+;-----------------------------------------------------------------------------
+INIT_XMM sse4
+cglobal interp_4tap_vert_pp_4x2, 4, 6, 8
+
+mov         r4d,       r4m
+sub         r0,        r1
+
+%ifdef PIC
+lea         r5,        [tab_ChromaCoeff]
+movd        m0,        [r5 + r4 * 4]
+%else
+movd        m0,        [tab_ChromaCoeff + r4 * 4]
+%endif
+
+pshufb      m0,        [tab_Cm]
+
+mova        m1,        [tab_c_512]
+
+movd        m2,        [r0]
+movd        m3,        [r0 + r1]
+movd        m4,        [r0 + 2 * r1]
+lea         r5,        [r0 + 2 * r1]
+movd        m5,        [r5 + r1]
+
+punpcklbw   m2,        m3
+punpcklbw   m6,        m4,        m5
+punpcklbw   m2,        m6
+
+pmaddubsw   m2,        m0
+
+movd        m6,        [r0 + 4 * r1]
+
+punpcklbw   m3,        m4
+punpcklbw   m7,        m5,        m6
+punpcklbw   m3,        m7
+
+pmaddubsw   m3,        m0
+
+phaddw      m2,        m3
+
+pmulhrsw    m2,        m1
+packuswb    m2,        m2
+movd        [r2],      m2
+pextrd      [r2 + r3], m2,  1
+RET