[x265] [PATCH Review Only] demo asm code for pixel_add_pp_c, 8x4 block

[praveen at multicorewareinc.com]

# HG changeset patch
# User Praveen Tiwari
# Date 1383639477 -19800
# Node ID 31e895d679b47fafc0e8f0d46d49e26ae4f51382
# Parent  686b5b50279715bcfd15af8603e52c59de7d1b40
demo asm code for pixel_add_pp_c, 8x4 block

diff -r 686b5b502797 -r 31e895d679b4 source/common/x86/blockcopy8.asm
--- a/source/common/x86/blockcopy8.asm	Mon Nov 04 22:50:48 2013 -0600
+++ b/source/common/x86/blockcopy8.asm	Tue Nov 05 13:47:57 2013 +0530
@@ -796,3 +796,40 @@
 BLOCKCOPY_PP_W64_H2 64, 32
 BLOCKCOPY_PP_W64_H2 64, 48
 BLOCKCOPY_PP_W64_H2 64, 64
+
+;-----------------------------------------------------------------------------
+; void pixel_add_pp_c_8x4(pixel *dest, intptr_t destride, pixel *src0, pixel *src1, intptr_t srcstride0, intptr_t srcstride1);
+;-----------------------------------------------------------------------------
+INIT_XMM sse2
+cglobal pixel_add_pp_8x4, 4, 6, 8, dest, deststride, src0, src1
+
+mov        r4d,        r4m
+mov        r5d,        r5m
+
+movh       m0,    [r2]
+movh       m1,    [r3]
+
+movh       m2,    [r2 + r4]
+movh       m3,    [r3 + r5]
+
+movh       m4,    [r2 + 2 * r4]
+movh       m5,    [r3 + 2 * r5]
+
+lea        r2,    [r2 + 2 * r4]
+lea        r3,    [r3 + 2 * r5]
+
+movh       m6,    [r2 +  r4]
+movh       m7,    [r3 +  r5]
+
+paddusb    m0,    m1
+paddusb    m2,    m3
+paddusb    m4,    m5
+paddusb    m6,    m7
+
+movh    [r0],             m0
+movh    [r0 + r1],        m2
+movh    [r0 + 2 * r1],    m4
+
+lea     r0,               [r0 + 2 * r1]
+movh    [r0 + r1],        m6
+RET