[x265] [PATCH 4 of 6] avx2:'integral24v' asm code-> 7.30x faster than 'C' version

[vignesh at multicorewareinc.com]

# HG changeset patch
# User Jayashri Murugan <jayashri at multicorewareinc.com>
# Date 1494222918 -19800
#      Mon May 08 11:25:18 2017 +0530
# Node ID 12017e874c9adfd6dbba05a18c3244089d0a51b3
# Parent  ecea0022176fa6c7e8fb8dc7be7b182bc19bab68
avx2:'integral24v' asm code-> 7.30x faster than 'C' version

    integral_init24v  7.30x    207.58          1515.26

diff -r ecea0022176f -r 12017e874c9a source/common/x86/seaintegral.asm
--- a/source/common/x86/seaintegral.asm	Mon May 08 11:02:46 2017 +0530
+++ b/source/common/x86/seaintegral.asm	Mon May 08 11:25:18 2017 +0530
@@ -111,8 +111,22 @@
 ;void integral_init24v_c(uint32_t *sum24, intptr_t stride)
 ;-----------------------------------------------------------------------------
 INIT_YMM avx2
-cglobal integral24v, 2, 2, 0
- 
+cglobal integral24v, 2, 4, 2
+    mov r2, r1
+    mov r3, r1
+    shl r2, 6
+    shl r3, 5
+    add r2, r3
+
+.loop
+    movu    m0, [r0]
+    movu    m1, [r0 + r2]
+    psubd   m1, m0
+    movu    [r0], m1
+    add     r0, 32
+    sub     r1, 8
+    cmp     r1, 0
+    jnz     .loop
     RET
 
 ;-----------------------------------------------------------------------------