[x265] [PATCH 1 of 6] avx2:'integral8v' asm code-> 7.32x faster than 'C' version

[vignesh at multicorewareinc.com]

# HG changeset patch
# User Vignesh Vijayakumar
# Date 1494225121 -19800
#      Mon May 08 12:02:01 2017 +0530
# Node ID 26b97f77ec3f8c23903a08537acbcb21625df472
# Parent  41611825c2f4661536500e1306db7d8c4bf7fd07
avx2:'integral8v' asm code-> 7.32x faster than 'C' version

    integral_init8v  7.32x    201.03          1470.88

diff -r 41611825c2f4 -r 26b97f77ec3f source/common/x86/seaintegral.asm
--- a/source/common/x86/seaintegral.asm	Thu May 04 19:13:48 2017 +0530
+++ b/source/common/x86/seaintegral.asm	Mon May 08 12:02:01 2017 +0530
@@ -51,8 +51,19 @@
 ;void integral_init8v_c(uint32_t *sum8, intptr_t stride)
 ;-----------------------------------------------------------------------------
 INIT_YMM avx2
-cglobal integral8v, 2, 2, 0
- 
+cglobal integral8v, 2, 3, 2
+    mov r2, r1
+    shl r2, 5
+
+.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
 
 ;-----------------------------------------------------------------------------