[x265] [PATCH] blockcopy_sp_6x8, optimized asm code

[praveen at multicorewareinc.com]

# HG changeset patch
# User Praveen Tiwari
# Date 1383927623 -19800
# Node ID 1f96f7e693f551be88259cc5f8dc156ae1690cd0
# Parent  eb99ce96cf7f261816694b6083b8d4d0d986d7df
blockcopy_sp_6x8, optimized asm code

diff -r eb99ce96cf7f -r 1f96f7e693f5 source/common/x86/blockcopy8.asm
--- a/source/common/x86/blockcopy8.asm	Fri Nov 08 21:21:21 2013 +0530
+++ b/source/common/x86/blockcopy8.asm	Fri Nov 08 21:50:23 2013 +0530
@@ -1048,47 +1048,32 @@
 ;-----------------------------------------------------------------------------
 %macro BLOCKCOPY_SP_W6_H4 2
 INIT_XMM sse4
-cglobal blockcopy_sp_6x8, 4, 7, 8, dest, destStride, src, srcStride
+cglobal blockcopy_sp_6x8, 4, 6, 2, dest, destStride, src, srcStride
 
-mov       r5d,    %2
+mov            r4d,           %2/2
 
-add       r3,      r3
-
-mova      m0,      [tab_Vm]
+add            r3,            r3
 
 .loop
-     movu      m1,      [r2]
-     movu      m2,      [r2 + r3]
-     movu      m3,      [r2 + 2 * r3]
-     lea       r4,      [r2 + 2 * r3]
-     movu      m4,      [r4 + r3]
+     movu       m0,           [r2]
+     movu       m1,           [r2 + r3]
 
-     pshufb    m1,      m0
-     pshufb    m2,      m0
-     pshufb    m3,      m0
-     pshufb    m4,      m0
 
-     movd      [r0],              m1
-     pextrw    r6,                m1,    2
-     mov       [r0 + 4],          r6w
+     packuswb   m0,           m1
 
-     movd      [r0 + r1],         m2
-     pextrw    r6,                m2,    2
-     mov       [r0 + r1 + 4],     r6w
+     movd      [r0],          m0
+     pextrw    r5,            m0,    2
+     mov       [r0 + 4],      r5w
 
-     movd      [r0 + 2 * r1],     m3
-     pextrw    r6,                m3,    2
-     mov       [r0 + 2 * r1 + 4], r6w
+     pextrw    r5,            m0,    6
+     pshufd     m0,           m0,    2
+     movd      [r0 + r1],     m0
+     mov       [r0 + r1 + 4], r5w
 
-     lea       r4,                [r0 + 2 * r1]
-     movd      [r4 + r1],         m4
-     pextrw    r6,                m4,    2
-     mov       [r4 + r1 + 4],     r6w
+     lea        r0,           [r0 + 2 * r1]
+     lea        r2,           [r2 + 2 * r3]
 
-     lea        r0,               [r0 + 4 * r1]
-     lea        r2,               [r2 + 4 * r3]
-
-     sub        r5d,              4
+     dec        r4d
      jnz        .loop
 
 RET