Use packssdw which is SSE2 not packusdw which is SSSE4.

BUG=none
TEST=Sobel* on AMD cpu
R=ryanpetrie@google.com

Review URL: https://webrtc-codereview.appspot.com/3069004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@829 16f28f9a-4ce2-e073-06de-1de4eb20be90

README.chromium

 Name: libyuv
 URL: http://code.google.com/p/libyuv/
-Version: 828
+Version: 829
 License: BSD
 License File: LICENSE
 

include/libyuv/version.h

@@ -11,6 +11,6 @@
 #ifndef INCLUDE_LIBYUV_VERSION_H_  // NOLINT
 #define INCLUDE_LIBYUV_VERSION_H_
 
-#define LIBYUV_VERSION 828
+#define LIBYUV_VERSION 829
 
 #endif  // INCLUDE_LIBYUV_VERSION_H_  NOLINT

source/row_posix.cc

@@ -5878,7 +5878,7 @@ void ARGBToBayerGGRow_SSE2(const uint8* src_argb, uint8* dst_bayer,
     "psrld     $0x8,%%xmm1                     \n"
     "pand      %%xmm5,%%xmm0                   \n"
     "pand      %%xmm5,%%xmm1                   \n"
-    "packusdw  %%xmm1,%%xmm0                   \n"
+    "packssdw  %%xmm1,%%xmm0                   \n"
     "packuswb  %%xmm1,%%xmm0                   \n"
     "sub       $0x8,%2                         \n"
     "movq      %%xmm0," MEMACCESS(1) "         \n"

source/row_win.cc

@@ -6662,7 +6662,7 @@ void ARGBToBayerGGRow_SSE2(const uint8* src_argb, uint8* dst_bayer,
     psrld      xmm1, 8
     pand       xmm0, xmm5
     pand       xmm1, xmm5
-    packusdw   xmm0, xmm1
+    packssdw   xmm0, xmm1
     packuswb   xmm0, xmm1
     sub        ecx, 8
     movq       qword ptr [edx], xmm0