--- old/src/cpu/x86/vm/assembler_x86.hpp	2015-07-22 17:47:47.288625200 -0700
+++ new/src/cpu/x86/vm/assembler_x86.hpp	2015-07-22 17:47:46.456625200 -0700
@@ -1653,10 +1653,14 @@
   // SSE 4.1 extract
   void pextrd(Register dst, XMMRegister src, int imm8);
   void pextrq(Register dst, XMMRegister src, int imm8);
+  // SSE 2 extract
+  void pextrw(Register dst, XMMRegister src, int imm8);
 
   // SSE 4.1 insert
   void pinsrd(XMMRegister dst, Register src, int imm8);
   void pinsrq(XMMRegister dst, Register src, int imm8);
+  // SSE 2 insert
+  void pinsrw(XMMRegister dst, Register src, int imm8);
 
   // SSE4.1 packed move
   void pmovzxbw(XMMRegister dst, XMMRegister src);
@@ -1906,6 +1910,7 @@
 
   // Multiply Packed Floating-Point Values
   void mulpd(XMMRegister dst, XMMRegister src);
+  void mulpd(XMMRegister dst, Address src);
   void mulps(XMMRegister dst, XMMRegister src);
   void vmulpd(XMMRegister dst, XMMRegister nds, XMMRegister src, int vector_len);
   void vmulps(XMMRegister dst, XMMRegister nds, XMMRegister src, int vector_len);
@@ -1928,6 +1933,9 @@
   void vandpd(XMMRegister dst, XMMRegister nds, Address src, int vector_len);
   void vandps(XMMRegister dst, XMMRegister nds, Address src, int vector_len);
 
+  void unpckhpd(XMMRegister dst, XMMRegister src);
+  void unpcklpd(XMMRegister dst, XMMRegister src);
+
   // Bitwise Logical XOR of Packed Floating-Point Values
   void xorpd(XMMRegister dst, XMMRegister src);
   void xorps(XMMRegister dst, XMMRegister src);
@@ -2020,6 +2028,7 @@
 
   // And packed integers
   void pand(XMMRegister dst, XMMRegister src);
+  void pandn(XMMRegister dst, XMMRegister src);
   void vpand(XMMRegister dst, XMMRegister nds, XMMRegister src, int vector_len);
   void vpand(XMMRegister dst, XMMRegister nds, Address src, int vector_len);