[x64] introduce vex prefix version of float64 arithmetic binop

test/cctest/test-disasm-x64.cc

Index: test/cctest/test-disasm-x64.cc
diff --git a/test/cctest/test-disasm-x64.cc b/test/cctest/test-disasm-x64.cc
index 2ff1c93a5719d54990d7710f54657bcb8f972542..6cd58ec209bb92a3ff8e0e4123ae236e6b894f92 100644
--- a/test/cctest/test-disasm-x64.cc
+++ b/test/cctest/test-disasm-x64.cc
@@ -475,6 +475,21 @@ TEST(DisasmX64) {
     }
   }
 
+  // AVX instruction
+  {
+    if (CpuFeatures::IsSupported(AVX)) {
+      CpuFeatureScope scope(&assm, AVX);
+      __ vaddsd(xmm0, xmm1, xmm2);
+      __ vaddsd(xmm0, xmm1, Operand(rbx, rcx, times_4, 10000));
+      __ vmulsd(xmm0, xmm1, xmm2);
+      __ vmulsd(xmm0, xmm1, Operand(rbx, rcx, times_4, 10000));
+      __ vsubsd(xmm0, xmm1, xmm2);
+      __ vsubsd(xmm0, xmm1, Operand(rbx, rcx, times_4, 10000));
+      __ vdivsd(xmm0, xmm1, xmm2);
+      __ vdivsd(xmm0, xmm1, Operand(rbx, rcx, times_4, 10000));
+    }
+  }
+
   // FMA3 instruction
   {
     if (CpuFeatures::IsSupported(FMA3)) {