X86: Implement MOVSS.

src/arch/x86/isa/decoder/two_byte_opcodes.isa

@@ -306,8 +306,8 @@
                     }
                     // repe (0xF3)
                     0x4: decode OPCODE_OP_BOTTOM3 {
-                        0x0: WarnUnimpl::movss_Vd_Wd();
+                        0x0: MOVSS(Vd,Wd);
-                        0x1: WarnUnimpl::movss_Wd_Vd();
+                        0x1: MOVSS(Wd,Vd);
                         0x2: WarnUnimpl::movsldup_Vo_Wo();
                         0x6: WarnUnimpl::movshdup_Vo_Wo();
                         default: UD2();

src/arch/x86/isa/insts/simd128/floating_point/data_transfer/move.py

@@ -249,7 +249,31 @@ def macroop MOVLHPS_XMM_XMM {
     movfp xmmh, xmmlm, dataSize=8
 };
 
-# MOVSS
+def macroop MOVSS_XMM_XMM {
+    movfp xmml, xmmlm, dataSize=4
+};
+
+def macroop MOVSS_XMM_M {
+    lfpimm xmml, 0
+    lfpimm xmmh, 0
+    ldfp xmml, seg, sib, disp, dataSize=4
+};
+
+def macroop MOVSS_XMM_P {
+    rdip t7
+    lfpimm xmml, 0
+    lfpimm xmmh, 0
+    ldfp xmml, seg, riprel, disp, dataSize=4
+};
+
+def macroop MOVSS_M_XMM {
+    stfp xmml, seg, sib, disp, dataSize=4
+};
+
+def macroop MOVSS_P_XMM {
+    rdip t7
+    stfp xmml, seg, riprel, disp, dataSize=4
+};
 
 def macroop MOVSD_XMM_M {
     # Zero xmmh