Cleanup: use xorps to zero out an xmm register on ia32 too.

src/ia32/full-codegen-ia32.cc

 // Copyright 2011 the V8 project authors. All rights reserved.
 // Redistribution and use in source and binary forms, with or without
 // modification, are permitted provided that the following conditions are
 // met:
 //
 //     * Redistributions of source code must retain the above copyright
 //       notice, this list of conditions and the following disclaimer.
 //     * Redistributions in binary form must reproduce the above
 //       copyright notice, this list of conditions and the following
 //       disclaimer in the documentation and/or other materials provided
@@ skipping 2776 matching lines @@
   // Convert 32 random bits in eax to 0.(32 random bits) in a double
   // by computing:
   // ( 1.(20 0s)(32 random bits) x 2^20 ) - (1.0 x 2^20)).
   // This is implemented on both SSE2 and FPU.
   if (CpuFeatures::IsSupported(SSE2)) {
     CpuFeatures::Scope fscope(SSE2);
     __ mov(ebx, Immediate(0x49800000));  // 1.0 x 2^20 as single.
     __ movd(xmm1, Operand(ebx));
     __ movd(xmm0, Operand(eax));
     __ cvtss2sd(xmm1, xmm1);
-    __ pxor(xmm0, xmm1);
+    __ xorps(xmm0, xmm1);
     __ subsd(xmm0, xmm1);
     __ movdbl(FieldOperand(edi, HeapNumber::kValueOffset), xmm0);
   } else {
     // 0x4130000000000000 is 1.0 x 2^20 as a double.
     __ mov(FieldOperand(edi, HeapNumber::kExponentOffset),
            Immediate(0x41300000));
     __ mov(FieldOperand(edi, HeapNumber::kMantissaOffset), eax);
     __ fld_d(FieldOperand(edi, HeapNumber::kValueOffset));
     __ mov(FieldOperand(edi, HeapNumber::kMantissaOffset), Immediate(0));
     __ fld_d(FieldOperand(edi, HeapNumber::kValueOffset));
@@ skipping 1510 matching lines @@
   // And return.
   __ ret(0);
 }
 
 
 #undef __
 
 } }  // namespace v8::internal
 
 #endif  // V8_TARGET_ARCH_IA32