Introduce andps for IA32/X64

src/ia32/lithium-codegen-ia32.cc

 // Copyright 2012 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 3865 matching lines @@
 
   ASSERT(instr->value()->Equals(instr->result()));
   Representation r = instr->hydrogen()->value()->representation();
 
   CpuFeatureScope scope(masm(), SSE2);
   if (r.IsDouble()) {
     XMMRegister scratch = double_scratch0();
     XMMRegister input_reg = ToDoubleRegister(instr->value());
     __ xorps(scratch, scratch);
     __ subsd(scratch, input_reg);
-    __ pand(input_reg, scratch);
+    __ andps(input_reg, scratch);

[Benedikt Meurer, 2013/10/28 11:23:18]

Why andps and not andpd?

[Weiliang, 2013/10/28 12:41:10]

For bit-wise operation, I think there is no much difference between andps and
andpd. But andps is one byte less. 

   } else if (r.IsSmiOrInteger32()) {
     EmitIntegerMathAbs(instr);
   } else {  // Tagged case.
     DeferredMathAbsTaggedHeapNumber* deferred =
         new(zone()) DeferredMathAbsTaggedHeapNumber(this, instr, x87_stack_);
     Register input_reg = ToRegister(instr->value());
     // Smi check.
     __ JumpIfNotSmi(input_reg, deferred->entry());
     EmitIntegerMathAbs(instr);
     __ bind(deferred->exit());
@@ skipping 2481 matching lines @@
                               FixedArray::kHeaderSize - kPointerSize));
   __ bind(&done);
 }
 
 
 #undef __
 
 } }  // namespace v8::internal
 
 #endif  // V8_TARGET_ARCH_IA32