New class for Date objects: caches individual date components.

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 2857 matching lines @@
   // If the object is not a value type, return the object.
   __ CmpObjectType(eax, JS_VALUE_TYPE, ebx);
   __ j(not_equal, &done, Label::kNear);
   __ mov(eax, FieldOperand(eax, JSValue::kValueOffset));
 
   __ bind(&done);
   context()->Plug(eax);
 }
 
 
+void FullCodeGenerator::EmitDateField(CallRuntime* expr) {
+  ZoneList<Expression*>* args = expr->arguments();
+  ASSERT(args->length() == 2);
+  ASSERT_NE(NULL, args->at(1)->AsLiteral());
+  int index = Smi::cast(*(args->at(1)->AsLiteral()->handle()))->value();
+
+  VisitForAccumulatorValue(args->at(0));  // Load the object.
+
+  Label done;
+  // If the object is a smi return the object.
+  __ JumpIfSmi(eax, &done, Label::kNear);

[ulan, 2012/01/25 13:08:48]

As discussed offline, it's better to crash if DateField is called not on JSDate.

[rossberg, 2012/01/25 15:48:37]

I removed all (non-debug-mode) type checks, and checked through all call-sites.
Actually uncovered a hidden bug in the existing Date.UTC.

+  // If the object is not a date type, return the object.
+  __ CmpObjectType(eax, JS_DATE_TYPE, ebx);
+  __ j(not_equal, &done, Label::kNear);
+  __ mov(eax, FieldOperand(eax, JSDate::kYearOffset + kPointerSize * index));
+
+  __ bind(&done);
+  context()->Plug(eax);
+}
+
+
 void FullCodeGenerator::EmitMathPow(CallRuntime* expr) {
   // Load the arguments on the stack and call the runtime function.
   ZoneList<Expression*>* args = expr->arguments();
   ASSERT(args->length() == 2);
   VisitForStackValue(args->at(0));
   VisitForStackValue(args->at(1));
 
   if (CpuFeatures::IsSupported(SSE2)) {
     MathPowStub stub(MathPowStub::ON_STACK);
     __ CallStub(&stub);
@@ skipping 26 matching lines @@
   // Update the write barrier.  Save the value as it will be
   // overwritten by the write barrier code and is needed afterward.
   __ mov(edx, eax);
   __ RecordWriteField(ebx, JSValue::kValueOffset, edx, ecx, kDontSaveFPRegs);
 
   __ bind(&done);
   context()->Plug(eax);
 }
 
 
+void FullCodeGenerator::EmitSetDateField(CallRuntime* expr) {
+  ZoneList<Expression*>* args = expr->arguments();
+  ASSERT(args->length() == 3);
+  ASSERT_NE(NULL, args->at(1)->AsLiteral());
+  int index = Smi::cast(*(args->at(1)->AsLiteral()->handle()))->value();
+
+  VisitForStackValue(args->at(0));  // Load the object.

[ulan, 2012/01/25 13:08:48]

Is it possible/faster to do 
VisitForAccumulatorValue(args->at(0)) and then mov ebx, eax?

[rossberg, 2012/01/25 15:48:37]

No, the second Visit can clobber any register. Keep in mind that the full
code-gen implements a stack machine.

+  VisitForAccumulatorValue(args->at(2));  // Load the value.
+  __ pop(ebx);  // eax = value. ebx = object.
+
+  Label done;
+  // If the object is a smi, return the value.
+  __ JumpIfSmi(ebx, &done, Label::kNear);

[ulan, 2012/01/25 13:08:48]

As discussed offline, it's better to crash if DateField is called not on JSDate.

[rossberg, 2012/01/25 15:48:37]

Done.

+
+  // If the object is not a value type, return the value.
+  __ CmpObjectType(ebx, JS_DATE_TYPE, ecx);
+  __ j(not_equal, &done, Label::kNear);
+
+  // Store the value.
+  __ mov(FieldOperand(ebx, JSDate::kYearOffset + kPointerSize * index), eax);
+
+  // Update the write barrier.  Save the value as it will be
+  // overwritten by the write barrier code and is needed afterward.
+  __ mov(edx, eax);
+  __ RecordWriteField(ebx, JSDate::kYearOffset + kPointerSize * index,

[ulan, 2012/01/25 13:08:48]

Can we avoid the write barrier by making sure that each field is set to either a
smi or NaN?

[rossberg, 2012/01/25 15:48:37]

Done.

+                      edx, ecx, kDontSaveFPRegs);
+
+  __ bind(&done);
+  context()->Plug(eax);
+}
+
+
 void FullCodeGenerator::EmitNumberToString(CallRuntime* expr) {
   ZoneList<Expression*>* args = expr->arguments();
   ASSERT_EQ(args->length(), 1);
 
   // Load the argument on the stack and call the stub.
   VisitForStackValue(args->at(0));
 
   NumberToStringStub stub;
   __ CallStub(&stub);
   context()->Plug(eax);
@@ skipping 1446 matching lines @@
   *context_length = 0;
   return previous_;
 }
 
 
 #undef __
 
 } }  // namespace v8::internal
 
 #endif  // V8_TARGET_ARCH_IA32