blob: 86a7e833835b56747f4c8765e893b8873465031a [file] [log] [blame]
// 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
// with the distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived
// from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#include "v8.h"
#if defined(V8_TARGET_ARCH_X64)
#include "x64/lithium-codegen-x64.h"
#include "code-stubs.h"
#include "stub-cache.h"
namespace v8 {
namespace internal {
// When invoking builtins, we need to record the safepoint in the middle of
// the invoke instruction sequence generated by the macro assembler.
class SafepointGenerator : public CallWrapper {
public:
SafepointGenerator(LCodeGen* codegen,
LPointerMap* pointers,
int deoptimization_index)
: codegen_(codegen),
pointers_(pointers),
deoptimization_index_(deoptimization_index) { }
virtual ~SafepointGenerator() { }
virtual void BeforeCall(int call_size) {
ASSERT(call_size >= 0);
// Ensure that we have enough space after the previous safepoint position
// for the jump generated there.
int call_end = codegen_->masm()->pc_offset() + call_size;
int prev_jump_end = codegen_->LastSafepointEnd() + kMinSafepointSize;
if (call_end < prev_jump_end) {
int padding_size = prev_jump_end - call_end;
STATIC_ASSERT(kMinSafepointSize <= 9); // One multibyte nop is enough.
codegen_->masm()->nop(padding_size);
}
}
virtual void AfterCall() {
codegen_->RecordSafepoint(pointers_, deoptimization_index_);
}
private:
static const int kMinSafepointSize =
MacroAssembler::kShortCallInstructionLength;
LCodeGen* codegen_;
LPointerMap* pointers_;
int deoptimization_index_;
};
#define __ masm()->
bool LCodeGen::GenerateCode() {
HPhase phase("Code generation", chunk());
ASSERT(is_unused());
status_ = GENERATING;
return GeneratePrologue() &&
GenerateBody() &&
GenerateDeferredCode() &&
GenerateJumpTable() &&
GenerateSafepointTable();
}
void LCodeGen::FinishCode(Handle<Code> code) {
ASSERT(is_done());
code->set_stack_slots(StackSlotCount());
code->set_safepoint_table_offset(safepoints_.GetCodeOffset());
PopulateDeoptimizationData(code);
Deoptimizer::EnsureRelocSpaceForLazyDeoptimization(code);
}
void LCodeGen::Abort(const char* format, ...) {
if (FLAG_trace_bailout) {
SmartPointer<char> name(info()->shared_info()->DebugName()->ToCString());
PrintF("Aborting LCodeGen in @\"%s\": ", *name);
va_list arguments;
va_start(arguments, format);
OS::VPrint(format, arguments);
va_end(arguments);
PrintF("\n");
}
status_ = ABORTED;
}
void LCodeGen::Comment(const char* format, ...) {
if (!FLAG_code_comments) return;
char buffer[4 * KB];
StringBuilder builder(buffer, ARRAY_SIZE(buffer));
va_list arguments;
va_start(arguments, format);
builder.AddFormattedList(format, arguments);
va_end(arguments);
// Copy the string before recording it in the assembler to avoid
// issues when the stack allocated buffer goes out of scope.
int length = builder.position();
Vector<char> copy = Vector<char>::New(length + 1);
memcpy(copy.start(), builder.Finalize(), copy.length());
masm()->RecordComment(copy.start());
}
bool LCodeGen::GeneratePrologue() {
ASSERT(is_generating());
#ifdef DEBUG
if (strlen(FLAG_stop_at) > 0 &&
info_->function()->name()->IsEqualTo(CStrVector(FLAG_stop_at))) {
__ int3();
}
#endif
__ push(rbp); // Caller's frame pointer.
__ movq(rbp, rsp);
__ push(rsi); // Callee's context.
__ push(rdi); // Callee's JS function.
// Reserve space for the stack slots needed by the code.
int slots = StackSlotCount();
if (slots > 0) {
if (FLAG_debug_code) {
__ movl(rax, Immediate(slots));
__ movq(kScratchRegister, kSlotsZapValue, RelocInfo::NONE);
Label loop;
__ bind(&loop);
__ push(kScratchRegister);
__ decl(rax);
__ j(not_zero, &loop);
} else {
__ subq(rsp, Immediate(slots * kPointerSize));
#ifdef _MSC_VER
// On windows, you may not access the stack more than one page below
// the most recently mapped page. To make the allocated area randomly
// accessible, we write to each page in turn (the value is irrelevant).
const int kPageSize = 4 * KB;
for (int offset = slots * kPointerSize - kPageSize;
offset > 0;
offset -= kPageSize) {
__ movq(Operand(rsp, offset), rax);
}
#endif
}
}
// Possibly allocate a local context.
int heap_slots = scope()->num_heap_slots() - Context::MIN_CONTEXT_SLOTS;
if (heap_slots > 0) {
Comment(";;; Allocate local context");
// Argument to NewContext is the function, which is still in rdi.
__ push(rdi);
if (heap_slots <= FastNewContextStub::kMaximumSlots) {
FastNewContextStub stub(heap_slots);
__ CallStub(&stub);
} else {
__ CallRuntime(Runtime::kNewContext, 1);
}
RecordSafepoint(Safepoint::kNoDeoptimizationIndex);
// Context is returned in both rax and rsi. It replaces the context
// passed to us. It's saved in the stack and kept live in rsi.
__ movq(Operand(rbp, StandardFrameConstants::kContextOffset), rsi);
// Copy any necessary parameters into the context.
int num_parameters = scope()->num_parameters();
for (int i = 0; i < num_parameters; i++) {
Slot* slot = scope()->parameter(i)->AsSlot();
if (slot != NULL && slot->type() == Slot::CONTEXT) {
int parameter_offset = StandardFrameConstants::kCallerSPOffset +
(num_parameters - 1 - i) * kPointerSize;
// Load parameter from stack.
__ movq(rax, Operand(rbp, parameter_offset));
// Store it in the context.
int context_offset = Context::SlotOffset(slot->index());
__ movq(Operand(rsi, context_offset), rax);
// Update the write barrier. This clobbers all involved
// registers, so we have use a third register to avoid
// clobbering rsi.
__ movq(rcx, rsi);
__ RecordWrite(rcx, context_offset, rax, rbx);
}
}
Comment(";;; End allocate local context");
}
// Trace the call.
if (FLAG_trace) {
__ CallRuntime(Runtime::kTraceEnter, 0);
}
return !is_aborted();
}
bool LCodeGen::GenerateBody() {
ASSERT(is_generating());
bool emit_instructions = true;
for (current_instruction_ = 0;
!is_aborted() && current_instruction_ < instructions_->length();
current_instruction_++) {
LInstruction* instr = instructions_->at(current_instruction_);
if (instr->IsLabel()) {
LLabel* label = LLabel::cast(instr);
emit_instructions = !label->HasReplacement();
}
if (emit_instructions) {
Comment(";;; @%d: %s.", current_instruction_, instr->Mnemonic());
instr->CompileToNative(this);
}
}
return !is_aborted();
}
LInstruction* LCodeGen::GetNextInstruction() {
if (current_instruction_ < instructions_->length() - 1) {
return instructions_->at(current_instruction_ + 1);
} else {
return NULL;
}
}
bool LCodeGen::GenerateJumpTable() {
for (int i = 0; i < jump_table_.length(); i++) {
__ bind(&jump_table_[i].label);
__ Jump(jump_table_[i].address, RelocInfo::RUNTIME_ENTRY);
}
return !is_aborted();
}
bool LCodeGen::GenerateDeferredCode() {
ASSERT(is_generating());
for (int i = 0; !is_aborted() && i < deferred_.length(); i++) {
LDeferredCode* code = deferred_[i];
__ bind(code->entry());
code->Generate();
__ jmp(code->exit());
}
// Deferred code is the last part of the instruction sequence. Mark
// the generated code as done unless we bailed out.
if (!is_aborted()) status_ = DONE;
return !is_aborted();
}
bool LCodeGen::GenerateSafepointTable() {
ASSERT(is_done());
// Ensure that there is space at the end of the code to write a number
// of jump instructions, as well as to afford writing a call near the end
// of the code.
// The jumps are used when there isn't room in the code stream to write
// a long call instruction. Instead it writes a shorter call to a
// jump instruction in the same code object.
// The calls are used when lazy deoptimizing a function and calls to a
// deoptimization function.
int short_deopts = safepoints_.CountShortDeoptimizationIntervals(
static_cast<unsigned>(MacroAssembler::kJumpInstructionLength));
int byte_count = (short_deopts) * MacroAssembler::kJumpInstructionLength;
while (byte_count-- > 0) {
__ int3();
}
safepoints_.Emit(masm(), StackSlotCount());
return !is_aborted();
}
Register LCodeGen::ToRegister(int index) const {
return Register::FromAllocationIndex(index);
}
XMMRegister LCodeGen::ToDoubleRegister(int index) const {
return XMMRegister::FromAllocationIndex(index);
}
Register LCodeGen::ToRegister(LOperand* op) const {
ASSERT(op->IsRegister());
return ToRegister(op->index());
}
XMMRegister LCodeGen::ToDoubleRegister(LOperand* op) const {
ASSERT(op->IsDoubleRegister());
return ToDoubleRegister(op->index());
}
bool LCodeGen::IsInteger32Constant(LConstantOperand* op) const {
return op->IsConstantOperand() &&
chunk_->LookupLiteralRepresentation(op).IsInteger32();
}
bool LCodeGen::IsTaggedConstant(LConstantOperand* op) const {
return op->IsConstantOperand() &&
chunk_->LookupLiteralRepresentation(op).IsTagged();
}
int LCodeGen::ToInteger32(LConstantOperand* op) const {
Handle<Object> value = chunk_->LookupLiteral(op);
ASSERT(chunk_->LookupLiteralRepresentation(op).IsInteger32());
ASSERT(static_cast<double>(static_cast<int32_t>(value->Number())) ==
value->Number());
return static_cast<int32_t>(value->Number());
}
Handle<Object> LCodeGen::ToHandle(LConstantOperand* op) const {
Handle<Object> literal = chunk_->LookupLiteral(op);
ASSERT(chunk_->LookupLiteralRepresentation(op).IsTagged());
return literal;
}
Operand LCodeGen::ToOperand(LOperand* op) const {
// Does not handle registers. In X64 assembler, plain registers are not
// representable as an Operand.
ASSERT(op->IsStackSlot() || op->IsDoubleStackSlot());
int index = op->index();
if (index >= 0) {
// Local or spill slot. Skip the frame pointer, function, and
// context in the fixed part of the frame.
return Operand(rbp, -(index + 3) * kPointerSize);
} else {
// Incoming parameter. Skip the return address.
return Operand(rbp, -(index - 1) * kPointerSize);
}
}
void LCodeGen::WriteTranslation(LEnvironment* environment,
Translation* translation) {
if (environment == NULL) return;
// The translation includes one command per value in the environment.
int translation_size = environment->values()->length();
// The output frame height does not include the parameters.
int height = translation_size - environment->parameter_count();
WriteTranslation(environment->outer(), translation);
int closure_id = DefineDeoptimizationLiteral(environment->closure());
translation->BeginFrame(environment->ast_id(), closure_id, height);
for (int i = 0; i < translation_size; ++i) {
LOperand* value = environment->values()->at(i);
// spilled_registers_ and spilled_double_registers_ are either
// both NULL or both set.
if (environment->spilled_registers() != NULL && value != NULL) {
if (value->IsRegister() &&
environment->spilled_registers()[value->index()] != NULL) {
translation->MarkDuplicate();
AddToTranslation(translation,
environment->spilled_registers()[value->index()],
environment->HasTaggedValueAt(i));
} else if (
value->IsDoubleRegister() &&
environment->spilled_double_registers()[value->index()] != NULL) {
translation->MarkDuplicate();
AddToTranslation(
translation,
environment->spilled_double_registers()[value->index()],
false);
}
}
AddToTranslation(translation, value, environment->HasTaggedValueAt(i));
}
}
void LCodeGen::AddToTranslation(Translation* translation,
LOperand* op,
bool is_tagged) {
if (op == NULL) {
// TODO(twuerthinger): Introduce marker operands to indicate that this value
// is not present and must be reconstructed from the deoptimizer. Currently
// this is only used for the arguments object.
translation->StoreArgumentsObject();
} else if (op->IsStackSlot()) {
if (is_tagged) {
translation->StoreStackSlot(op->index());
} else {
translation->StoreInt32StackSlot(op->index());
}
} else if (op->IsDoubleStackSlot()) {
translation->StoreDoubleStackSlot(op->index());
} else if (op->IsArgument()) {
ASSERT(is_tagged);
int src_index = StackSlotCount() + op->index();
translation->StoreStackSlot(src_index);
} else if (op->IsRegister()) {
Register reg = ToRegister(op);
if (is_tagged) {
translation->StoreRegister(reg);
} else {
translation->StoreInt32Register(reg);
}
} else if (op->IsDoubleRegister()) {
XMMRegister reg = ToDoubleRegister(op);
translation->StoreDoubleRegister(reg);
} else if (op->IsConstantOperand()) {
Handle<Object> literal = chunk()->LookupLiteral(LConstantOperand::cast(op));
int src_index = DefineDeoptimizationLiteral(literal);
translation->StoreLiteral(src_index);
} else {
UNREACHABLE();
}
}
void LCodeGen::CallCode(Handle<Code> code,
RelocInfo::Mode mode,
LInstruction* instr) {
ASSERT(instr != NULL);
LPointerMap* pointers = instr->pointer_map();
RecordPosition(pointers->position());
__ call(code, mode);
RegisterLazyDeoptimization(instr);
// Signal that we don't inline smi code before these stubs in the
// optimizing code generator.
if (code->kind() == Code::TYPE_RECORDING_BINARY_OP_IC ||
code->kind() == Code::COMPARE_IC) {
__ nop();
}
}
void LCodeGen::CallRuntime(const Runtime::Function* function,
int num_arguments,
LInstruction* instr) {
ASSERT(instr != NULL);
ASSERT(instr->HasPointerMap());
LPointerMap* pointers = instr->pointer_map();
RecordPosition(pointers->position());
__ CallRuntime(function, num_arguments);
RegisterLazyDeoptimization(instr);
}
void LCodeGen::RegisterLazyDeoptimization(LInstruction* instr) {
// Create the environment to bailout to. If the call has side effects
// execution has to continue after the call otherwise execution can continue
// from a previous bailout point repeating the call.
LEnvironment* deoptimization_environment;
if (instr->HasDeoptimizationEnvironment()) {
deoptimization_environment = instr->deoptimization_environment();
} else {
deoptimization_environment = instr->environment();
}
RegisterEnvironmentForDeoptimization(deoptimization_environment);
RecordSafepoint(instr->pointer_map(),
deoptimization_environment->deoptimization_index());
}
void LCodeGen::RegisterEnvironmentForDeoptimization(LEnvironment* environment) {
if (!environment->HasBeenRegistered()) {
// Physical stack frame layout:
// -x ............. -4 0 ..................................... y
// [incoming arguments] [spill slots] [pushed outgoing arguments]
// Layout of the environment:
// 0 ..................................................... size-1
// [parameters] [locals] [expression stack including arguments]
// Layout of the translation:
// 0 ........................................................ size - 1 + 4
// [expression stack including arguments] [locals] [4 words] [parameters]
// |>------------ translation_size ------------<|
int frame_count = 0;
for (LEnvironment* e = environment; e != NULL; e = e->outer()) {
++frame_count;
}
Translation translation(&translations_, frame_count);
WriteTranslation(environment, &translation);
int deoptimization_index = deoptimizations_.length();
environment->Register(deoptimization_index, translation.index());
deoptimizations_.Add(environment);
}
}
void LCodeGen::DeoptimizeIf(Condition cc, LEnvironment* environment) {
RegisterEnvironmentForDeoptimization(environment);
ASSERT(environment->HasBeenRegistered());
int id = environment->deoptimization_index();
Address entry = Deoptimizer::GetDeoptimizationEntry(id, Deoptimizer::EAGER);
ASSERT(entry != NULL);
if (entry == NULL) {
Abort("bailout was not prepared");
return;
}
if (cc == no_condition) {
__ Jump(entry, RelocInfo::RUNTIME_ENTRY);
} else {
// We often have several deopts to the same entry, reuse the last
// jump entry if this is the case.
if (jump_table_.is_empty() ||
jump_table_.last().address != entry) {
jump_table_.Add(entry);
}
__ j(cc, &jump_table_.last().label);
}
}
void LCodeGen::PopulateDeoptimizationData(Handle<Code> code) {
int length = deoptimizations_.length();
if (length == 0) return;
ASSERT(FLAG_deopt);
Handle<DeoptimizationInputData> data =
factory()->NewDeoptimizationInputData(length, TENURED);
Handle<ByteArray> translations = translations_.CreateByteArray();
data->SetTranslationByteArray(*translations);
data->SetInlinedFunctionCount(Smi::FromInt(inlined_function_count_));
Handle<FixedArray> literals =
factory()->NewFixedArray(deoptimization_literals_.length(), TENURED);
for (int i = 0; i < deoptimization_literals_.length(); i++) {
literals->set(i, *deoptimization_literals_[i]);
}
data->SetLiteralArray(*literals);
data->SetOsrAstId(Smi::FromInt(info_->osr_ast_id()));
data->SetOsrPcOffset(Smi::FromInt(osr_pc_offset_));
// Populate the deoptimization entries.
for (int i = 0; i < length; i++) {
LEnvironment* env = deoptimizations_[i];
data->SetAstId(i, Smi::FromInt(env->ast_id()));
data->SetTranslationIndex(i, Smi::FromInt(env->translation_index()));
data->SetArgumentsStackHeight(i,
Smi::FromInt(env->arguments_stack_height()));
}
code->set_deoptimization_data(*data);
}
int LCodeGen::DefineDeoptimizationLiteral(Handle<Object> literal) {
int result = deoptimization_literals_.length();
for (int i = 0; i < deoptimization_literals_.length(); ++i) {
if (deoptimization_literals_[i].is_identical_to(literal)) return i;
}
deoptimization_literals_.Add(literal);
return result;
}
void LCodeGen::PopulateDeoptimizationLiteralsWithInlinedFunctions() {
ASSERT(deoptimization_literals_.length() == 0);
const ZoneList<Handle<JSFunction> >* inlined_closures =
chunk()->inlined_closures();
for (int i = 0, length = inlined_closures->length();
i < length;
i++) {
DefineDeoptimizationLiteral(inlined_closures->at(i));
}
inlined_function_count_ = deoptimization_literals_.length();
}
void LCodeGen::RecordSafepoint(
LPointerMap* pointers,
Safepoint::Kind kind,
int arguments,
int deoptimization_index) {
const ZoneList<LOperand*>* operands = pointers->operands();
Safepoint safepoint = safepoints_.DefineSafepoint(masm(),
kind, arguments, deoptimization_index);
for (int i = 0; i < operands->length(); i++) {
LOperand* pointer = operands->at(i);
if (pointer->IsStackSlot()) {
safepoint.DefinePointerSlot(pointer->index());
} else if (pointer->IsRegister() && (kind & Safepoint::kWithRegisters)) {
safepoint.DefinePointerRegister(ToRegister(pointer));
}
}
if (kind & Safepoint::kWithRegisters) {
// Register rsi always contains a pointer to the context.
safepoint.DefinePointerRegister(rsi);
}
}
void LCodeGen::RecordSafepoint(LPointerMap* pointers,
int deoptimization_index) {
RecordSafepoint(pointers, Safepoint::kSimple, 0, deoptimization_index);
}
void LCodeGen::RecordSafepoint(int deoptimization_index) {
LPointerMap empty_pointers(RelocInfo::kNoPosition);
RecordSafepoint(&empty_pointers, deoptimization_index);
}
void LCodeGen::RecordSafepointWithRegisters(LPointerMap* pointers,
int arguments,
int deoptimization_index) {
RecordSafepoint(pointers, Safepoint::kWithRegisters, arguments,
deoptimization_index);
}
void LCodeGen::RecordPosition(int position) {
if (!FLAG_debug_info || position == RelocInfo::kNoPosition) return;
masm()->positions_recorder()->RecordPosition(position);
}
void LCodeGen::DoLabel(LLabel* label) {
if (label->is_loop_header()) {
Comment(";;; B%d - LOOP entry", label->block_id());
} else {
Comment(";;; B%d", label->block_id());
}
__ bind(label->label());
current_block_ = label->block_id();
LCodeGen::DoGap(label);
}
void LCodeGen::DoParallelMove(LParallelMove* move) {
resolver_.Resolve(move);
}
void LCodeGen::DoGap(LGap* gap) {
for (int i = LGap::FIRST_INNER_POSITION;
i <= LGap::LAST_INNER_POSITION;
i++) {
LGap::InnerPosition inner_pos = static_cast<LGap::InnerPosition>(i);
LParallelMove* move = gap->GetParallelMove(inner_pos);
if (move != NULL) DoParallelMove(move);
}
LInstruction* next = GetNextInstruction();
if (next != NULL && next->IsLazyBailout()) {
int pc = masm()->pc_offset();
safepoints_.SetPcAfterGap(pc);
}
}
void LCodeGen::DoParameter(LParameter* instr) {
// Nothing to do.
}
void LCodeGen::DoCallStub(LCallStub* instr) {
ASSERT(ToRegister(instr->result()).is(rax));
switch (instr->hydrogen()->major_key()) {
case CodeStub::RegExpConstructResult: {
RegExpConstructResultStub stub;
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
break;
}
case CodeStub::RegExpExec: {
RegExpExecStub stub;
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
break;
}
case CodeStub::SubString: {
SubStringStub stub;
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
break;
}
case CodeStub::NumberToString: {
NumberToStringStub stub;
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
break;
}
case CodeStub::StringAdd: {
StringAddStub stub(NO_STRING_ADD_FLAGS);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
break;
}
case CodeStub::StringCompare: {
StringCompareStub stub;
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
break;
}
case CodeStub::TranscendentalCache: {
TranscendentalCacheStub stub(instr->transcendental_type(),
TranscendentalCacheStub::TAGGED);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
break;
}
default:
UNREACHABLE();
}
}
void LCodeGen::DoUnknownOSRValue(LUnknownOSRValue* instr) {
// Nothing to do.
}
void LCodeGen::DoModI(LModI* instr) {
if (instr->hydrogen()->HasPowerOf2Divisor()) {
Register dividend = ToRegister(instr->InputAt(0));
int32_t divisor =
HConstant::cast(instr->hydrogen()->right())->Integer32Value();
if (divisor < 0) divisor = -divisor;
NearLabel positive_dividend, done;
__ testl(dividend, dividend);
__ j(not_sign, &positive_dividend);
__ negl(dividend);
__ andl(dividend, Immediate(divisor - 1));
__ negl(dividend);
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
__ j(not_zero, &done);
DeoptimizeIf(no_condition, instr->environment());
}
__ bind(&positive_dividend);
__ andl(dividend, Immediate(divisor - 1));
__ bind(&done);
} else {
LOperand* right = instr->InputAt(1);
Register right_reg = ToRegister(right);
ASSERT(ToRegister(instr->result()).is(rdx));
ASSERT(ToRegister(instr->InputAt(0)).is(rax));
ASSERT(!right_reg.is(rax));
ASSERT(!right_reg.is(rdx));
// Check for x % 0.
if (instr->hydrogen()->CheckFlag(HValue::kCanBeDivByZero)) {
__ testl(right_reg, right_reg);
DeoptimizeIf(zero, instr->environment());
}
// Sign extend eax to edx.
// (We are using only the low 32 bits of the values.)
__ cdq();
// Check for (0 % -x) that will produce negative zero.
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
NearLabel positive_left;
NearLabel done;
__ testl(rax, rax);
__ j(not_sign, &positive_left);
__ idivl(right_reg);
// Test the remainder for 0, because then the result would be -0.
__ testl(rdx, rdx);
__ j(not_zero, &done);
DeoptimizeIf(no_condition, instr->environment());
__ bind(&positive_left);
__ idivl(right_reg);
__ bind(&done);
} else {
__ idivl(right_reg);
}
}
}
void LCodeGen::DoDivI(LDivI* instr) {
LOperand* right = instr->InputAt(1);
ASSERT(ToRegister(instr->result()).is(rax));
ASSERT(ToRegister(instr->InputAt(0)).is(rax));
ASSERT(!ToRegister(instr->InputAt(1)).is(rax));
ASSERT(!ToRegister(instr->InputAt(1)).is(rdx));
Register left_reg = rax;
// Check for x / 0.
Register right_reg = ToRegister(right);
if (instr->hydrogen()->CheckFlag(HValue::kCanBeDivByZero)) {
__ testl(right_reg, right_reg);
DeoptimizeIf(zero, instr->environment());
}
// Check for (0 / -x) that will produce negative zero.
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
NearLabel left_not_zero;
__ testl(left_reg, left_reg);
__ j(not_zero, &left_not_zero);
__ testl(right_reg, right_reg);
DeoptimizeIf(sign, instr->environment());
__ bind(&left_not_zero);
}
// Check for (-kMinInt / -1).
if (instr->hydrogen()->CheckFlag(HValue::kCanOverflow)) {
NearLabel left_not_min_int;
__ cmpl(left_reg, Immediate(kMinInt));
__ j(not_zero, &left_not_min_int);
__ cmpl(right_reg, Immediate(-1));
DeoptimizeIf(zero, instr->environment());
__ bind(&left_not_min_int);
}
// Sign extend to rdx.
__ cdq();
__ idivl(right_reg);
// Deoptimize if remainder is not 0.
__ testl(rdx, rdx);
DeoptimizeIf(not_zero, instr->environment());
}
void LCodeGen::DoMulI(LMulI* instr) {
Register left = ToRegister(instr->InputAt(0));
LOperand* right = instr->InputAt(1);
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
__ movl(kScratchRegister, left);
}
bool can_overflow =
instr->hydrogen()->CheckFlag(HValue::kCanOverflow);
if (right->IsConstantOperand()) {
int right_value = ToInteger32(LConstantOperand::cast(right));
if (right_value == -1) {
__ negl(left);
} else if (right_value == 0) {
__ xorl(left, left);
} else if (right_value == 2) {
__ addl(left, left);
} else if (!can_overflow) {
// If the multiplication is known to not overflow, we
// can use operations that don't set the overflow flag
// correctly.
switch (right_value) {
case 1:
// Do nothing.
break;
case 3:
__ leal(left, Operand(left, left, times_2, 0));
break;
case 4:
__ shll(left, Immediate(2));
break;
case 5:
__ leal(left, Operand(left, left, times_4, 0));
break;
case 8:
__ shll(left, Immediate(3));
break;
case 9:
__ leal(left, Operand(left, left, times_8, 0));
break;
case 16:
__ shll(left, Immediate(4));
break;
default:
__ imull(left, left, Immediate(right_value));
break;
}
} else {
__ imull(left, left, Immediate(right_value));
}
} else if (right->IsStackSlot()) {
__ imull(left, ToOperand(right));
} else {
__ imull(left, ToRegister(right));
}
if (can_overflow) {
DeoptimizeIf(overflow, instr->environment());
}
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
// Bail out if the result is supposed to be negative zero.
NearLabel done;
__ testl(left, left);
__ j(not_zero, &done);
if (right->IsConstantOperand()) {
if (ToInteger32(LConstantOperand::cast(right)) <= 0) {
DeoptimizeIf(no_condition, instr->environment());
}
} else if (right->IsStackSlot()) {
__ or_(kScratchRegister, ToOperand(right));
DeoptimizeIf(sign, instr->environment());
} else {
// Test the non-zero operand for negative sign.
__ or_(kScratchRegister, ToRegister(right));
DeoptimizeIf(sign, instr->environment());
}
__ bind(&done);
}
}
void LCodeGen::DoBitI(LBitI* instr) {
LOperand* left = instr->InputAt(0);
LOperand* right = instr->InputAt(1);
ASSERT(left->Equals(instr->result()));
ASSERT(left->IsRegister());
if (right->IsConstantOperand()) {
int right_operand = ToInteger32(LConstantOperand::cast(right));
switch (instr->op()) {
case Token::BIT_AND:
__ andl(ToRegister(left), Immediate(right_operand));
break;
case Token::BIT_OR:
__ orl(ToRegister(left), Immediate(right_operand));
break;
case Token::BIT_XOR:
__ xorl(ToRegister(left), Immediate(right_operand));
break;
default:
UNREACHABLE();
break;
}
} else if (right->IsStackSlot()) {
switch (instr->op()) {
case Token::BIT_AND:
__ andl(ToRegister(left), ToOperand(right));
break;
case Token::BIT_OR:
__ orl(ToRegister(left), ToOperand(right));
break;
case Token::BIT_XOR:
__ xorl(ToRegister(left), ToOperand(right));
break;
default:
UNREACHABLE();
break;
}
} else {
ASSERT(right->IsRegister());
switch (instr->op()) {
case Token::BIT_AND:
__ andl(ToRegister(left), ToRegister(right));
break;
case Token::BIT_OR:
__ orl(ToRegister(left), ToRegister(right));
break;
case Token::BIT_XOR:
__ xorl(ToRegister(left), ToRegister(right));
break;
default:
UNREACHABLE();
break;
}
}
}
void LCodeGen::DoShiftI(LShiftI* instr) {
LOperand* left = instr->InputAt(0);
LOperand* right = instr->InputAt(1);
ASSERT(left->Equals(instr->result()));
ASSERT(left->IsRegister());
if (right->IsRegister()) {
ASSERT(ToRegister(right).is(rcx));
switch (instr->op()) {
case Token::SAR:
__ sarl_cl(ToRegister(left));
break;
case Token::SHR:
__ shrl_cl(ToRegister(left));
if (instr->can_deopt()) {
__ testl(ToRegister(left), ToRegister(left));
DeoptimizeIf(negative, instr->environment());
}
break;
case Token::SHL:
__ shll_cl(ToRegister(left));
break;
default:
UNREACHABLE();
break;
}
} else {
int value = ToInteger32(LConstantOperand::cast(right));
uint8_t shift_count = static_cast<uint8_t>(value & 0x1F);
switch (instr->op()) {
case Token::SAR:
if (shift_count != 0) {
__ sarl(ToRegister(left), Immediate(shift_count));
}
break;
case Token::SHR:
if (shift_count == 0 && instr->can_deopt()) {
__ testl(ToRegister(left), ToRegister(left));
DeoptimizeIf(negative, instr->environment());
} else {
__ shrl(ToRegister(left), Immediate(shift_count));
}
break;
case Token::SHL:
if (shift_count != 0) {
__ shll(ToRegister(left), Immediate(shift_count));
}
break;
default:
UNREACHABLE();
break;
}
}
}
void LCodeGen::DoSubI(LSubI* instr) {
LOperand* left = instr->InputAt(0);
LOperand* right = instr->InputAt(1);
ASSERT(left->Equals(instr->result()));
if (right->IsConstantOperand()) {
__ subl(ToRegister(left),
Immediate(ToInteger32(LConstantOperand::cast(right))));
} else if (right->IsRegister()) {
__ subl(ToRegister(left), ToRegister(right));
} else {
__ subl(ToRegister(left), ToOperand(right));
}
if (instr->hydrogen()->CheckFlag(HValue::kCanOverflow)) {
DeoptimizeIf(overflow, instr->environment());
}
}
void LCodeGen::DoConstantI(LConstantI* instr) {
ASSERT(instr->result()->IsRegister());
__ movl(ToRegister(instr->result()), Immediate(instr->value()));
}
void LCodeGen::DoConstantD(LConstantD* instr) {
ASSERT(instr->result()->IsDoubleRegister());
XMMRegister res = ToDoubleRegister(instr->result());
double v = instr->value();
uint64_t int_val = BitCast<uint64_t, double>(v);
// Use xor to produce +0.0 in a fast and compact way, but avoid to
// do so if the constant is -0.0.
if (int_val == 0) {
__ xorpd(res, res);
} else {
Register tmp = ToRegister(instr->TempAt(0));
__ Set(tmp, int_val);
__ movq(res, tmp);
}
}
void LCodeGen::DoConstantT(LConstantT* instr) {
ASSERT(instr->result()->IsRegister());
__ Move(ToRegister(instr->result()), instr->value());
}
void LCodeGen::DoJSArrayLength(LJSArrayLength* instr) {
Register result = ToRegister(instr->result());
Register array = ToRegister(instr->InputAt(0));
__ movq(result, FieldOperand(array, JSArray::kLengthOffset));
}
void LCodeGen::DoFixedArrayLength(LFixedArrayLength* instr) {
Register result = ToRegister(instr->result());
Register array = ToRegister(instr->InputAt(0));
__ movq(result, FieldOperand(array, FixedArray::kLengthOffset));
}
void LCodeGen::DoExternalArrayLength(LExternalArrayLength* instr) {
Register result = ToRegister(instr->result());
Register array = ToRegister(instr->InputAt(0));
__ movl(result, FieldOperand(array, ExternalPixelArray::kLengthOffset));
}
void LCodeGen::DoValueOf(LValueOf* instr) {
Register input = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
ASSERT(input.is(result));
NearLabel done;
// If the object is a smi return the object.
__ JumpIfSmi(input, &done);
// If the object is not a value type, return the object.
__ CmpObjectType(input, JS_VALUE_TYPE, kScratchRegister);
__ j(not_equal, &done);
__ movq(result, FieldOperand(input, JSValue::kValueOffset));
__ bind(&done);
}
void LCodeGen::DoBitNotI(LBitNotI* instr) {
LOperand* input = instr->InputAt(0);
ASSERT(input->Equals(instr->result()));
__ not_(ToRegister(input));
}
void LCodeGen::DoThrow(LThrow* instr) {
__ push(ToRegister(instr->InputAt(0)));
CallRuntime(Runtime::kThrow, 1, instr);
if (FLAG_debug_code) {
Comment("Unreachable code.");
__ int3();
}
}
void LCodeGen::DoAddI(LAddI* instr) {
LOperand* left = instr->InputAt(0);
LOperand* right = instr->InputAt(1);
ASSERT(left->Equals(instr->result()));
if (right->IsConstantOperand()) {
__ addl(ToRegister(left),
Immediate(ToInteger32(LConstantOperand::cast(right))));
} else if (right->IsRegister()) {
__ addl(ToRegister(left), ToRegister(right));
} else {
__ addl(ToRegister(left), ToOperand(right));
}
if (instr->hydrogen()->CheckFlag(HValue::kCanOverflow)) {
DeoptimizeIf(overflow, instr->environment());
}
}
void LCodeGen::DoArithmeticD(LArithmeticD* instr) {
XMMRegister left = ToDoubleRegister(instr->InputAt(0));
XMMRegister right = ToDoubleRegister(instr->InputAt(1));
XMMRegister result = ToDoubleRegister(instr->result());
// All operations except MOD are computed in-place.
ASSERT(instr->op() == Token::MOD || left.is(result));
switch (instr->op()) {
case Token::ADD:
__ addsd(left, right);
break;
case Token::SUB:
__ subsd(left, right);
break;
case Token::MUL:
__ mulsd(left, right);
break;
case Token::DIV:
__ divsd(left, right);
break;
case Token::MOD:
__ PrepareCallCFunction(2);
__ movsd(xmm0, left);
ASSERT(right.is(xmm1));
__ CallCFunction(
ExternalReference::double_fp_operation(Token::MOD, isolate()), 2);
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
__ movsd(result, xmm0);
break;
default:
UNREACHABLE();
break;
}
}
void LCodeGen::DoArithmeticT(LArithmeticT* instr) {
ASSERT(ToRegister(instr->InputAt(0)).is(rdx));
ASSERT(ToRegister(instr->InputAt(1)).is(rax));
ASSERT(ToRegister(instr->result()).is(rax));
TypeRecordingBinaryOpStub stub(instr->op(), NO_OVERWRITE);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
}
int LCodeGen::GetNextEmittedBlock(int block) {
for (int i = block + 1; i < graph()->blocks()->length(); ++i) {
LLabel* label = chunk_->GetLabel(i);
if (!label->HasReplacement()) return i;
}
return -1;
}
void LCodeGen::EmitBranch(int left_block, int right_block, Condition cc) {
int next_block = GetNextEmittedBlock(current_block_);
right_block = chunk_->LookupDestination(right_block);
left_block = chunk_->LookupDestination(left_block);
if (right_block == left_block) {
EmitGoto(left_block);
} else if (left_block == next_block) {
__ j(NegateCondition(cc), chunk_->GetAssemblyLabel(right_block));
} else if (right_block == next_block) {
__ j(cc, chunk_->GetAssemblyLabel(left_block));
} else {
__ j(cc, chunk_->GetAssemblyLabel(left_block));
if (cc != always) {
__ jmp(chunk_->GetAssemblyLabel(right_block));
}
}
}
void LCodeGen::DoBranch(LBranch* instr) {
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
Representation r = instr->hydrogen()->representation();
if (r.IsInteger32()) {
Register reg = ToRegister(instr->InputAt(0));
__ testl(reg, reg);
EmitBranch(true_block, false_block, not_zero);
} else if (r.IsDouble()) {
XMMRegister reg = ToDoubleRegister(instr->InputAt(0));
__ xorpd(xmm0, xmm0);
__ ucomisd(reg, xmm0);
EmitBranch(true_block, false_block, not_equal);
} else {
ASSERT(r.IsTagged());
Register reg = ToRegister(instr->InputAt(0));
HType type = instr->hydrogen()->type();
if (type.IsBoolean()) {
__ CompareRoot(reg, Heap::kTrueValueRootIndex);
EmitBranch(true_block, false_block, equal);
} else if (type.IsSmi()) {
__ SmiCompare(reg, Smi::FromInt(0));
EmitBranch(true_block, false_block, not_equal);
} else {
Label* true_label = chunk_->GetAssemblyLabel(true_block);
Label* false_label = chunk_->GetAssemblyLabel(false_block);
__ CompareRoot(reg, Heap::kUndefinedValueRootIndex);
__ j(equal, false_label);
__ CompareRoot(reg, Heap::kTrueValueRootIndex);
__ j(equal, true_label);
__ CompareRoot(reg, Heap::kFalseValueRootIndex);
__ j(equal, false_label);
__ Cmp(reg, Smi::FromInt(0));
__ j(equal, false_label);
__ JumpIfSmi(reg, true_label);
// Test for double values. Plus/minus zero and NaN are false.
NearLabel call_stub;
__ CompareRoot(FieldOperand(reg, HeapObject::kMapOffset),
Heap::kHeapNumberMapRootIndex);
__ j(not_equal, &call_stub);
// HeapNumber => false iff +0, -0, or NaN. These three cases set the
// zero flag when compared to zero using ucomisd.
__ xorpd(xmm0, xmm0);
__ ucomisd(xmm0, FieldOperand(reg, HeapNumber::kValueOffset));
__ j(zero, false_label);
__ jmp(true_label);
// The conversion stub doesn't cause garbage collections so it's
// safe to not record a safepoint after the call.
__ bind(&call_stub);
ToBooleanStub stub;
__ Pushad();
__ push(reg);
__ CallStub(&stub);
__ testq(rax, rax);
__ Popad();
EmitBranch(true_block, false_block, not_zero);
}
}
}
void LCodeGen::EmitGoto(int block, LDeferredCode* deferred_stack_check) {
block = chunk_->LookupDestination(block);
int next_block = GetNextEmittedBlock(current_block_);
if (block != next_block) {
// Perform stack overflow check if this goto needs it before jumping.
if (deferred_stack_check != NULL) {
__ CompareRoot(rsp, Heap::kStackLimitRootIndex);
__ j(above_equal, chunk_->GetAssemblyLabel(block));
__ jmp(deferred_stack_check->entry());
deferred_stack_check->SetExit(chunk_->GetAssemblyLabel(block));
} else {
__ jmp(chunk_->GetAssemblyLabel(block));
}
}
}
void LCodeGen::DoDeferredStackCheck(LGoto* instr) {
__ Pushad();
__ CallRuntimeSaveDoubles(Runtime::kStackGuard);
RecordSafepointWithRegisters(
instr->pointer_map(), 0, Safepoint::kNoDeoptimizationIndex);
__ Popad();
}
void LCodeGen::DoGoto(LGoto* instr) {
class DeferredStackCheck: public LDeferredCode {
public:
DeferredStackCheck(LCodeGen* codegen, LGoto* instr)
: LDeferredCode(codegen), instr_(instr) { }
virtual void Generate() { codegen()->DoDeferredStackCheck(instr_); }
private:
LGoto* instr_;
};
DeferredStackCheck* deferred = NULL;
if (instr->include_stack_check()) {
deferred = new DeferredStackCheck(this, instr);
}
EmitGoto(instr->block_id(), deferred);
}
inline Condition LCodeGen::TokenToCondition(Token::Value op, bool is_unsigned) {
Condition cond = no_condition;
switch (op) {
case Token::EQ:
case Token::EQ_STRICT:
cond = equal;
break;
case Token::LT:
cond = is_unsigned ? below : less;
break;
case Token::GT:
cond = is_unsigned ? above : greater;
break;
case Token::LTE:
cond = is_unsigned ? below_equal : less_equal;
break;
case Token::GTE:
cond = is_unsigned ? above_equal : greater_equal;
break;
case Token::IN:
case Token::INSTANCEOF:
default:
UNREACHABLE();
}
return cond;
}
void LCodeGen::EmitCmpI(LOperand* left, LOperand* right) {
if (right->IsConstantOperand()) {
int32_t value = ToInteger32(LConstantOperand::cast(right));
if (left->IsRegister()) {
__ cmpl(ToRegister(left), Immediate(value));
} else {
__ cmpl(ToOperand(left), Immediate(value));
}
} else if (right->IsRegister()) {
__ cmpl(ToRegister(left), ToRegister(right));
} else {
__ cmpl(ToRegister(left), ToOperand(right));
}
}
void LCodeGen::DoCmpID(LCmpID* instr) {
LOperand* left = instr->InputAt(0);
LOperand* right = instr->InputAt(1);
LOperand* result = instr->result();
NearLabel unordered;
if (instr->is_double()) {
// Don't base result on EFLAGS when a NaN is involved. Instead
// jump to the unordered case, which produces a false value.
__ ucomisd(ToDoubleRegister(left), ToDoubleRegister(right));
__ j(parity_even, &unordered);
} else {
EmitCmpI(left, right);
}
NearLabel done;
Condition cc = TokenToCondition(instr->op(), instr->is_double());
__ LoadRoot(ToRegister(result), Heap::kTrueValueRootIndex);
__ j(cc, &done);
__ bind(&unordered);
__ LoadRoot(ToRegister(result), Heap::kFalseValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoCmpIDAndBranch(LCmpIDAndBranch* instr) {
LOperand* left = instr->InputAt(0);
LOperand* right = instr->InputAt(1);
int false_block = chunk_->LookupDestination(instr->false_block_id());
int true_block = chunk_->LookupDestination(instr->true_block_id());
if (instr->is_double()) {
// Don't base result on EFLAGS when a NaN is involved. Instead
// jump to the false block.
__ ucomisd(ToDoubleRegister(left), ToDoubleRegister(right));
__ j(parity_even, chunk_->GetAssemblyLabel(false_block));
} else {
EmitCmpI(left, right);
}
Condition cc = TokenToCondition(instr->op(), instr->is_double());
EmitBranch(true_block, false_block, cc);
}
void LCodeGen::DoCmpJSObjectEq(LCmpJSObjectEq* instr) {
Register left = ToRegister(instr->InputAt(0));
Register right = ToRegister(instr->InputAt(1));
Register result = ToRegister(instr->result());
NearLabel different, done;
__ cmpq(left, right);
__ j(not_equal, &different);
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ jmp(&done);
__ bind(&different);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoCmpJSObjectEqAndBranch(LCmpJSObjectEqAndBranch* instr) {
Register left = ToRegister(instr->InputAt(0));
Register right = ToRegister(instr->InputAt(1));
int false_block = chunk_->LookupDestination(instr->false_block_id());
int true_block = chunk_->LookupDestination(instr->true_block_id());
__ cmpq(left, right);
EmitBranch(true_block, false_block, equal);
}
void LCodeGen::DoIsNull(LIsNull* instr) {
Register reg = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
// If the expression is known to be a smi, then it's
// definitely not null. Materialize false.
// Consider adding other type and representation tests too.
if (instr->hydrogen()->value()->type().IsSmi()) {
__ LoadRoot(result, Heap::kFalseValueRootIndex);
return;
}
__ CompareRoot(reg, Heap::kNullValueRootIndex);
if (instr->is_strict()) {
__ movl(result, Immediate(Heap::kTrueValueRootIndex));
NearLabel load;
__ j(equal, &load);
__ movl(result, Immediate(Heap::kFalseValueRootIndex));
__ bind(&load);
__ LoadRootIndexed(result, result, 0);
} else {
NearLabel true_value, false_value, done;
__ j(equal, &true_value);
__ CompareRoot(reg, Heap::kUndefinedValueRootIndex);
__ j(equal, &true_value);
__ JumpIfSmi(reg, &false_value);
// Check for undetectable objects by looking in the bit field in
// the map. The object has already been smi checked.
Register scratch = result;
__ movq(scratch, FieldOperand(reg, HeapObject::kMapOffset));
__ testb(FieldOperand(scratch, Map::kBitFieldOffset),
Immediate(1 << Map::kIsUndetectable));
__ j(not_zero, &true_value);
__ bind(&false_value);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ jmp(&done);
__ bind(&true_value);
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ bind(&done);
}
}
void LCodeGen::DoIsNullAndBranch(LIsNullAndBranch* instr) {
Register reg = ToRegister(instr->InputAt(0));
int false_block = chunk_->LookupDestination(instr->false_block_id());
if (instr->hydrogen()->representation().IsSpecialization() ||
instr->hydrogen()->type().IsSmi()) {
// If the expression is known to untagged or smi, then it's definitely
// not null, and it can't be a an undetectable object.
// Jump directly to the false block.
EmitGoto(false_block);
return;
}
int true_block = chunk_->LookupDestination(instr->true_block_id());
__ CompareRoot(reg, Heap::kNullValueRootIndex);
if (instr->is_strict()) {
EmitBranch(true_block, false_block, equal);
} else {
Label* true_label = chunk_->GetAssemblyLabel(true_block);
Label* false_label = chunk_->GetAssemblyLabel(false_block);
__ j(equal, true_label);
__ CompareRoot(reg, Heap::kUndefinedValueRootIndex);
__ j(equal, true_label);
__ JumpIfSmi(reg, false_label);
// Check for undetectable objects by looking in the bit field in
// the map. The object has already been smi checked.
Register scratch = ToRegister(instr->TempAt(0));
__ movq(scratch, FieldOperand(reg, HeapObject::kMapOffset));
__ testb(FieldOperand(scratch, Map::kBitFieldOffset),
Immediate(1 << Map::kIsUndetectable));
EmitBranch(true_block, false_block, not_zero);
}
}
Condition LCodeGen::EmitIsObject(Register input,
Label* is_not_object,
Label* is_object) {
ASSERT(!input.is(kScratchRegister));
__ JumpIfSmi(input, is_not_object);
__ CompareRoot(input, Heap::kNullValueRootIndex);
__ j(equal, is_object);
__ movq(kScratchRegister, FieldOperand(input, HeapObject::kMapOffset));
// Undetectable objects behave like undefined.
__ testb(FieldOperand(kScratchRegister, Map::kBitFieldOffset),
Immediate(1 << Map::kIsUndetectable));
__ j(not_zero, is_not_object);
__ movzxbl(kScratchRegister,
FieldOperand(kScratchRegister, Map::kInstanceTypeOffset));
__ cmpb(kScratchRegister, Immediate(FIRST_JS_OBJECT_TYPE));
__ j(below, is_not_object);
__ cmpb(kScratchRegister, Immediate(LAST_JS_OBJECT_TYPE));
return below_equal;
}
void LCodeGen::DoIsObject(LIsObject* instr) {
Register reg = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
Label is_false, is_true, done;
Condition true_cond = EmitIsObject(reg, &is_false, &is_true);
__ j(true_cond, &is_true);
__ bind(&is_false);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ jmp(&done);
__ bind(&is_true);
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoIsObjectAndBranch(LIsObjectAndBranch* instr) {
Register reg = ToRegister(instr->InputAt(0));
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
Label* true_label = chunk_->GetAssemblyLabel(true_block);
Label* false_label = chunk_->GetAssemblyLabel(false_block);
Condition true_cond = EmitIsObject(reg, false_label, true_label);
EmitBranch(true_block, false_block, true_cond);
}
void LCodeGen::DoIsSmi(LIsSmi* instr) {
LOperand* input_operand = instr->InputAt(0);
Register result = ToRegister(instr->result());
if (input_operand->IsRegister()) {
Register input = ToRegister(input_operand);
__ CheckSmiToIndicator(result, input);
} else {
Operand input = ToOperand(instr->InputAt(0));
__ CheckSmiToIndicator(result, input);
}
// result is zero if input is a smi, and one otherwise.
ASSERT(Heap::kFalseValueRootIndex == Heap::kTrueValueRootIndex + 1);
__ LoadRootIndexed(result, result, Heap::kTrueValueRootIndex);
}
void LCodeGen::DoIsSmiAndBranch(LIsSmiAndBranch* instr) {
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
Condition is_smi;
if (instr->InputAt(0)->IsRegister()) {
Register input = ToRegister(instr->InputAt(0));
is_smi = masm()->CheckSmi(input);
} else {
Operand input = ToOperand(instr->InputAt(0));
is_smi = masm()->CheckSmi(input);
}
EmitBranch(true_block, false_block, is_smi);
}
static InstanceType TestType(HHasInstanceType* instr) {
InstanceType from = instr->from();
InstanceType to = instr->to();
if (from == FIRST_TYPE) return to;
ASSERT(from == to || to == LAST_TYPE);
return from;
}
static Condition BranchCondition(HHasInstanceType* instr) {
InstanceType from = instr->from();
InstanceType to = instr->to();
if (from == to) return equal;
if (to == LAST_TYPE) return above_equal;
if (from == FIRST_TYPE) return below_equal;
UNREACHABLE();
return equal;
}
void LCodeGen::DoHasInstanceType(LHasInstanceType* instr) {
Register input = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
ASSERT(instr->hydrogen()->value()->representation().IsTagged());
__ testl(input, Immediate(kSmiTagMask));
NearLabel done, is_false;
__ j(zero, &is_false);
__ CmpObjectType(input, TestType(instr->hydrogen()), result);
__ j(NegateCondition(BranchCondition(instr->hydrogen())), &is_false);
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ jmp(&done);
__ bind(&is_false);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoHasInstanceTypeAndBranch(LHasInstanceTypeAndBranch* instr) {
Register input = ToRegister(instr->InputAt(0));
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
Label* false_label = chunk_->GetAssemblyLabel(false_block);
__ JumpIfSmi(input, false_label);
__ CmpObjectType(input, TestType(instr->hydrogen()), kScratchRegister);
EmitBranch(true_block, false_block, BranchCondition(instr->hydrogen()));
}
void LCodeGen::DoGetCachedArrayIndex(LGetCachedArrayIndex* instr) {
Register input = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
if (FLAG_debug_code) {
__ AbortIfNotString(input);
}
__ movl(result, FieldOperand(input, String::kHashFieldOffset));
ASSERT(String::kHashShift >= kSmiTagSize);
__ IndexFromHash(result, result);
}
void LCodeGen::DoHasCachedArrayIndex(LHasCachedArrayIndex* instr) {
Register input = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
ASSERT(instr->hydrogen()->value()->representation().IsTagged());
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ testl(FieldOperand(input, String::kHashFieldOffset),
Immediate(String::kContainsCachedArrayIndexMask));
NearLabel done;
__ j(zero, &done);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoHasCachedArrayIndexAndBranch(
LHasCachedArrayIndexAndBranch* instr) {
Register input = ToRegister(instr->InputAt(0));
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
__ testl(FieldOperand(input, String::kHashFieldOffset),
Immediate(String::kContainsCachedArrayIndexMask));
EmitBranch(true_block, false_block, equal);
}
// Branches to a label or falls through with the answer in the z flag.
// Trashes the temp register and possibly input (if it and temp are aliased).
void LCodeGen::EmitClassOfTest(Label* is_true,
Label* is_false,
Handle<String> class_name,
Register input,
Register temp) {
__ JumpIfSmi(input, is_false);
__ CmpObjectType(input, FIRST_JS_OBJECT_TYPE, temp);
__ j(below, is_false);
// Map is now in temp.
// Functions have class 'Function'.
__ CmpInstanceType(temp, JS_FUNCTION_TYPE);
if (class_name->IsEqualTo(CStrVector("Function"))) {
__ j(equal, is_true);
} else {
__ j(equal, is_false);
}
// Check if the constructor in the map is a function.
__ movq(temp, FieldOperand(temp, Map::kConstructorOffset));
// As long as JS_FUNCTION_TYPE is the last instance type and it is
// right after LAST_JS_OBJECT_TYPE, we can avoid checking for
// LAST_JS_OBJECT_TYPE.
ASSERT(LAST_TYPE == JS_FUNCTION_TYPE);
ASSERT(JS_FUNCTION_TYPE == LAST_JS_OBJECT_TYPE + 1);
// Objects with a non-function constructor have class 'Object'.
__ CmpObjectType(temp, JS_FUNCTION_TYPE, kScratchRegister);
if (class_name->IsEqualTo(CStrVector("Object"))) {
__ j(not_equal, is_true);
} else {
__ j(not_equal, is_false);
}
// temp now contains the constructor function. Grab the
// instance class name from there.
__ movq(temp, FieldOperand(temp, JSFunction::kSharedFunctionInfoOffset));
__ movq(temp, FieldOperand(temp,
SharedFunctionInfo::kInstanceClassNameOffset));
// The class name we are testing against is a symbol because it's a literal.
// The name in the constructor is a symbol because of the way the context is
// booted. This routine isn't expected to work for random API-created
// classes and it doesn't have to because you can't access it with natives
// syntax. Since both sides are symbols it is sufficient to use an identity
// comparison.
ASSERT(class_name->IsSymbol());
__ Cmp(temp, class_name);
// End with the answer in the z flag.
}
void LCodeGen::DoClassOfTest(LClassOfTest* instr) {
Register input = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
ASSERT(input.is(result));
Register temp = ToRegister(instr->TempAt(0));
Handle<String> class_name = instr->hydrogen()->class_name();
NearLabel done;
Label is_true, is_false;
EmitClassOfTest(&is_true, &is_false, class_name, input, temp);
__ j(not_equal, &is_false);
__ bind(&is_true);
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ jmp(&done);
__ bind(&is_false);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoClassOfTestAndBranch(LClassOfTestAndBranch* instr) {
Register input = ToRegister(instr->InputAt(0));
Register temp = ToRegister(instr->TempAt(0));
Handle<String> class_name = instr->hydrogen()->class_name();
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
Label* true_label = chunk_->GetAssemblyLabel(true_block);
Label* false_label = chunk_->GetAssemblyLabel(false_block);
EmitClassOfTest(true_label, false_label, class_name, input, temp);
EmitBranch(true_block, false_block, equal);
}
void LCodeGen::DoCmpMapAndBranch(LCmpMapAndBranch* instr) {
Register reg = ToRegister(instr->InputAt(0));
int true_block = instr->true_block_id();
int false_block = instr->false_block_id();
__ Cmp(FieldOperand(reg, HeapObject::kMapOffset), instr->map());
EmitBranch(true_block, false_block, equal);
}
void LCodeGen::DoInstanceOf(LInstanceOf* instr) {
InstanceofStub stub(InstanceofStub::kNoFlags);
__ push(ToRegister(instr->InputAt(0)));
__ push(ToRegister(instr->InputAt(1)));
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
NearLabel true_value, done;
__ testq(rax, rax);
__ j(zero, &true_value);
__ LoadRoot(ToRegister(instr->result()), Heap::kFalseValueRootIndex);
__ jmp(&done);
__ bind(&true_value);
__ LoadRoot(ToRegister(instr->result()), Heap::kTrueValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoInstanceOfAndBranch(LInstanceOfAndBranch* instr) {
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
InstanceofStub stub(InstanceofStub::kNoFlags);
__ push(ToRegister(instr->InputAt(0)));
__ push(ToRegister(instr->InputAt(1)));
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
__ testq(rax, rax);
EmitBranch(true_block, false_block, zero);
}
void LCodeGen::DoInstanceOfKnownGlobal(LInstanceOfKnownGlobal* instr) {
class DeferredInstanceOfKnownGlobal: public LDeferredCode {
public:
DeferredInstanceOfKnownGlobal(LCodeGen* codegen,
LInstanceOfKnownGlobal* instr)
: LDeferredCode(codegen), instr_(instr) { }
virtual void Generate() {
codegen()->DoDeferredLInstanceOfKnownGlobal(instr_, &map_check_);
}
Label* map_check() { return &map_check_; }
private:
LInstanceOfKnownGlobal* instr_;
Label map_check_;
};
DeferredInstanceOfKnownGlobal* deferred;
deferred = new DeferredInstanceOfKnownGlobal(this, instr);
Label done, false_result;
Register object = ToRegister(instr->InputAt(0));
// A Smi is not an instance of anything.
__ JumpIfSmi(object, &false_result);
// This is the inlined call site instanceof cache. The two occurences of the
// hole value will be patched to the last map/result pair generated by the
// instanceof stub.
NearLabel cache_miss;
// Use a temp register to avoid memory operands with variable lengths.
Register map = ToRegister(instr->TempAt(0));
__ movq(map, FieldOperand(object, HeapObject::kMapOffset));
__ bind(deferred->map_check()); // Label for calculating code patching.
__ movq(kScratchRegister, factory()->the_hole_value(),
RelocInfo::EMBEDDED_OBJECT);
__ cmpq(map, kScratchRegister); // Patched to cached map.
__ j(not_equal, &cache_miss);
// Patched to load either true or false.
__ LoadRoot(ToRegister(instr->result()), Heap::kTheHoleValueRootIndex);
#ifdef DEBUG
// Check that the code size between patch label and patch sites is invariant.
Label end_of_patched_code;
__ bind(&end_of_patched_code);
ASSERT(true);
#endif
__ jmp(&done);
// The inlined call site cache did not match. Check for null and string
// before calling the deferred code.
__ bind(&cache_miss); // Null is not an instance of anything.
__ CompareRoot(object, Heap::kNullValueRootIndex);
__ j(equal, &false_result);
// String values are not instances of anything.
__ JumpIfNotString(object, kScratchRegister, deferred->entry());
__ bind(&false_result);
__ LoadRoot(ToRegister(instr->result()), Heap::kFalseValueRootIndex);
__ bind(deferred->exit());
__ bind(&done);
}
void LCodeGen::DoDeferredLInstanceOfKnownGlobal(LInstanceOfKnownGlobal* instr,
Label* map_check) {
__ PushSafepointRegisters();
InstanceofStub::Flags flags = static_cast<InstanceofStub::Flags>(
InstanceofStub::kNoFlags | InstanceofStub::kCallSiteInlineCheck);
InstanceofStub stub(flags);
__ push(ToRegister(instr->InputAt(0)));
__ Push(instr->function());
Register temp = ToRegister(instr->TempAt(0));
ASSERT(temp.is(rdi));
static const int kAdditionalDelta = 16;
int delta =
masm_->SizeOfCodeGeneratedSince(map_check) + kAdditionalDelta;
__ movq(temp, Immediate(delta));
__ push(temp);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
__ movq(kScratchRegister, rax);
__ PopSafepointRegisters();
__ testq(kScratchRegister, kScratchRegister);
Label load_false;
Label done;
__ j(not_zero, &load_false);
__ LoadRoot(rax, Heap::kTrueValueRootIndex);
__ jmp(&done);
__ bind(&load_false);
__ LoadRoot(rax, Heap::kFalseValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoCmpT(LCmpT* instr) {
Token::Value op = instr->op();
Handle<Code> ic = CompareIC::GetUninitialized(op);
CallCode(ic, RelocInfo::CODE_TARGET, instr);
Condition condition = TokenToCondition(op, false);
if (op == Token::GT || op == Token::LTE) {
condition = ReverseCondition(condition);
}
NearLabel true_value, done;
__ testq(rax, rax);
__ j(condition, &true_value);
__ LoadRoot(ToRegister(instr->result()), Heap::kFalseValueRootIndex);
__ jmp(&done);
__ bind(&true_value);
__ LoadRoot(ToRegister(instr->result()), Heap::kTrueValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoCmpTAndBranch(LCmpTAndBranch* instr) {
Token::Value op = instr->op();
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
Handle<Code> ic = CompareIC::GetUninitialized(op);
CallCode(ic, RelocInfo::CODE_TARGET, instr);
// The compare stub expects compare condition and the input operands
// reversed for GT and LTE.
Condition condition = TokenToCondition(op, false);
if (op == Token::GT || op == Token::LTE) {
condition = ReverseCondition(condition);
}
__ testq(rax, rax);
EmitBranch(true_block, false_block, condition);
}
void LCodeGen::DoReturn(LReturn* instr) {
if (FLAG_trace) {
// Preserve the return value on the stack and rely on the runtime
// call to return the value in the same register.
__ push(rax);
__ CallRuntime(Runtime::kTraceExit, 1);
}
__ movq(rsp, rbp);
__ pop(rbp);
__ Ret((ParameterCount() + 1) * kPointerSize, rcx);
}
void LCodeGen::DoLoadGlobal(LLoadGlobal* instr) {
Register result = ToRegister(instr->result());
if (result.is(rax)) {
__ load_rax(instr->hydrogen()->cell().location(),
RelocInfo::GLOBAL_PROPERTY_CELL);
} else {
__ movq(result, instr->hydrogen()->cell(), RelocInfo::GLOBAL_PROPERTY_CELL);
__ movq(result, Operand(result, 0));
}
if (instr->hydrogen()->check_hole_value()) {
__ CompareRoot(result, Heap::kTheHoleValueRootIndex);
DeoptimizeIf(equal, instr->environment());
}
}
void LCodeGen::DoStoreGlobal(LStoreGlobal* instr) {
Register value = ToRegister(instr->InputAt(0));
Register temp = ToRegister(instr->TempAt(0));
ASSERT(!value.is(temp));
bool check_hole = instr->hydrogen()->check_hole_value();
if (!check_hole && value.is(rax)) {
__ store_rax(instr->hydrogen()->cell().location(),
RelocInfo::GLOBAL_PROPERTY_CELL);
return;
}
// If the cell we are storing to contains the hole it could have
// been deleted from the property dictionary. In that case, we need
// to update the property details in the property dictionary to mark
// it as no longer deleted. We deoptimize in that case.
__ movq(temp, instr->hydrogen()->cell(), RelocInfo::GLOBAL_PROPERTY_CELL);
if (check_hole) {
__ CompareRoot(Operand(temp, 0), Heap::kTheHoleValueRootIndex);
DeoptimizeIf(equal, instr->environment());
}
__ movq(Operand(temp, 0), value);
}
void LCodeGen::DoLoadContextSlot(LLoadContextSlot* instr) {
Register context = ToRegister(instr->context());
Register result = ToRegister(instr->result());
__ movq(result, ContextOperand(context, instr->slot_index()));
}
void LCodeGen::DoStoreContextSlot(LStoreContextSlot* instr) {
Register context = ToRegister(instr->context());
Register value = ToRegister(instr->value());
__ movq(ContextOperand(context, instr->slot_index()), value);
if (instr->needs_write_barrier()) {
int offset = Context::SlotOffset(instr->slot_index());
Register scratch = ToRegister(instr->TempAt(0));
__ RecordWrite(context, offset, value, scratch);
}
}
void LCodeGen::DoLoadNamedField(LLoadNamedField* instr) {
Register object = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
if (instr->hydrogen()->is_in_object()) {
__ movq(result, FieldOperand(object, instr->hydrogen()->offset()));
} else {
__ movq(result, FieldOperand(object, JSObject::kPropertiesOffset));
__ movq(result, FieldOperand(result, instr->hydrogen()->offset()));
}
}
void LCodeGen::EmitLoadField(Register result,
Register object,
Handle<Map> type,
Handle<String> name) {
LookupResult lookup;
type->LookupInDescriptors(NULL, *name, &lookup);
ASSERT(lookup.IsProperty() && lookup.type() == FIELD);
int index = lookup.GetLocalFieldIndexFromMap(*type);
int offset = index * kPointerSize;
if (index < 0) {
// Negative property indices are in-object properties, indexed
// from the end of the fixed part of the object.
__ movq(result, FieldOperand(object, offset + type->instance_size()));
} else {
// Non-negative property indices are in the properties array.
__ movq(result, FieldOperand(object, JSObject::kPropertiesOffset));
__ movq(result, FieldOperand(result, offset + FixedArray::kHeaderSize));
}
}
void LCodeGen::DoLoadNamedFieldPolymorphic(LLoadNamedFieldPolymorphic* instr) {
Register object = ToRegister(instr->object());
Register result = ToRegister(instr->result());
int map_count = instr->hydrogen()->types()->length();
Handle<String> name = instr->hydrogen()->name();
if (map_count == 0) {
ASSERT(instr->hydrogen()->need_generic());
__ Move(rcx, instr->hydrogen()->name());
Handle<Code> ic = isolate()->builtins()->LoadIC_Initialize();
CallCode(ic, RelocInfo::CODE_TARGET, instr);
} else {
NearLabel done;
for (int i = 0; i < map_count - 1; ++i) {
Handle<Map> map = instr->hydrogen()->types()->at(i);
NearLabel next;
__ Cmp(FieldOperand(object, HeapObject::kMapOffset), map);
__ j(not_equal, &next);
EmitLoadField(result, object, map, name);
__ jmp(&done);
__ bind(&next);
}
Handle<Map> map = instr->hydrogen()->types()->last();
__ Cmp(FieldOperand(object, HeapObject::kMapOffset), map);
if (instr->hydrogen()->need_generic()) {
NearLabel generic;
__ j(not_equal, &generic);
EmitLoadField(result, object, map, name);
__ jmp(&done);
__ bind(&generic);
__ Move(rcx, instr->hydrogen()->name());
Handle<Code> ic = isolate()->builtins()->LoadIC_Initialize();
CallCode(ic, RelocInfo::CODE_TARGET, instr);
} else {
DeoptimizeIf(not_equal, instr->environment());
EmitLoadField(result, object, map, name);
}
__ bind(&done);
}
}
void LCodeGen::DoLoadNamedGeneric(LLoadNamedGeneric* instr) {
ASSERT(ToRegister(instr->object()).is(rax));
ASSERT(ToRegister(instr->result()).is(rax));
__ Move(rcx, instr->name());
Handle<Code> ic = isolate()->builtins()->LoadIC_Initialize();
CallCode(ic, RelocInfo::CODE_TARGET, instr);
}
void LCodeGen::DoLoadFunctionPrototype(LLoadFunctionPrototype* instr) {
Register function = ToRegister(instr->function());
Register result = ToRegister(instr->result());
// Check that the function really is a function.
__ CmpObjectType(function, JS_FUNCTION_TYPE, result);
DeoptimizeIf(not_equal, instr->environment());
// Check whether the function has an instance prototype.
NearLabel non_instance;
__ testb(FieldOperand(result, Map::kBitFieldOffset),
Immediate(1 << Map::kHasNonInstancePrototype));
__ j(not_zero, &non_instance);
// Get the prototype or initial map from the function.
__ movq(result,
FieldOperand(function, JSFunction::kPrototypeOrInitialMapOffset));
// Check that the function has a prototype or an initial map.
__ CompareRoot(result, Heap::kTheHoleValueRootIndex);
DeoptimizeIf(equal, instr->environment());
// If the function does not have an initial map, we're done.
NearLabel done;
__ CmpObjectType(result, MAP_TYPE, kScratchRegister);
__ j(not_equal, &done);
// Get the prototype from the initial map.
__ movq(result, FieldOperand(result, Map::kPrototypeOffset));
__ jmp(&done);
// Non-instance prototype: Fetch prototype from constructor field
// in the function's map.
__ bind(&non_instance);
__ movq(result, FieldOperand(result, Map::kConstructorOffset));
// All done.
__ bind(&done);
}
void LCodeGen::DoLoadElements(LLoadElements* instr) {
Register result = ToRegister(instr->result());
Register input = ToRegister(instr->InputAt(0));
__ movq(result, FieldOperand(input, JSObject::kElementsOffset));
if (FLAG_debug_code) {
NearLabel done;
__ CompareRoot(FieldOperand(result, HeapObject::kMapOffset),
Heap::kFixedArrayMapRootIndex);
__ j(equal, &done);
__ CompareRoot(FieldOperand(result, HeapObject::kMapOffset),
Heap::kFixedCOWArrayMapRootIndex);
__ j(equal, &done);
Register temp((result.is(rax)) ? rbx : rax);
__ push(temp);
__ movq(temp, FieldOperand(result, HeapObject::kMapOffset));
__ movzxbq(temp, FieldOperand(temp, Map::kInstanceTypeOffset));
__ subq(temp, Immediate(FIRST_EXTERNAL_ARRAY_TYPE));
__ cmpq(temp, Immediate(kExternalArrayTypeCount));
__ pop(temp);
__ Check(below, "Check for fast elements failed.");
__ bind(&done);
}
}
void LCodeGen::DoLoadExternalArrayPointer(
LLoadExternalArrayPointer* instr) {
Register result = ToRegister(instr->result());
Register input = ToRegister(instr->InputAt(0));
__ movq(result, FieldOperand(input,
ExternalPixelArray::kExternalPointerOffset));
}
void LCodeGen::DoAccessArgumentsAt(LAccessArgumentsAt* instr) {
Register arguments = ToRegister(instr->arguments());
Register length = ToRegister(instr->length());
Register result = ToRegister(instr->result());
if (instr->index()->IsRegister()) {
__ subl(length, ToRegister(instr->index()));
} else {
__ subl(length, ToOperand(instr->index()));
}
DeoptimizeIf(below_equal, instr->environment());
// There are two words between the frame pointer and the last argument.
// Subtracting from length accounts for one of them add one more.
__ movq(result, Operand(arguments, length, times_pointer_size, kPointerSize));
}
void LCodeGen::DoLoadKeyedFastElement(LLoadKeyedFastElement* instr) {
Register elements = ToRegister(instr->elements());
Register key = ToRegister(instr->key());
Register result = ToRegister(instr->result());
ASSERT(result.is(elements));
// Load the result.
__ movq(result, FieldOperand(elements,
key,
times_pointer_size,
FixedArray::kHeaderSize));
// Check for the hole value.
__ CompareRoot(result, Heap::kTheHoleValueRootIndex);
DeoptimizeIf(equal, instr->environment());
}
void LCodeGen::DoLoadKeyedSpecializedArrayElement(
LLoadKeyedSpecializedArrayElement* instr) {
Register external_pointer = ToRegister(instr->external_pointer());
Register key = ToRegister(instr->key());
ExternalArrayType array_type = instr->array_type();
if (array_type == kExternalFloatArray) {
XMMRegister result(ToDoubleRegister(instr->result()));
__ movss(result, Operand(external_pointer, key, times_4, 0));
__ cvtss2sd(result, result);
} else {
Register result(ToRegister(instr->result()));
switch (array_type) {
case kExternalByteArray:
__ movsxbq(result, Operand(external_pointer, key, times_1, 0));
break;
case kExternalUnsignedByteArray:
case kExternalPixelArray:
__ movzxbq(result, Operand(external_pointer, key, times_1, 0));
break;
case kExternalShortArray:
__ movsxwq(result, Operand(external_pointer, key, times_2, 0));
break;
case kExternalUnsignedShortArray:
__ movzxwq(result, Operand(external_pointer, key, times_2, 0));
break;
case kExternalIntArray:
__ movsxlq(result, Operand(external_pointer, key, times_4, 0));
break;
case kExternalUnsignedIntArray:
__ movl(result, Operand(external_pointer, key, times_4, 0));
__ testl(result, result);
// TODO(danno): we could be more clever here, perhaps having a special
// version of the stub that detects if the overflow case actually
// happens, and generate code that returns a double rather than int.
DeoptimizeIf(negative, instr->environment());
break;
case kExternalFloatArray:
UNREACHABLE();
break;
}
}
}
void LCodeGen::DoLoadKeyedGeneric(LLoadKeyedGeneric* instr) {
ASSERT(ToRegister(instr->object()).is(rdx));
ASSERT(ToRegister(instr->key()).is(rax));
Handle<Code> ic = isolate()->builtins()->KeyedLoadIC_Initialize();
CallCode(ic, RelocInfo::CODE_TARGET, instr);
}
void LCodeGen::DoArgumentsElements(LArgumentsElements* instr) {
Register result = ToRegister(instr->result());
// Check for arguments adapter frame.
NearLabel done, adapted;
__ movq(result, Operand(rbp, StandardFrameConstants::kCallerFPOffset));
__ Cmp(Operand(result, StandardFrameConstants::kContextOffset),
Smi::FromInt(StackFrame::ARGUMENTS_ADAPTOR));
__ j(equal, &adapted);
// No arguments adaptor frame.
__ movq(result, rbp);
__ jmp(&done);
// Arguments adaptor frame present.
__ bind(&adapted);
__ movq(result, Operand(rbp, StandardFrameConstants::kCallerFPOffset));
// Result is the frame pointer for the frame if not adapted and for the real
// frame below the adaptor frame if adapted.
__ bind(&done);
}
void LCodeGen::DoArgumentsLength(LArgumentsLength* instr) {
Register result = ToRegister(instr->result());
NearLabel done;
// If no arguments adaptor frame the number of arguments is fixed.
if (instr->InputAt(0)->IsRegister()) {
__ cmpq(rbp, ToRegister(instr->InputAt(0)));
} else {
__ cmpq(rbp, ToOperand(instr->InputAt(0)));
}
__ movq(result, Immediate(scope()->num_parameters()));
__ j(equal, &done);
// Arguments adaptor frame present. Get argument length from there.
__ movq(result, Operand(rbp, StandardFrameConstants::kCallerFPOffset));
__ movq(result, Operand(result,
ArgumentsAdaptorFrameConstants::kLengthOffset));
__ SmiToInteger32(result, result);
// Argument length is in result register.
__ bind(&done);
}
void LCodeGen::DoApplyArguments(LApplyArguments* instr) {
Register receiver = ToRegister(instr->receiver());
Register function = ToRegister(instr->function());
Register length = ToRegister(instr->length());
Register elements = ToRegister(instr->elements());
ASSERT(receiver.is(rax)); // Used for parameter count.
ASSERT(function.is(rdi)); // Required by InvokeFunction.
ASSERT(ToRegister(instr->result()).is(rax));
// If the receiver is null or undefined, we have to pass the global object
// as a receiver.
NearLabel global_object, receiver_ok;
__ CompareRoot(receiver, Heap::kNullValueRootIndex);
__ j(equal, &global_object);
__ CompareRoot(receiver, Heap::kUndefinedValueRootIndex);
__ j(equal, &global_object);
// The receiver should be a JS object.
Condition is_smi = __ CheckSmi(receiver);
DeoptimizeIf(is_smi, instr->environment());
__ CmpObjectType(receiver, FIRST_JS_OBJECT_TYPE, kScratchRegister);
DeoptimizeIf(below, instr->environment());
__ jmp(&receiver_ok);
__ bind(&global_object);
// TODO(kmillikin): We have a hydrogen value for the global object. See
// if it's better to use it than to explicitly fetch it from the context
// here.
__ movq(receiver, Operand(rbp, StandardFrameConstants::kContextOffset));
__ movq(receiver, ContextOperand(receiver, Context::GLOBAL_INDEX));
__ bind(&receiver_ok);
// Copy the arguments to this function possibly from the
// adaptor frame below it.
const uint32_t kArgumentsLimit = 1 * KB;
__ cmpq(length, Immediate(kArgumentsLimit));
DeoptimizeIf(above, instr->environment());
__ push(receiver);
__ movq(receiver, length);
// Loop through the arguments pushing them onto the execution
// stack.
NearLabel invoke, loop;
// length is a small non-negative integer, due to the test above.
__ testl(length, length);
__ j(zero, &invoke);
__ bind(&loop);
__ push(Operand(elements, length, times_pointer_size, 1 * kPointerSize));
__ decl(length);
__ j(not_zero, &loop);
// Invoke the function.
__ bind(&invoke);
ASSERT(instr->HasPointerMap() && instr->HasDeoptimizationEnvironment());
LPointerMap* pointers = instr->pointer_map();
LEnvironment* env = instr->deoptimization_environment();
RecordPosition(pointers->position());
RegisterEnvironmentForDeoptimization(env);
SafepointGenerator safepoint_generator(this,
pointers,
env->deoptimization_index());
v8::internal::ParameterCount actual(rax);
__ InvokeFunction(function, actual, CALL_FUNCTION, &safepoint_generator);
}
void LCodeGen::DoPushArgument(LPushArgument* instr) {
LOperand* argument = instr->InputAt(0);
if (argument->IsConstantOperand()) {
EmitPushConstantOperand(argument);
} else if (argument->IsRegister()) {
__ push(ToRegister(argument));
} else {
ASSERT(!argument->IsDoubleRegister());
__ push(ToOperand(argument));
}
}
void LCodeGen::DoContext(LContext* instr) {
Register result = ToRegister(instr->result());
__ movq(result, Operand(rbp, StandardFrameConstants::kContextOffset));
}
void LCodeGen::DoOuterContext(LOuterContext* instr) {
Register context = ToRegister(instr->context());
Register result = ToRegister(instr->result());
__ movq(result,
Operand(context, Context::SlotOffset(Context::CLOSURE_INDEX)));
__ movq(result, FieldOperand(result, JSFunction::kContextOffset));
}
void LCodeGen::DoGlobalObject(LGlobalObject* instr) {
Register result = ToRegister(instr->result());
__ movq(result, GlobalObjectOperand());
}
void LCodeGen::DoGlobalReceiver(LGlobalReceiver* instr) {
Register global = ToRegister(instr->global());
Register result = ToRegister(instr->result());
__ movq(result, FieldOperand(global, GlobalObject::kGlobalReceiverOffset));
}
void LCodeGen::CallKnownFunction(Handle<JSFunction> function,
int arity,
LInstruction* instr) {
// Change context if needed.
bool change_context =
(info()->closure()->context() != function->context()) ||
scope()->contains_with() ||
(scope()->num_heap_slots() > 0);
if (change_context) {
__ movq(rsi, FieldOperand(rdi, JSFunction::kContextOffset));
}
// Set rax to arguments count if adaption is not needed. Assumes that rax
// is available to write to at this point.
if (!function->NeedsArgumentsAdaption()) {
__ Set(rax, arity);
}
LPointerMap* pointers = instr->pointer_map();
RecordPosition(pointers->position());
// Invoke function.
if (*function == *info()->closure()) {
__ CallSelf();
} else {
__ call(FieldOperand(rdi, JSFunction::kCodeEntryOffset));
}
// Setup deoptimization.
RegisterLazyDeoptimization(instr);
// Restore context.
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
}
void LCodeGen::DoCallConstantFunction(LCallConstantFunction* instr) {
ASSERT(ToRegister(instr->result()).is(rax));
__ Move(rdi, instr->function());
CallKnownFunction(instr->function(), instr->arity(), instr);
}
void LCodeGen::DoDeferredMathAbsTaggedHeapNumber(LUnaryMathOperation* instr) {
Register input_reg = ToRegister(instr->InputAt(0));
__ CompareRoot(FieldOperand(input_reg, HeapObject::kMapOffset),
Heap::kHeapNumberMapRootIndex);
DeoptimizeIf(not_equal, instr->environment());
Label done;
Register tmp = input_reg.is(rax) ? rcx : rax;
Register tmp2 = tmp.is(rcx) ? rdx : input_reg.is(rcx) ? rdx : rcx;
// Preserve the value of all registers.
__ PushSafepointRegisters();
Label negative;
__ movl(tmp, FieldOperand(input_reg, HeapNumber::kExponentOffset));
// Check the sign of the argument. If the argument is positive, just
// return it. We do not need to patch the stack since |input| and
// |result| are the same register and |input| will be restored
// unchanged by popping safepoint registers.
__ testl(tmp, Immediate(HeapNumber::kSignMask));
__ j(not_zero, &negative);
__ jmp(&done);
__ bind(&negative);
Label allocated, slow;
__ AllocateHeapNumber(tmp, tmp2, &slow);
__ jmp(&allocated);
// Slow case: Call the runtime system to do the number allocation.
__ bind(&slow);
__ CallRuntimeSaveDoubles(Runtime::kAllocateHeapNumber);
RecordSafepointWithRegisters(
instr->pointer_map(), 0, Safepoint::kNoDeoptimizationIndex);
// Set the pointer to the new heap number in tmp.
if (!tmp.is(rax)) {
__ movq(tmp, rax);
}
// Restore input_reg after call to runtime.
__ LoadFromSafepointRegisterSlot(input_reg, input_reg);
__ bind(&allocated);
__ movq(tmp2, FieldOperand(input_reg, HeapNumber::kValueOffset));
__ shl(tmp2, Immediate(1));
__ shr(tmp2, Immediate(1));
__ movq(FieldOperand(tmp, HeapNumber::kValueOffset), tmp2);
__ StoreToSafepointRegisterSlot(input_reg, tmp);
__ bind(&done);
__ PopSafepointRegisters();
}
void LCodeGen::EmitIntegerMathAbs(LUnaryMathOperation* instr) {
Register input_reg = ToRegister(instr->InputAt(0));
__ testl(input_reg, input_reg);
Label is_positive;
__ j(not_sign, &is_positive);
__ negl(input_reg); // Sets flags.
DeoptimizeIf(negative, instr->environment());
__ bind(&is_positive);
}
void LCodeGen::DoMathAbs(LUnaryMathOperation* instr) {
// Class for deferred case.
class DeferredMathAbsTaggedHeapNumber: public LDeferredCode {
public:
DeferredMathAbsTaggedHeapNumber(LCodeGen* codegen,
LUnaryMathOperation* instr)
: LDeferredCode(codegen), instr_(instr) { }
virtual void Generate() {
codegen()->DoDeferredMathAbsTaggedHeapNumber(instr_);
}
private:
LUnaryMathOperation* instr_;
};
ASSERT(instr->InputAt(0)->Equals(instr->result()));
Representation r = instr->hydrogen()->value()->representation();
if (r.IsDouble()) {
XMMRegister scratch = xmm0;
XMMRegister input_reg = ToDoubleRegister(instr->InputAt(0));
__ xorpd(scratch, scratch);
__ subsd(scratch, input_reg);
__ andpd(input_reg, scratch);
} else if (r.IsInteger32()) {
EmitIntegerMathAbs(instr);
} else { // Tagged case.
DeferredMathAbsTaggedHeapNumber* deferred =
new DeferredMathAbsTaggedHeapNumber(this, instr);
Register input_reg = ToRegister(instr->InputAt(0));
// Smi check.
__ JumpIfNotSmi(input_reg, deferred->entry());
EmitIntegerMathAbs(instr);
__ bind(deferred->exit());
}
}
void LCodeGen::DoMathFloor(LUnaryMathOperation* instr) {
XMMRegister xmm_scratch = xmm0;
Register output_reg = ToRegister(instr->result());
XMMRegister input_reg = ToDoubleRegister(instr->InputAt(0));
__ xorpd(xmm_scratch, xmm_scratch); // Zero the register.
__ ucomisd(input_reg, xmm_scratch);
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
DeoptimizeIf(below_equal, instr->environment());
} else {
DeoptimizeIf(below, instr->environment());
}
// Use truncating instruction (OK because input is positive).
__ cvttsd2si(output_reg, input_reg);
// Overflow is signalled with minint.
__ cmpl(output_reg, Immediate(0x80000000));
DeoptimizeIf(equal, instr->environment());
}
void LCodeGen::DoMathRound(LUnaryMathOperation* instr) {
const XMMRegister xmm_scratch = xmm0;
Register output_reg = ToRegister(instr->result());
XMMRegister input_reg = ToDoubleRegister(instr->InputAt(0));
// xmm_scratch = 0.5
__ movq(kScratchRegister, V8_INT64_C(0x3FE0000000000000), RelocInfo::NONE);
__ movq(xmm_scratch, kScratchRegister);
// input = input + 0.5
__ addsd(input_reg, xmm_scratch);
// We need to return -0 for the input range [-0.5, 0[, otherwise
// compute Math.floor(value + 0.5).
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
__ ucomisd(input_reg, xmm_scratch);
DeoptimizeIf(below_equal, instr->environment());
} else {
// If we don't need to bailout on -0, we check only bailout
// on negative inputs.
__ xorpd(xmm_scratch, xmm_scratch); // Zero the register.
__ ucomisd(input_reg, xmm_scratch);
DeoptimizeIf(below, instr->environment());
}
// Compute Math.floor(value + 0.5).
// Use truncating instruction (OK because input is positive).
__ cvttsd2si(output_reg, input_reg);
// Overflow is signalled with minint.
__ cmpl(output_reg, Immediate(0x80000000));
DeoptimizeIf(equal, instr->environment());
}
void LCodeGen::DoMathSqrt(LUnaryMathOperation* instr) {
XMMRegister input_reg = ToDoubleRegister(instr->InputAt(0));
ASSERT(ToDoubleRegister(instr->result()).is(input_reg));
__ sqrtsd(input_reg, input_reg);
}
void LCodeGen::DoMathPowHalf(LUnaryMathOperation* instr) {
XMMRegister xmm_scratch = xmm0;
XMMRegister input_reg = ToDoubleRegister(instr->InputAt(0));
ASSERT(ToDoubleRegister(instr->result()).is(input_reg));
__ xorpd(xmm_scratch, xmm_scratch);
__ addsd(input_reg, xmm_scratch); // Convert -0 to +0.
__ sqrtsd(input_reg, input_reg);
}
void LCodeGen::DoPower(LPower* instr) {
LOperand* left = instr->InputAt(0);
XMMRegister left_reg = ToDoubleRegister(left);
ASSERT(!left_reg.is(xmm1));
LOperand* right = instr->InputAt(1);
XMMRegister result_reg = ToDoubleRegister(instr->result());
Representation exponent_type = instr->hydrogen()->right()->representation();
if (exponent_type.IsDouble()) {
__ PrepareCallCFunction(2);
// Move arguments to correct registers
__ movsd(xmm0, left_reg);
ASSERT(ToDoubleRegister(right).is(xmm1));
__ CallCFunction(
ExternalReference::power_double_double_function(isolate()), 2);
} else if (exponent_type.IsInteger32()) {
__ PrepareCallCFunction(2);
// Move arguments to correct registers: xmm0 and edi (not rdi).
// On Windows, the registers are xmm0 and edx.
__ movsd(xmm0, left_reg);
#ifdef _WIN64
ASSERT(ToRegister(right).is(rdx));
#else
ASSERT(ToRegister(right).is(rdi));
#endif
__ CallCFunction(
ExternalReference::power_double_int_function(isolate()), 2);
} else {
ASSERT(exponent_type.IsTagged());
CpuFeatures::Scope scope(SSE2);
Register right_reg = ToRegister(right);
Label non_smi, call;
__ JumpIfNotSmi(right_reg, &non_smi);
__ SmiToInteger32(right_reg, right_reg);
__ cvtlsi2sd(xmm1, right_reg);
__ jmp(&call);
__ bind(&non_smi);
__ CmpObjectType(right_reg, HEAP_NUMBER_TYPE , kScratchRegister);
DeoptimizeIf(not_equal, instr->environment());
__ movsd(xmm1, FieldOperand(right_reg, HeapNumber::kValueOffset));
__ bind(&call);
__ PrepareCallCFunction(2);
// Move arguments to correct registers xmm0 and xmm1.
__ movsd(xmm0, left_reg);
// Right argument is already in xmm1.
__ CallCFunction(
ExternalReference::power_double_double_function(isolate()), 2);
}
// Return value is in xmm0.
__ movsd(result_reg, xmm0);
// Restore context register.
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
}
void LCodeGen::DoMathLog(LUnaryMathOperation* instr) {
ASSERT(ToDoubleRegister(instr->result()).is(xmm1));
TranscendentalCacheStub stub(TranscendentalCache::LOG,
TranscendentalCacheStub::UNTAGGED);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
}
void LCodeGen::DoMathCos(LUnaryMathOperation* instr) {
ASSERT(ToDoubleRegister(instr->result()).is(xmm1));
TranscendentalCacheStub stub(TranscendentalCache::COS,
TranscendentalCacheStub::UNTAGGED);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
}
void LCodeGen::DoMathSin(LUnaryMathOperation* instr) {
ASSERT(ToDoubleRegister(instr->result()).is(xmm1));
TranscendentalCacheStub stub(TranscendentalCache::SIN,
TranscendentalCacheStub::UNTAGGED);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
}
void LCodeGen::DoUnaryMathOperation(LUnaryMathOperation* instr) {
switch (instr->op()) {
case kMathAbs:
DoMathAbs(instr);
break;
case kMathFloor:
DoMathFloor(instr);
break;
case kMathRound:
DoMathRound(instr);
break;
case kMathSqrt:
DoMathSqrt(instr);
break;
case kMathPowHalf:
DoMathPowHalf(instr);
break;
case kMathCos:
DoMathCos(instr);
break;
case kMathSin:
DoMathSin(instr);
break;
case kMathLog:
DoMathLog(instr);
break;
default:
UNREACHABLE();
}
}
void LCodeGen::DoCallKeyed(LCallKeyed* instr) {
ASSERT(ToRegister(instr->key()).is(rcx));
ASSERT(ToRegister(instr->result()).is(rax));
int arity = instr->arity();
Handle<Code> ic = isolate()->stub_cache()->ComputeKeyedCallInitialize(
arity, NOT_IN_LOOP);
CallCode(ic, RelocInfo::CODE_TARGET, instr);
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
}
void LCodeGen::DoCallNamed(LCallNamed* instr) {
ASSERT(ToRegister(instr->result()).is(rax));
int arity = instr->arity();
Handle<Code> ic = isolate()->stub_cache()->ComputeCallInitialize(
arity, NOT_IN_LOOP);
__ Move(rcx, instr->name());
CallCode(ic, RelocInfo::CODE_TARGET, instr);
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
}
void LCodeGen::DoCallFunction(LCallFunction* instr) {
ASSERT(ToRegister(instr->result()).is(rax));
int arity = instr->arity();
CallFunctionStub stub(arity, NOT_IN_LOOP, RECEIVER_MIGHT_BE_VALUE);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
__ Drop(1);
}
void LCodeGen::DoCallGlobal(LCallGlobal* instr) {
ASSERT(ToRegister(instr->result()).is(rax));
int arity = instr->arity();
Handle<Code> ic = isolate()->stub_cache()->ComputeCallInitialize(
arity, NOT_IN_LOOP);
__ Move(rcx, instr->name());
CallCode(ic, RelocInfo::CODE_TARGET_CONTEXT, instr);
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
}
void LCodeGen::DoCallKnownGlobal(LCallKnownGlobal* instr) {
ASSERT(ToRegister(instr->result()).is(rax));
__ Move(rdi, instr->target());
CallKnownFunction(instr->target(), instr->arity(), instr);
}
void LCodeGen::DoCallNew(LCallNew* instr) {
ASSERT(ToRegister(instr->InputAt(0)).is(rdi));
ASSERT(ToRegister(instr->result()).is(rax));
Handle<Code> builtin = isolate()->builtins()->JSConstructCall();
__ Set(rax, instr->arity());
CallCode(builtin, RelocInfo::CONSTRUCT_CALL, instr);
}
void LCodeGen::DoCallRuntime(LCallRuntime* instr) {
CallRuntime(instr->function(), instr->arity(), instr);
}
void LCodeGen::DoStoreNamedField(LStoreNamedField* instr) {
Register object = ToRegister(instr->object());
Register value = ToRegister(instr->value());
int offset = instr->offset();
if (!instr->transition().is_null()) {
__ Move(FieldOperand(object, HeapObject::kMapOffset), instr->transition());
}
// Do the store.
if (instr->is_in_object()) {
__ movq(FieldOperand(object, offset), value);
if (instr->needs_write_barrier()) {
Register temp = ToRegister(instr->TempAt(0));
// Update the write barrier for the object for in-object properties.
__ RecordWrite(object, offset, value, temp);
}
} else {
Register temp = ToRegister(instr->TempAt(0));
__ movq(temp, FieldOperand(object, JSObject::kPropertiesOffset));
__ movq(FieldOperand(temp, offset), value);
if (instr->needs_write_barrier()) {
// Update the write barrier for the properties array.
// object is used as a scratch register.
__ RecordWrite(temp, offset, value, object);
}
}
}
void LCodeGen::DoStoreNamedGeneric(LStoreNamedGeneric* instr) {
ASSERT(ToRegister(instr->object()).is(rdx));
ASSERT(ToRegister(instr->value()).is(rax));
__ Move(rcx, instr->hydrogen()->name());
Handle<Code> ic = info_->is_strict()
? isolate()->builtins()->StoreIC_Initialize_Strict()
: isolate()->builtins()->StoreIC_Initialize();
CallCode(ic, RelocInfo::CODE_TARGET, instr);
}
void LCodeGen::DoStoreKeyedSpecializedArrayElement(
LStoreKeyedSpecializedArrayElement* instr) {
Register external_pointer = ToRegister(instr->external_pointer());
Register key = ToRegister(instr->key());
ExternalArrayType array_type = instr->array_type();
if (array_type == kExternalFloatArray) {
XMMRegister value(ToDoubleRegister(instr->value()));
__ cvtsd2ss(value, value);
__ movss(Operand(external_pointer, key, times_4, 0), value);
} else {
Register value(ToRegister(instr->value()));
switch (array_type) {
case kExternalPixelArray:
{ // Clamp the value to [0..255].
NearLabel done;
__ testl(value, Immediate(0xFFFFFF00));
__ j(zero, &done);
__ setcc(negative, value); // 1 if negative, 0 if positive.
__ decb(value); // 0 if negative, 255 if positive.
__ bind(&done);
__ movb(Operand(external_pointer, key, times_1, 0), value);
}
break;
case kExternalByteArray:
case kExternalUnsignedByteArray:
__ movb(Operand(external_pointer, key, times_1, 0), value);
break;
case kExternalShortArray:
case kExternalUnsignedShortArray:
__ movw(Operand(external_pointer, key, times_2, 0), value);
break;
case kExternalIntArray:
case kExternalUnsignedIntArray:
__ movl(Operand(external_pointer, key, times_4, 0), value);
break;
case kExternalFloatArray:
UNREACHABLE();
break;
}
}
}
void LCodeGen::DoBoundsCheck(LBoundsCheck* instr) {
if (instr->length()->IsRegister()) {
__ cmpq(ToRegister(instr->index()), ToRegister(instr->length()));
} else {
__ cmpq(ToRegister(instr->index()), ToOperand(instr->length()));
}
DeoptimizeIf(above_equal, instr->environment());
}
void LCodeGen::DoStoreKeyedFastElement(LStoreKeyedFastElement* instr) {
Register value = ToRegister(instr->value());
Register elements = ToRegister(instr->object());
Register key = instr->key()->IsRegister() ? ToRegister(instr->key()) : no_reg;
// Do the store.
if (instr->key()->IsConstantOperand()) {
ASSERT(!instr->hydrogen()->NeedsWriteBarrier());
LConstantOperand* const_operand = LConstantOperand::cast(instr->key());
int offset =
ToInteger32(const_operand) * kPointerSize + FixedArray::kHeaderSize;
__ movq(FieldOperand(elements, offset), value);
} else {
__ movq(FieldOperand(elements,
key,
times_pointer_size,
FixedArray::kHeaderSize),
value);
}
if (instr->hydrogen()->NeedsWriteBarrier()) {
// Compute address of modified element and store it into key register.
__ lea(key, FieldOperand(elements,
key,
times_pointer_size,
FixedArray::kHeaderSize));
__ RecordWrite(elements, key, value);
}
}
void LCodeGen::DoStoreKeyedGeneric(LStoreKeyedGeneric* instr) {
ASSERT(ToRegister(instr->object()).is(rdx));
ASSERT(ToRegister(instr->key()).is(rcx));
ASSERT(ToRegister(instr->value()).is(rax));
Handle<Code> ic = info_->is_strict()
? isolate()->builtins()->KeyedStoreIC_Initialize_Strict()
: isolate()->builtins()->KeyedStoreIC_Initialize();
CallCode(ic, RelocInfo::CODE_TARGET, instr);
}
void LCodeGen::DoStringCharCodeAt(LStringCharCodeAt* instr) {
class DeferredStringCharCodeAt: public LDeferredCode {
public:
DeferredStringCharCodeAt(LCodeGen* codegen, LStringCharCodeAt* instr)
: LDeferredCode(codegen), instr_(instr) { }
virtual void Generate() { codegen()->DoDeferredStringCharCodeAt(instr_); }
private:
LStringCharCodeAt* instr_;
};
Register string = ToRegister(instr->string());
Register index = no_reg;
int const_index = -1;
if (instr->index()->IsConstantOperand()) {
const_index = ToInteger32(LConstantOperand::cast(instr->index()));
STATIC_ASSERT(String::kMaxLength <= Smi::kMaxValue);
if (!Smi::IsValid(const_index)) {
// Guaranteed to be out of bounds because of the assert above.
// So the bounds check that must dominate this instruction must
// have deoptimized already.
if (FLAG_debug_code) {
__ Abort("StringCharCodeAt: out of bounds index.");
}
// No code needs to be generated.
return;
}
} else {
index = ToRegister(instr->index());
}
Register result = ToRegister(instr->result());
DeferredStringCharCodeAt* deferred =
new DeferredStringCharCodeAt(this, instr);
NearLabel flat_string, ascii_string, done;
// Fetch the instance type of the receiver into result register.
__ movq(result, FieldOperand(string, HeapObject::kMapOffset));
__ movzxbl(result, FieldOperand(result, Map::kInstanceTypeOffset));
// We need special handling for non-sequential strings.
STATIC_ASSERT(kSeqStringTag == 0);
__ testb(result, Immediate(kStringRepresentationMask));
__ j(zero, &flat_string);
// Handle cons strings and go to deferred code for the rest.
__ testb(result, Immediate(kIsConsStringMask));
__ j(zero, deferred->entry());
// ConsString.
// Check whether the right hand side is the empty string (i.e. if
// this is really a flat string in a cons string). If that is not
// the case we would rather go to the runtime system now to flatten
// the string.
__ CompareRoot(FieldOperand(string, ConsString::kSecondOffset),
Heap::kEmptyStringRootIndex);
__ j(not_equal, deferred->entry());
// Get the first of the two strings and load its instance type.
__ movq(string, FieldOperand(string, ConsString::kFirstOffset));
__ movq(result, FieldOperand(string, HeapObject::kMapOffset));
__ movzxbl(result, FieldOperand(result, Map::kInstanceTypeOffset));
// If the first cons component is also non-flat, then go to runtime.
STATIC_ASSERT(kSeqStringTag == 0);
__ testb(result, Immediate(kStringRepresentationMask));
__ j(not_zero, deferred->entry());
// Check for ASCII or two-byte string.
__ bind(&flat_string);
STATIC_ASSERT(kAsciiStringTag != 0);
__ testb(result, Immediate(kStringEncodingMask));
__ j(not_zero, &ascii_string);
// Two-byte string.
// Load the two-byte character code into the result register.
STATIC_ASSERT(kSmiTag == 0 && kSmiTagSize == 1);
if (instr->index()->IsConstantOperand()) {
__ movzxwl(result,
FieldOperand(string,
SeqTwoByteString::kHeaderSize +
(kUC16Size * const_index)));
} else {
__ movzxwl(result, FieldOperand(string,
index,
times_2,
SeqTwoByteString::kHeaderSize));
}
__ jmp(&done);
// ASCII string.
// Load the byte into the result register.
__ bind(&ascii_string);
if (instr->index()->IsConstantOperand()) {
__ movzxbl(result, FieldOperand(string,
SeqAsciiString::kHeaderSize + const_index));
} else {
__ movzxbl(result, FieldOperand(string,
index,
times_1,
SeqAsciiString::kHeaderSize));
}
__ bind(&done);
__ bind(deferred->exit());
}
void LCodeGen::DoDeferredStringCharCodeAt(LStringCharCodeAt* instr) {
Register string = ToRegister(instr->string());
Register result = ToRegister(instr->result());
// TODO(3095996): Get rid of this. For now, we need to make the
// result register contain a valid pointer because it is already
// contained in the register pointer map.
__ Set(result, 0);
__ PushSafepointRegisters();
__ push(string);
// Push the index as a smi. This is safe because of the checks in
// DoStringCharCodeAt above.
STATIC_ASSERT(String::kMaxLength <= Smi::kMaxValue);
if (instr->index()->IsConstantOperand()) {
int const_index = ToInteger32(LConstantOperand::cast(instr->index()));
__ Push(Smi::FromInt(const_index));
} else {
Register index = ToRegister(instr->index());
__ Integer32ToSmi(index, index);
__ push(index);
}
__ movq(rsi, Operand(rbp, StandardFrameConstants::kContextOffset));
__ CallRuntimeSaveDoubles(Runtime::kStringCharCodeAt);
RecordSafepointWithRegisters(
instr->pointer_map(), 2, Safepoint::kNoDeoptimizationIndex);
if (FLAG_debug_code) {
__ AbortIfNotSmi(rax);
}
__ SmiToInteger32(rax, rax);
__ StoreToSafepointRegisterSlot(result, rax);
__ PopSafepointRegisters();
}
void LCodeGen::DoStringCharFromCode(LStringCharFromCode* instr) {
class DeferredStringCharFromCode: public LDeferredCode {
public:
DeferredStringCharFromCode(LCodeGen* codegen, LStringCharFromCode* instr)
: LDeferredCode(codegen), instr_(instr) { }
virtual void Generate() { codegen()->DoDeferredStringCharFromCode(instr_); }
private:
LStringCharFromCode* instr_;
};
DeferredStringCharFromCode* deferred =
new DeferredStringCharFromCode(this, instr);
ASSERT(instr->hydrogen()->value()->representation().IsInteger32());
Register char_code = ToRegister(instr->char_code());
Register result = ToRegister(instr->result());
ASSERT(!char_code.is(result));
__ cmpl(char_code, Immediate(String::kMaxAsciiCharCode));
__ j(above, deferred->entry());
__ LoadRoot(result, Heap::kSingleCharacterStringCacheRootIndex);
__ movq(result, FieldOperand(result,
char_code, times_pointer_size,
FixedArray::kHeaderSize));
__ CompareRoot(result, Heap::kUndefinedValueRootIndex);
__ j(equal, deferred->entry());
__ bind(deferred->exit());
}
void LCodeGen::DoDeferredStringCharFromCode(LStringCharFromCode* instr) {
Register char_code = ToRegister(instr->char_code());
Register result = ToRegister(instr->result());
// TODO(3095996): Get rid of this. For now, we need to make the
// result register contain a valid pointer because it is already
// contained in the register pointer map.
__ Set(result, 0);
__ PushSafepointRegisters();
__ Integer32ToSmi(char_code, char_code);
__ push(char_code);
__ CallRuntimeSaveDoubles(Runtime::kCharFromCode);
RecordSafepointWithRegisters(
instr->pointer_map(), 1, Safepoint::kNoDeoptimizationIndex);
__ StoreToSafepointRegisterSlot(result, rax);
__ PopSafepointRegisters();
}
void LCodeGen::DoStringLength(LStringLength* instr) {
Register string = ToRegister(instr->string());
Register result = ToRegister(instr->result());
__ movq(result, FieldOperand(string, String::kLengthOffset));
}
void LCodeGen::DoInteger32ToDouble(LInteger32ToDouble* instr) {
LOperand* input = instr->InputAt(0);
ASSERT(input->IsRegister() || input->IsStackSlot());
LOperand* output = instr->result();
ASSERT(output->IsDoubleRegister());
if (input->IsRegister()) {
__ cvtlsi2sd(ToDoubleRegister(output), ToRegister(input));
} else {
__ cvtlsi2sd(ToDoubleRegister(output), ToOperand(input));
}
}
void LCodeGen::DoNumberTagI(LNumberTagI* instr) {
LOperand* input = instr->InputAt(0);
ASSERT(input->IsRegister() && input->Equals(instr->result()));
Register reg = ToRegister(input);
__ Integer32ToSmi(reg, reg);
}
void LCodeGen::DoNumberTagD(LNumberTagD* instr) {
class DeferredNumberTagD: public LDeferredCode {
public:
DeferredNumberTagD(LCodeGen* codegen, LNumberTagD* instr)
: LDeferredCode(codegen), instr_(instr) { }
virtual void Generate() { codegen()->DoDeferredNumberTagD(instr_); }
private:
LNumberTagD* instr_;
};
XMMRegister input_reg = ToDoubleRegister(instr->InputAt(0));
Register reg = ToRegister(instr->result());
Register tmp = ToRegister(instr->TempAt(0));
DeferredNumberTagD* deferred = new DeferredNumberTagD(this, instr);
if (FLAG_inline_new) {
__ AllocateHeapNumber(reg, tmp, deferred->entry());
} else {
__ jmp(deferred->entry());
}
__ bind(deferred->exit());
__ movsd(FieldOperand(reg, HeapNumber::kValueOffset), input_reg);
}
void LCodeGen::DoDeferredNumberTagD(LNumberTagD* instr) {
// TODO(3095996): Get rid of this. For now, we need to make the
// result register contain a valid pointer because it is already
// contained in the register pointer map.
Register reg = ToRegister(instr->result());
__ Move(reg, Smi::FromInt(0));
__ PushSafepointRegisters();
__ CallRuntimeSaveDoubles(Runtime::kAllocateHeapNumber);
RecordSafepointWithRegisters(
instr->pointer_map(), 0, Safepoint::kNoDeoptimizationIndex);
// Ensure that value in rax survives popping registers.
__ movq(kScratchRegister, rax);
__ PopSafepointRegisters();
__ movq(reg, kScratchRegister);
}
void LCodeGen::DoSmiTag(LSmiTag* instr) {
ASSERT(instr->InputAt(0)->Equals(instr->result()));
Register input = ToRegister(instr->InputAt(0));
ASSERT(!instr->hydrogen_value()->CheckFlag(HValue::kCanOverflow));
__ Integer32ToSmi(input, input);
}
void LCodeGen::DoSmiUntag(LSmiUntag* instr) {
ASSERT(instr->InputAt(0)->Equals(instr->result()));
Register input = ToRegister(instr->InputAt(0));
if (instr->needs_check()) {
Condition is_smi = __ CheckSmi(input);
DeoptimizeIf(NegateCondition(is_smi), instr->environment());
}
__ SmiToInteger32(input, input);
}
void LCodeGen::EmitNumberUntagD(Register input_reg,
XMMRegister result_reg,
LEnvironment* env) {
NearLabel load_smi, heap_number, done;
// Smi check.
__ JumpIfSmi(input_reg, &load_smi);
// Heap number map check.
__ CompareRoot(FieldOperand(input_reg, HeapObject::kMapOffset),
Heap::kHeapNumberMapRootIndex);
__ j(equal, &heap_number);
__ CompareRoot(input_reg, Heap::kUndefinedValueRootIndex);
DeoptimizeIf(not_equal, env);
// Convert undefined to NaN. Compute NaN as 0/0.
__ xorpd(result_reg, result_reg);
__ divsd(result_reg, result_reg);
__ jmp(&done);
// Heap number to XMM conversion.
__ bind(&heap_number);
__ movsd(result_reg, FieldOperand(input_reg, HeapNumber::kValueOffset));
__ jmp(&done);
// Smi to XMM conversion
__ bind(&load_smi);
__ SmiToInteger32(kScratchRegister, input_reg);
__ cvtlsi2sd(result_reg, kScratchRegister);
__ bind(&done);
}
class DeferredTaggedToI: public LDeferredCode {
public:
DeferredTaggedToI(LCodeGen* codegen, LTaggedToI* instr)
: LDeferredCode(codegen), instr_(instr) { }
virtual void Generate() { codegen()->DoDeferredTaggedToI(instr_); }
private:
LTaggedToI* instr_;
};
void LCodeGen::DoDeferredTaggedToI(LTaggedToI* instr) {
NearLabel done, heap_number;
Register input_reg = ToRegister(instr->InputAt(0));
// Heap number map check.
__ CompareRoot(FieldOperand(input_reg, HeapObject::kMapOffset),
Heap::kHeapNumberMapRootIndex);
if (instr->truncating()) {
__ j(equal, &heap_number);
// Check for undefined. Undefined is converted to zero for truncating
// conversions.
__ CompareRoot(input_reg, Heap::kUndefinedValueRootIndex);
DeoptimizeIf(not_equal, instr->environment());
__ movl(input_reg, Immediate(0));
__ jmp(&done);
__ bind(&heap_number);
__ movsd(xmm0, FieldOperand(input_reg, HeapNumber::kValueOffset));
__ cvttsd2siq(input_reg, xmm0);
__ Set(kScratchRegister, V8_UINT64_C(0x8000000000000000));
__ cmpl(input_reg, kScratchRegister);
DeoptimizeIf(equal, instr->environment());
} else {
// Deoptimize if we don't have a heap number.
DeoptimizeIf(not_equal, instr->environment());
XMMRegister xmm_temp = ToDoubleRegister(instr->TempAt(0));
__ movsd(xmm0, FieldOperand(input_reg, HeapNumber::kValueOffset));
__ cvttsd2si(input_reg, xmm0);
__ cvtlsi2sd(xmm_temp, input_reg);
__ ucomisd(xmm0, xmm_temp);
DeoptimizeIf(not_equal, instr->environment());
DeoptimizeIf(parity_even, instr->environment()); // NaN.
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
__ testl(input_reg, input_reg);
__ j(not_zero, &done);
__ movmskpd(input_reg, xmm0);
__ andl(input_reg, Immediate(1));
DeoptimizeIf(not_zero, instr->environment());
}
}
__ bind(&done);
}
void LCodeGen::DoTaggedToI(LTaggedToI* instr) {
LOperand* input = instr->InputAt(0);
ASSERT(input->IsRegister());
ASSERT(input->Equals(instr->result()));
Register input_reg = ToRegister(input);
DeferredTaggedToI* deferred = new DeferredTaggedToI(this, instr);
__ JumpIfNotSmi(input_reg, deferred->entry());
__ SmiToInteger32(input_reg, input_reg);
__ bind(deferred->exit());
}
void LCodeGen::DoNumberUntagD(LNumberUntagD* instr) {
LOperand* input = instr->InputAt(0);
ASSERT(input->IsRegister());
LOperand* result = instr->result();
ASSERT(result->IsDoubleRegister());
Register input_reg = ToRegister(input);
XMMRegister result_reg = ToDoubleRegister(result);
EmitNumberUntagD(input_reg, result_reg, instr->environment());
}
void LCodeGen::DoDoubleToI(LDoubleToI* instr) {
LOperand* input = instr->InputAt(0);
ASSERT(input->IsDoubleRegister());
LOperand* result = instr->result();
ASSERT(result->IsRegister());
XMMRegister input_reg = ToDoubleRegister(input);
Register result_reg = ToRegister(result);
if (instr->truncating()) {
// Performs a truncating conversion of a floating point number as used by
// the JS bitwise operations.
__ cvttsd2siq(result_reg, input_reg);
__ movq(kScratchRegister, V8_INT64_C(0x8000000000000000), RelocInfo::NONE);
__ cmpl(result_reg, kScratchRegister);
DeoptimizeIf(equal, instr->environment());
} else {
__ cvttsd2si(result_reg, input_reg);
__ cvtlsi2sd(xmm0, result_reg);
__ ucomisd(xmm0, input_reg);
DeoptimizeIf(not_equal, instr->environment());
DeoptimizeIf(parity_even, instr->environment()); // NaN.
if (instr->hydrogen()->CheckFlag(HValue::kBailoutOnMinusZero)) {
NearLabel done;
// The integer converted back is equal to the original. We
// only have to test if we got -0 as an input.
__ testl(result_reg, result_reg);
__ j(not_zero, &done);
__ movmskpd(result_reg, input_reg);
// Bit 0 contains the sign of the double in input_reg.
// If input was positive, we are ok and return 0, otherwise
// deoptimize.
__ andl(result_reg, Immediate(1));
DeoptimizeIf(not_zero, instr->environment());
__ bind(&done);
}
}
}
void LCodeGen::DoCheckSmi(LCheckSmi* instr) {
LOperand* input = instr->InputAt(0);
Condition cc = masm()->CheckSmi(ToRegister(input));
DeoptimizeIf(NegateCondition(cc), instr->environment());
}
void LCodeGen::DoCheckNonSmi(LCheckNonSmi* instr) {
LOperand* input = instr->InputAt(0);
Condition cc = masm()->CheckSmi(ToRegister(input));
DeoptimizeIf(cc, instr->environment());
}
void LCodeGen::DoCheckInstanceType(LCheckInstanceType* instr) {
Register input = ToRegister(instr->InputAt(0));
InstanceType first = instr->hydrogen()->first();
InstanceType last = instr->hydrogen()->last();
__ movq(kScratchRegister, FieldOperand(input, HeapObject::kMapOffset));
// If there is only one type in the interval check for equality.
if (first == last) {
__ cmpb(FieldOperand(kScratchRegister, Map::kInstanceTypeOffset),
Immediate(static_cast<int8_t>(first)));
DeoptimizeIf(not_equal, instr->environment());
} else if (first == FIRST_STRING_TYPE && last == LAST_STRING_TYPE) {
// String has a dedicated bit in instance type.
__ testb(FieldOperand(kScratchRegister, Map::kInstanceTypeOffset),
Immediate(kIsNotStringMask));
DeoptimizeIf(not_zero, instr->environment());
} else {
__ cmpb(FieldOperand(kScratchRegister, Map::kInstanceTypeOffset),
Immediate(static_cast<int8_t>(first)));
DeoptimizeIf(below, instr->environment());
// Omit check for the last type.
if (last != LAST_TYPE) {
__ cmpb(FieldOperand(kScratchRegister, Map::kInstanceTypeOffset),
Immediate(static_cast<int8_t>(last)));
DeoptimizeIf(above, instr->environment());
}
}
}
void LCodeGen::DoCheckFunction(LCheckFunction* instr) {
ASSERT(instr->InputAt(0)->IsRegister());
Register reg = ToRegister(instr->InputAt(0));
__ Cmp(reg, instr->hydrogen()->target());
DeoptimizeIf(not_equal, instr->environment());
}
void LCodeGen::DoCheckMap(LCheckMap* instr) {
LOperand* input = instr->InputAt(0);
ASSERT(input->IsRegister());
Register reg = ToRegister(input);
__ Cmp(FieldOperand(reg, HeapObject::kMapOffset),
instr->hydrogen()->map());
DeoptimizeIf(not_equal, instr->environment());
}
void LCodeGen::LoadHeapObject(Register result, Handle<HeapObject> object) {
if (heap()->InNewSpace(*object)) {
Handle<JSGlobalPropertyCell> cell =
factory()->NewJSGlobalPropertyCell(object);
__ movq(result, cell, RelocInfo::GLOBAL_PROPERTY_CELL);
__ movq(result, Operand(result, 0));
} else {
__ Move(result, object);
}
}
void LCodeGen::DoCheckPrototypeMaps(LCheckPrototypeMaps* instr) {
Register reg = ToRegister(instr->TempAt(0));
Handle<JSObject> holder = instr->holder();
Handle<JSObject> current_prototype = instr->prototype();
// Load prototype object.
LoadHeapObject(reg, current_prototype);
// Check prototype maps up to the holder.
while (!current_prototype.is_identical_to(holder)) {
__ Cmp(FieldOperand(reg, HeapObject::kMapOffset),
Handle<Map>(current_prototype->map()));
DeoptimizeIf(not_equal, instr->environment());
current_prototype =
Handle<JSObject>(JSObject::cast(current_prototype->GetPrototype()));
// Load next prototype object.
LoadHeapObject(reg, current_prototype);
}
// Check the holder map.
__ Cmp(FieldOperand(reg, HeapObject::kMapOffset),
Handle<Map>(current_prototype->map()));
DeoptimizeIf(not_equal, instr->environment());
}
void LCodeGen::DoArrayLiteral(LArrayLiteral* instr) {
// Setup the parameters to the stub/runtime call.
__ movq(rax, Operand(rbp, JavaScriptFrameConstants::kFunctionOffset));
__ push(FieldOperand(rax, JSFunction::kLiteralsOffset));
__ Push(Smi::FromInt(instr->hydrogen()->literal_index()));
__ Push(instr->hydrogen()->constant_elements());
// Pick the right runtime function or stub to call.
int length = instr->hydrogen()->length();
if (instr->hydrogen()->IsCopyOnWrite()) {
ASSERT(instr->hydrogen()->depth() == 1);
FastCloneShallowArrayStub::Mode mode =
FastCloneShallowArrayStub::COPY_ON_WRITE_ELEMENTS;
FastCloneShallowArrayStub stub(mode, length);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
} else if (instr->hydrogen()->depth() > 1) {
CallRuntime(Runtime::kCreateArrayLiteral, 3, instr);
} else if (length > FastCloneShallowArrayStub::kMaximumClonedLength) {
CallRuntime(Runtime::kCreateArrayLiteralShallow, 3, instr);
} else {
FastCloneShallowArrayStub::Mode mode =
FastCloneShallowArrayStub::CLONE_ELEMENTS;
FastCloneShallowArrayStub stub(mode, length);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
}
}
void LCodeGen::DoObjectLiteral(LObjectLiteral* instr) {
// Setup the parameters to the stub/runtime call.
__ movq(rax, Operand(rbp, JavaScriptFrameConstants::kFunctionOffset));
__ push(FieldOperand(rax, JSFunction::kLiteralsOffset));
__ Push(Smi::FromInt(instr->hydrogen()->literal_index()));
__ Push(instr->hydrogen()->constant_properties());
__ Push(Smi::FromInt(instr->hydrogen()->fast_elements() ? 1 : 0));
// Pick the right runtime function to call.
if (instr->hydrogen()->depth() > 1) {
CallRuntime(Runtime::kCreateObjectLiteral, 4, instr);
} else {
CallRuntime(Runtime::kCreateObjectLiteralShallow, 4, instr);
}
}
void LCodeGen::DoToFastProperties(LToFastProperties* instr) {
ASSERT(ToRegister(instr->InputAt(0)).is(rax));
__ push(rax);
CallRuntime(Runtime::kToFastProperties, 1, instr);
}
void LCodeGen::DoRegExpLiteral(LRegExpLiteral* instr) {
NearLabel materialized;
// Registers will be used as follows:
// rdi = JS function.
// rcx = literals array.
// rbx = regexp literal.
// rax = regexp literal clone.
__ movq(rdi, Operand(rbp, JavaScriptFrameConstants::kFunctionOffset));
__ movq(rcx, FieldOperand(rdi, JSFunction::kLiteralsOffset));
int literal_offset = FixedArray::kHeaderSize +
instr->hydrogen()->literal_index() * kPointerSize;
__ movq(rbx, FieldOperand(rcx, literal_offset));
__ CompareRoot(rbx, Heap::kUndefinedValueRootIndex);
__ j(not_equal, &materialized);
// Create regexp literal using runtime function
// Result will be in rax.
__ push(rcx);
__ Push(Smi::FromInt(instr->hydrogen()->literal_index()));
__ Push(instr->hydrogen()->pattern());
__ Push(instr->hydrogen()->flags());
CallRuntime(Runtime::kMaterializeRegExpLiteral, 4, instr);
__ movq(rbx, rax);
__ bind(&materialized);
int size = JSRegExp::kSize + JSRegExp::kInObjectFieldCount * kPointerSize;
Label allocated, runtime_allocate;
__ AllocateInNewSpace(size, rax, rcx, rdx, &runtime_allocate, TAG_OBJECT);
__ jmp(&allocated);
__ bind(&runtime_allocate);
__ push(rbx);
__ Push(Smi::FromInt(size));
CallRuntime(Runtime::kAllocateInNewSpace, 1, instr);
__ pop(rbx);
__ bind(&allocated);
// Copy the content into the newly allocated memory.
// (Unroll copy loop once for better throughput).
for (int i = 0; i < size - kPointerSize; i += 2 * kPointerSize) {
__ movq(rdx, FieldOperand(rbx, i));
__ movq(rcx, FieldOperand(rbx, i + kPointerSize));
__ movq(FieldOperand(rax, i), rdx);
__ movq(FieldOperand(rax, i + kPointerSize), rcx);
}
if ((size % (2 * kPointerSize)) != 0) {
__ movq(rdx, FieldOperand(rbx, size - kPointerSize));
__ movq(FieldOperand(rax, size - kPointerSize), rdx);
}
}
void LCodeGen::DoFunctionLiteral(LFunctionLiteral* instr) {
// Use the fast case closure allocation code that allocates in new
// space for nested functions that don't need literals cloning.
Handle<SharedFunctionInfo> shared_info = instr->shared_info();
bool pretenure = instr->hydrogen()->pretenure();
if (!pretenure && shared_info->num_literals() == 0) {
FastNewClosureStub stub(
shared_info->strict_mode() ? kStrictMode : kNonStrictMode);
__ Push(shared_info);
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
} else {
__ push(rsi);
__ Push(shared_info);
__ PushRoot(pretenure ?
Heap::kTrueValueRootIndex :
Heap::kFalseValueRootIndex);
CallRuntime(Runtime::kNewClosure, 3, instr);
}
}
void LCodeGen::DoTypeof(LTypeof* instr) {
LOperand* input = instr->InputAt(0);
if (input->IsConstantOperand()) {
__ Push(ToHandle(LConstantOperand::cast(input)));
} else if (input->IsRegister()) {
__ push(ToRegister(input));
} else {
ASSERT(input->IsStackSlot());
__ push(ToOperand(input));
}
CallRuntime(Runtime::kTypeof, 1, instr);
}
void LCodeGen::DoTypeofIs(LTypeofIs* instr) {
Register input = ToRegister(instr->InputAt(0));
Register result = ToRegister(instr->result());
Label true_label;
Label false_label;
NearLabel done;
Condition final_branch_condition = EmitTypeofIs(&true_label,
&false_label,
input,
instr->type_literal());
__ j(final_branch_condition, &true_label);
__ bind(&false_label);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ jmp(&done);
__ bind(&true_label);
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ bind(&done);
}
void LCodeGen::EmitPushConstantOperand(LOperand* operand) {
ASSERT(operand->IsConstantOperand());
LConstantOperand* const_op = LConstantOperand::cast(operand);
Handle<Object> literal = chunk_->LookupLiteral(const_op);
Representation r = chunk_->LookupLiteralRepresentation(const_op);
if (r.IsInteger32()) {
ASSERT(literal->IsNumber());
__ push(Immediate(static_cast<int32_t>(literal->Number())));
} else if (r.IsDouble()) {
Abort("unsupported double immediate");
} else {
ASSERT(r.IsTagged());
__ Push(literal);
}
}
void LCodeGen::DoTypeofIsAndBranch(LTypeofIsAndBranch* instr) {
Register input = ToRegister(instr->InputAt(0));
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
Label* true_label = chunk_->GetAssemblyLabel(true_block);
Label* false_label = chunk_->GetAssemblyLabel(false_block);
Condition final_branch_condition = EmitTypeofIs(true_label,
false_label,
input,
instr->type_literal());
EmitBranch(true_block, false_block, final_branch_condition);
}
Condition LCodeGen::EmitTypeofIs(Label* true_label,
Label* false_label,
Register input,
Handle<String> type_name) {
Condition final_branch_condition = no_condition;
if (type_name->Equals(heap()->number_symbol())) {
__ JumpIfSmi(input, true_label);
__ CompareRoot(FieldOperand(input, HeapObject::kMapOffset),
Heap::kHeapNumberMapRootIndex);
final_branch_condition = equal;
} else if (type_name->Equals(heap()->string_symbol())) {
__ JumpIfSmi(input, false_label);
__ CmpObjectType(input, FIRST_NONSTRING_TYPE, input);
__ j(above_equal, false_label);
__ testb(FieldOperand(input, Map::kBitFieldOffset),
Immediate(1 << Map::kIsUndetectable));
final_branch_condition = zero;
} else if (type_name->Equals(heap()->boolean_symbol())) {
__ CompareRoot(input, Heap::kTrueValueRootIndex);
__ j(equal, true_label);
__ CompareRoot(input, Heap::kFalseValueRootIndex);
final_branch_condition = equal;
} else if (type_name->Equals(heap()->undefined_symbol())) {
__ CompareRoot(input, Heap::kUndefinedValueRootIndex);
__ j(equal, true_label);
__ JumpIfSmi(input, false_label);
// Check for undetectable objects => true.
__ movq(input, FieldOperand(input, HeapObject::kMapOffset));
__ testb(FieldOperand(input, Map::kBitFieldOffset),
Immediate(1 << Map::kIsUndetectable));
final_branch_condition = not_zero;
} else if (type_name->Equals(heap()->function_symbol())) {
__ JumpIfSmi(input, false_label);
__ CmpObjectType(input, FIRST_FUNCTION_CLASS_TYPE, input);
final_branch_condition = above_equal;
} else if (type_name->Equals(heap()->object_symbol())) {
__ JumpIfSmi(input, false_label);
__ CompareRoot(input, Heap::kNullValueRootIndex);
__ j(equal, true_label);
__ CmpObjectType(input, FIRST_JS_OBJECT_TYPE, input);
__ j(below, false_label);
__ CmpInstanceType(input, FIRST_FUNCTION_CLASS_TYPE);
__ j(above_equal, false_label);
// Check for undetectable objects => false.
__ testb(FieldOperand(input, Map::kBitFieldOffset),
Immediate(1 << Map::kIsUndetectable));
final_branch_condition = zero;
} else {
final_branch_condition = never;
__ jmp(false_label);
}
return final_branch_condition;
}
void LCodeGen::DoIsConstructCall(LIsConstructCall* instr) {
Register result = ToRegister(instr->result());
NearLabel true_label;
NearLabel false_label;
NearLabel done;
EmitIsConstructCall(result);
__ j(equal, &true_label);
__ LoadRoot(result, Heap::kFalseValueRootIndex);
__ jmp(&done);
__ bind(&true_label);
__ LoadRoot(result, Heap::kTrueValueRootIndex);
__ bind(&done);
}
void LCodeGen::DoIsConstructCallAndBranch(LIsConstructCallAndBranch* instr) {
Register temp = ToRegister(instr->TempAt(0));
int true_block = chunk_->LookupDestination(instr->true_block_id());
int false_block = chunk_->LookupDestination(instr->false_block_id());
EmitIsConstructCall(temp);
EmitBranch(true_block, false_block, equal);
}
void LCodeGen::EmitIsConstructCall(Register temp) {
// Get the frame pointer for the calling frame.
__ movq(temp, Operand(rbp, StandardFrameConstants::kCallerFPOffset));
// Skip the arguments adaptor frame if it exists.
NearLabel check_frame_marker;
__ Cmp(Operand(temp, StandardFrameConstants::kContextOffset),
Smi::FromInt(StackFrame::ARGUMENTS_ADAPTOR));
__ j(not_equal, &check_frame_marker);
__ movq(temp, Operand(rax, StandardFrameConstants::kCallerFPOffset));
// Check the marker in the calling frame.
__ bind(&check_frame_marker);
__ Cmp(Operand(temp, StandardFrameConstants::kMarkerOffset),
Smi::FromInt(StackFrame::CONSTRUCT));
}
void LCodeGen::DoLazyBailout(LLazyBailout* instr) {
// No code for lazy bailout instruction. Used to capture environment after a
// call for populating the safepoint data with deoptimization data.
}
void LCodeGen::DoDeoptimize(LDeoptimize* instr) {
DeoptimizeIf(no_condition, instr->environment());
}
void LCodeGen::DoDeleteProperty(LDeleteProperty* instr) {
LOperand* obj = instr->object();
LOperand* key = instr->key();
// Push object.
if (obj->IsRegister()) {
__ push(ToRegister(obj));
} else {
__ push(ToOperand(obj));
}
// Push key.
if (key->IsConstantOperand()) {
EmitPushConstantOperand(key);
} else if (key->IsRegister()) {
__ push(ToRegister(key));
} else {
__ push(ToOperand(key));
}
ASSERT(instr->HasPointerMap() && instr->HasDeoptimizationEnvironment());
LPointerMap* pointers = instr->pointer_map();
LEnvironment* env = instr->deoptimization_environment();
RecordPosition(pointers->position());
RegisterEnvironmentForDeoptimization(env);
// Create safepoint generator that will also ensure enough space in the
// reloc info for patching in deoptimization (since this is invoking a
// builtin)
SafepointGenerator safepoint_generator(this,
pointers,
env->deoptimization_index());
__ Push(Smi::FromInt(strict_mode_flag()));
__ InvokeBuiltin(Builtins::DELETE, CALL_FUNCTION, &safepoint_generator);
}
void LCodeGen::DoStackCheck(LStackCheck* instr) {
// Perform stack overflow check.
NearLabel done;
__ CompareRoot(rsp, Heap::kStackLimitRootIndex);
__ j(above_equal, &done);
StackCheckStub stub;
CallCode(stub.GetCode(), RelocInfo::CODE_TARGET, instr);
__ bind(&done);
}
void LCodeGen::DoOsrEntry(LOsrEntry* instr) {
// This is a pseudo-instruction that ensures that the environment here is
// properly registered for deoptimization and records the assembler's PC
// offset.
LEnvironment* environment = instr->environment();
environment->SetSpilledRegisters(instr->SpilledRegisterArray(),
instr->SpilledDoubleRegisterArray());
// If the environment were already registered, we would have no way of
// backpatching it with the spill slot operands.
ASSERT(!environment->HasBeenRegistered());
RegisterEnvironmentForDeoptimization(environment);
ASSERT(osr_pc_offset_ == -1);
osr_pc_offset_ = masm()->pc_offset();
}
#undef __
} } // namespace v8::internal
#endif // V8_TARGET_ARCH_X64