// Copyright 2014 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "src/deoptimizer.h"
#include "src/codegen.h"
#include "src/full-codegen/full-codegen.h"
#include "src/register-configuration.h"
#include "src/safepoint-table.h"
namespace v8 {
namespace internal {
// LAY + LGHI/LHI + BRCL
const int Deoptimizer::table_entry_size_ = 16;
int Deoptimizer::patch_size() {
#if V8_TARGET_ARCH_S390X
const int kCallInstructionSize = 16;
#else
const int kCallInstructionSize = 10;
#endif
return kCallInstructionSize;
}
void Deoptimizer::EnsureRelocSpaceForLazyDeoptimization(Handle<Code> code) {
// Empty because there is no need for relocation information for the code
// patching in Deoptimizer::PatchCodeForDeoptimization below.
}
void Deoptimizer::PatchCodeForDeoptimization(Isolate* isolate, Code* code) {
Address code_start_address = code->instruction_start();
// Invalidate the relocation information, as it will become invalid by the
// code patching below, and is not needed any more.
code->InvalidateRelocation();
if (FLAG_zap_code_space) {
// Fail hard and early if we enter this code object again.
byte* pointer = code->FindCodeAgeSequence();
if (pointer != NULL) {
pointer += kNoCodeAgeSequenceLength;
} else {
pointer = code->instruction_start();
}
CodePatcher patcher(isolate, pointer, 2);
patcher.masm()->bkpt(0);
DeoptimizationInputData* data =
DeoptimizationInputData::cast(code->deoptimization_data());
int osr_offset = data->OsrPcOffset()->value();
if (osr_offset > 0) {
CodePatcher osr_patcher(isolate, code->instruction_start() + osr_offset,
2);
osr_patcher.masm()->bkpt(0);
}
}
DeoptimizationInputData* deopt_data =
DeoptimizationInputData::cast(code->deoptimization_data());
#ifdef DEBUG
Address prev_call_address = NULL;
#endif
// For each LLazyBailout instruction insert a call to the corresponding
// deoptimization entry.
for (int i = 0; i < deopt_data->DeoptCount(); i++) {
if (deopt_data->Pc(i)->value() == -1) continue;
Address call_address = code_start_address + deopt_data->Pc(i)->value();
Address deopt_entry = GetDeoptimizationEntry(isolate, i, LAZY);
// We need calls to have a predictable size in the unoptimized code, but
// this is optimized code, so we don't have to have a predictable size.
int call_size_in_bytes = MacroAssembler::CallSizeNotPredictableCodeSize(
deopt_entry, kRelocInfo_NONEPTR);
DCHECK(call_size_in_bytes <= patch_size());
CodePatcher patcher(isolate, call_address, call_size_in_bytes);
patcher.masm()->Call(deopt_entry, kRelocInfo_NONEPTR);
DCHECK(prev_call_address == NULL ||
call_address >= prev_call_address + patch_size());
DCHECK(call_address + patch_size() <= code->instruction_end());
#ifdef DEBUG
prev_call_address = call_address;
#endif
}
}
void Deoptimizer::SetPlatformCompiledStubRegisters(
FrameDescription* output_frame, CodeStubDescriptor* descriptor) {
ApiFunction function(descriptor->deoptimization_handler());
ExternalReference xref(&function, ExternalReference::BUILTIN_CALL, isolate_);
intptr_t handler = reinterpret_cast<intptr_t>(xref.address());
int params = descriptor->GetHandlerParameterCount();
output_frame->SetRegister(r2.code(), params);
output_frame->SetRegister(r3.code(), handler);
}
void Deoptimizer::CopyDoubleRegisters(FrameDescription* output_frame) {
for (int i = 0; i < DoubleRegister::kNumRegisters; ++i) {
Float64 double_value = input_->GetDoubleRegister(i);
output_frame->SetDoubleRegister(i, double_value);
}
}
#define __ masm()->
// This code tries to be close to ia32 code so that any changes can be
// easily ported.
void Deoptimizer::TableEntryGenerator::Generate() {
GeneratePrologue();
// Save all the registers onto the stack
const int kNumberOfRegisters = Register::kNumRegisters;
RegList restored_regs = kJSCallerSaved | kCalleeSaved;
const int kDoubleRegsSize = kDoubleSize * DoubleRegister::kNumRegisters;
// Save all double registers before messing with them.
__ lay(sp, MemOperand(sp, -kDoubleRegsSize));
const RegisterConfiguration* config = RegisterConfiguration::Crankshaft();
for (int i = 0; i < config->num_allocatable_double_registers(); ++i) {
int code = config->GetAllocatableDoubleCode(i);
const DoubleRegister dreg = DoubleRegister::from_code(code);
int offset = code * kDoubleSize;
__ StoreDouble(dreg, MemOperand(sp, offset));
}
// Push all GPRs onto the stack
__ lay(sp, MemOperand(sp, -kNumberOfRegisters * kPointerSize));
__ StoreMultipleP(r0, sp, MemOperand(sp)); // Save all 16 registers
__ mov(ip, Operand(ExternalReference(Isolate::kCEntryFPAddress, isolate())));
__ StoreP(fp, MemOperand(ip));
const int kSavedRegistersAreaSize =
(kNumberOfRegisters * kPointerSize) + kDoubleRegsSize;
// Get the bailout id from the stack.
__ LoadP(r4, MemOperand(sp, kSavedRegistersAreaSize));
// Cleanse the Return address for 31-bit
__ CleanseP(r14);
// Get the address of the location in the code object (r5)(return
// address for lazy deoptimization) and compute the fp-to-sp delta in
// register r6.
__ LoadRR(r5, r14);
__ la(r6, MemOperand(sp, kSavedRegistersAreaSize + (1 * kPointerSize)));
__ SubP(r6, fp, r6);
// Allocate a new deoptimizer object.
// Pass six arguments in r2 to r7.
__ PrepareCallCFunction(6, r7);
__ LoadImmP(r2, Operand::Zero());
Label context_check;
__ LoadP(r3, MemOperand(fp, CommonFrameConstants::kContextOrFrameTypeOffset));
__ JumpIfSmi(r3, &context_check);
__ LoadP(r2, MemOperand(fp, JavaScriptFrameConstants::kFunctionOffset));
__ bind(&context_check);
__ LoadImmP(r3, Operand(type())); // bailout type,
// r4: bailout id already loaded.
// r5: code address or 0 already loaded.
// r6: Fp-to-sp delta.
// Parm6: isolate is passed on the stack.
__ mov(r7, Operand(ExternalReference::isolate_address(isolate())));
__ StoreP(r7, MemOperand(sp, kStackFrameExtraParamSlot * kPointerSize));
// Call Deoptimizer::New().
{
AllowExternalCallThatCantCauseGC scope(masm());
__ CallCFunction(ExternalReference::new_deoptimizer_function(isolate()), 6);
}
// Preserve "deoptimizer" object in register r2 and get the input
// frame descriptor pointer to r3 (deoptimizer->input_);
__ LoadP(r3, MemOperand(r2, Deoptimizer::input_offset()));
// Copy core registers into FrameDescription::registers_[kNumRegisters].
// DCHECK(Register::kNumRegisters == kNumberOfRegisters);
// __ mvc(MemOperand(r3, FrameDescription::registers_offset()),
// MemOperand(sp), kNumberOfRegisters * kPointerSize);
// Copy core registers into FrameDescription::registers_[kNumRegisters].
// TODO(john.yan): optimize the following code by using mvc instruction
DCHECK(Register::kNumRegisters == kNumberOfRegisters);
for (int i = 0; i < kNumberOfRegisters; i++) {
int offset = (i * kPointerSize) + FrameDescription::registers_offset();
__ LoadP(r4, MemOperand(sp, i * kPointerSize));
__ StoreP(r4, MemOperand(r3, offset));
}
int double_regs_offset = FrameDescription::double_registers_offset();
// Copy double registers to
// double_registers_[DoubleRegister::kNumRegisters]
for (int i = 0; i < config->num_allocatable_double_registers(); ++i) {
int code = config->GetAllocatableDoubleCode(i);
int dst_offset = code * kDoubleSize + double_regs_offset;
int src_offset = code * kDoubleSize + kNumberOfRegisters * kPointerSize;
// TODO(joransiu): MVC opportunity
__ LoadDouble(d0, MemOperand(sp, src_offset));
__ StoreDouble(d0, MemOperand(r3, dst_offset));
}
// Remove the bailout id and the saved registers from the stack.
__ la(sp, MemOperand(sp, kSavedRegistersAreaSize + (1 * kPointerSize)));
// Compute a pointer to the unwinding limit in register r4; that is
// the first stack slot not part of the input frame.
__ LoadP(r4, MemOperand(r3, FrameDescription::frame_size_offset()));
__ AddP(r4, sp);
// Unwind the stack down to - but not including - the unwinding
// limit and copy the contents of the activation frame to the input
// frame description.
__ la(r5, MemOperand(r3, FrameDescription::frame_content_offset()));
Label pop_loop;
Label pop_loop_header;
__ b(&pop_loop_header, Label::kNear);
__ bind(&pop_loop);
__ pop(r6);
__ StoreP(r6, MemOperand(r5, 0));
__ la(r5, MemOperand(r5, kPointerSize));
__ bind(&pop_loop_header);
__ CmpP(r4, sp);
__ bne(&pop_loop);
// Compute the output frame in the deoptimizer.
__ push(r2); // Preserve deoptimizer object across call.
// r2: deoptimizer object; r3: scratch.
__ PrepareCallCFunction(1, r3);
// Call Deoptimizer::ComputeOutputFrames().
{
AllowExternalCallThatCantCauseGC scope(masm());
__ CallCFunction(
ExternalReference::compute_output_frames_function(isolate()), 1);
}
__ pop(r2); // Restore deoptimizer object (class Deoptimizer).
__ LoadP(sp, MemOperand(r2, Deoptimizer::caller_frame_top_offset()));
// Replace the current (input) frame with the output frames.
Label outer_push_loop, inner_push_loop, outer_loop_header, inner_loop_header;
// Outer loop state: r6 = current "FrameDescription** output_",
// r3 = one past the last FrameDescription**.
__ LoadlW(r3, MemOperand(r2, Deoptimizer::output_count_offset()));
__ LoadP(r6, MemOperand(r2, Deoptimizer::output_offset())); // r6 is output_.
__ ShiftLeftP(r3, r3, Operand(kPointerSizeLog2));
__ AddP(r3, r6, r3);
__ b(&outer_loop_header, Label::kNear);
__ bind(&outer_push_loop);
// Inner loop state: r4 = current FrameDescription*, r5 = loop index.
__ LoadP(r4, MemOperand(r6, 0)); // output_[ix]
__ LoadP(r5, MemOperand(r4, FrameDescription::frame_size_offset()));
__ b(&inner_loop_header, Label::kNear);
__ bind(&inner_push_loop);
__ AddP(r5, Operand(-sizeof(intptr_t)));
__ AddP(r8, r4, r5);
__ LoadP(r8, MemOperand(r8, FrameDescription::frame_content_offset()));
__ push(r8);
__ bind(&inner_loop_header);
__ CmpP(r5, Operand::Zero());
__ bne(&inner_push_loop); // test for gt?
__ AddP(r6, r6, Operand(kPointerSize));
__ bind(&outer_loop_header);
__ CmpP(r6, r3);
__ blt(&outer_push_loop);
__ LoadP(r3, MemOperand(r2, Deoptimizer::input_offset()));
for (int i = 0; i < config->num_allocatable_double_registers(); ++i) {
int code = config->GetAllocatableDoubleCode(i);
const DoubleRegister dreg = DoubleRegister::from_code(code);
int src_offset = code * kDoubleSize + double_regs_offset;
__ ld(dreg, MemOperand(r3, src_offset));
}
// Push state, pc, and continuation from the last output frame.
__ LoadP(r8, MemOperand(r4, FrameDescription::state_offset()));
__ push(r8);
__ LoadP(r8, MemOperand(r4, FrameDescription::pc_offset()));
__ push(r8);
__ LoadP(r8, MemOperand(r4, FrameDescription::continuation_offset()));
__ push(r8);
// Restore the registers from the last output frame.
__ LoadRR(r1, r4);
for (int i = kNumberOfRegisters - 1; i > 0; i--) {
int offset = (i * kPointerSize) + FrameDescription::registers_offset();
if ((restored_regs & (1 << i)) != 0) {
__ LoadP(ToRegister(i), MemOperand(r1, offset));
}
}
__ InitializeRootRegister();
__ pop(ip); // get continuation, leave pc on stack
__ pop(r14);
__ Jump(ip);
__ stop("Unreachable.");
}
void Deoptimizer::TableEntryGenerator::GeneratePrologue() {
// Create a sequence of deoptimization entries. Note that any
// registers may be still live.
Label done;
for (int i = 0; i < count(); i++) {
int start = masm()->pc_offset();
USE(start);
__ lay(sp, MemOperand(sp, -kPointerSize));
__ LoadImmP(ip, Operand(i));
__ b(&done);
int end = masm()->pc_offset();
USE(end);
DCHECK(masm()->pc_offset() - start == table_entry_size_);
}
__ bind(&done);
__ StoreP(ip, MemOperand(sp));
}
void FrameDescription::SetCallerPc(unsigned offset, intptr_t value) {
SetFrameSlot(offset, value);
}
void FrameDescription::SetCallerFp(unsigned offset, intptr_t value) {
SetFrameSlot(offset, value);
}
void FrameDescription::SetCallerConstantPool(unsigned offset, intptr_t value) {
// No out-of-line constant pool support.
UNREACHABLE();
}
#undef __
} // namespace internal
} // namespace v8