src/mips64/macro-assembler-mips64.cc

// Copyright 2012 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 <limits.h>  // For LONG_MIN, LONG_MAX.

#include "src/v8.h"

#if V8_TARGET_ARCH_MIPS64

#include "src/base/division-by-constant.h"
#include "src/bootstrapper.h"
#include "src/codegen.h"
#include "src/cpu-profiler.h"
#include "src/debug.h"
#include "src/isolate-inl.h"
#include "src/runtime/runtime.h"

namespace v8 {
namespace internal {

MacroAssembler::MacroAssembler(Isolate* arg_isolate, void* buffer, int size)
    : Assembler(arg_isolate, buffer, size),
      generating_stub_(false),
      has_frame_(false),
      has_double_zero_reg_set_(false) {
  if (isolate() != NULL) {
    code_object_ = Handle<Object>(isolate()->heap()->undefined_value(),
                                  isolate());
  }
}


void MacroAssembler::Load(Register dst,
                          const MemOperand& src,
                          Representation r) {
  DCHECK(!r.IsDouble());
  if (r.IsInteger8()) {
    lb(dst, src);
  } else if (r.IsUInteger8()) {
    lbu(dst, src);
  } else if (r.IsInteger16()) {
    lh(dst, src);
  } else if (r.IsUInteger16()) {
    lhu(dst, src);
  } else if (r.IsInteger32()) {
    lw(dst, src);
  } else {
    ld(dst, src);
  }
}


void MacroAssembler::Store(Register src,
                           const MemOperand& dst,
                           Representation r) {
  DCHECK(!r.IsDouble());
  if (r.IsInteger8() || r.IsUInteger8()) {
    sb(src, dst);
  } else if (r.IsInteger16() || r.IsUInteger16()) {
    sh(src, dst);
  } else if (r.IsInteger32()) {
    sw(src, dst);
  } else {
    if (r.IsHeapObject()) {
      AssertNotSmi(src);
    } else if (r.IsSmi()) {
      AssertSmi(src);
    }
    sd(src, dst);
  }
}


void MacroAssembler::LoadRoot(Register destination,
                              Heap::RootListIndex index) {
  ld(destination, MemOperand(s6, index << kPointerSizeLog2));
}


void MacroAssembler::LoadRoot(Register destination,
                              Heap::RootListIndex index,
                              Condition cond,
                              Register src1, const Operand& src2) {
  Branch(2, NegateCondition(cond), src1, src2);
  ld(destination, MemOperand(s6, index << kPointerSizeLog2));
}


void MacroAssembler::StoreRoot(Register source,
                               Heap::RootListIndex index) {
  sd(source, MemOperand(s6, index << kPointerSizeLog2));
}


void MacroAssembler::StoreRoot(Register source,
                               Heap::RootListIndex index,
                               Condition cond,
                               Register src1, const Operand& src2) {
  Branch(2, NegateCondition(cond), src1, src2);
  sd(source, MemOperand(s6, index << kPointerSizeLog2));
}


// Push and pop all registers that can hold pointers.
void MacroAssembler::PushSafepointRegisters() {
  // Safepoints expect a block of kNumSafepointRegisters values on the
  // stack, so adjust the stack for unsaved registers.
  const int num_unsaved = kNumSafepointRegisters - kNumSafepointSavedRegisters;
  DCHECK(num_unsaved >= 0);
  if (num_unsaved > 0) {
    Dsubu(sp, sp, Operand(num_unsaved * kPointerSize));
  }
  MultiPush(kSafepointSavedRegisters);
}


void MacroAssembler::PopSafepointRegisters() {
  const int num_unsaved = kNumSafepointRegisters - kNumSafepointSavedRegisters;
  MultiPop(kSafepointSavedRegisters);
  if (num_unsaved > 0) {
    Daddu(sp, sp, Operand(num_unsaved * kPointerSize));
  }
}


void MacroAssembler::StoreToSafepointRegisterSlot(Register src, Register dst) {
  sd(src, SafepointRegisterSlot(dst));
}


void MacroAssembler::LoadFromSafepointRegisterSlot(Register dst, Register src) {
  ld(dst, SafepointRegisterSlot(src));
}


int MacroAssembler::SafepointRegisterStackIndex(int reg_code) {
  // The registers are pushed starting with the highest encoding,
  // which means that lowest encodings are closest to the stack pointer.
  return kSafepointRegisterStackIndexMap[reg_code];
}


MemOperand MacroAssembler::SafepointRegisterSlot(Register reg) {
  return MemOperand(sp, SafepointRegisterStackIndex(reg.code()) * kPointerSize);
}


MemOperand MacroAssembler::SafepointRegistersAndDoublesSlot(Register reg) {
  UNIMPLEMENTED_MIPS();
  // General purpose registers are pushed last on the stack.
  int doubles_size = FPURegister::NumAllocatableRegisters() * kDoubleSize;
  int register_offset = SafepointRegisterStackIndex(reg.code()) * kPointerSize;
  return MemOperand(sp, doubles_size + register_offset);
}


void MacroAssembler::InNewSpace(Register object,
                                Register scratch,
                                Condition cc,
                                Label* branch) {
  DCHECK(cc == eq || cc == ne);
  And(scratch, object, Operand(ExternalReference::new_space_mask(isolate())));
  Branch(branch, cc, scratch,
         Operand(ExternalReference::new_space_start(isolate())));
}


void MacroAssembler::RecordWriteField(
    Register object,
    int offset,
    Register value,
    Register dst,
    RAStatus ra_status,
    SaveFPRegsMode save_fp,
    RememberedSetAction remembered_set_action,
    SmiCheck smi_check,
    PointersToHereCheck pointers_to_here_check_for_value) {
  DCHECK(!AreAliased(value, dst, t8, object));
  // First, check if a write barrier is even needed. The tests below
  // catch stores of Smis.
  Label done;

  // Skip barrier if writing a smi.
  if (smi_check == INLINE_SMI_CHECK) {
    JumpIfSmi(value, &done);
  }

  // Although the object register is tagged, the offset is relative to the start
  // of the object, so so offset must be a multiple of kPointerSize.
  DCHECK(IsAligned(offset, kPointerSize));

  Daddu(dst, object, Operand(offset - kHeapObjectTag));
  if (emit_debug_code()) {
    Label ok;
    And(t8, dst, Operand((1 << kPointerSizeLog2) - 1));
    Branch(&ok, eq, t8, Operand(zero_reg));
    stop("Unaligned cell in write barrier");
    bind(&ok);
  }

  RecordWrite(object,
              dst,
              value,
              ra_status,
              save_fp,
              remembered_set_action,
              OMIT_SMI_CHECK,
              pointers_to_here_check_for_value);

  bind(&done);

  // Clobber clobbered input registers when running with the debug-code flag
  // turned on to provoke errors.
  if (emit_debug_code()) {
    li(value, Operand(bit_cast<int64_t>(kZapValue + 4)));
    li(dst, Operand(bit_cast<int64_t>(kZapValue + 8)));
  }
}


// Will clobber 4 registers: object, map, dst, ip.  The
// register 'object' contains a heap object pointer.
void MacroAssembler::RecordWriteForMap(Register object,
                                       Register map,
                                       Register dst,
                                       RAStatus ra_status,
                                       SaveFPRegsMode fp_mode) {
  if (emit_debug_code()) {
    DCHECK(!dst.is(at));
    ld(dst, FieldMemOperand(map, HeapObject::kMapOffset));
    Check(eq,
          kWrongAddressOrValuePassedToRecordWrite,
          dst,
          Operand(isolate()->factory()->meta_map()));
  }

  if (!FLAG_incremental_marking) {
    return;
  }

  if (emit_debug_code()) {
    ld(at, FieldMemOperand(object, HeapObject::kMapOffset));
    Check(eq,
          kWrongAddressOrValuePassedToRecordWrite,
          map,
          Operand(at));
  }

  Label done;

  // A single check of the map's pages interesting flag suffices, since it is
  // only set during incremental collection, and then it's also guaranteed that
  // the from object's page's interesting flag is also set.  This optimization
  // relies on the fact that maps can never be in new space.
  CheckPageFlag(map,
                map,  // Used as scratch.
                MemoryChunk::kPointersToHereAreInterestingMask,
                eq,
                &done);

  Daddu(dst, object, Operand(HeapObject::kMapOffset - kHeapObjectTag));
  if (emit_debug_code()) {
    Label ok;
    And(at, dst, Operand((1 << kPointerSizeLog2) - 1));
    Branch(&ok, eq, at, Operand(zero_reg));
    stop("Unaligned cell in write barrier");
    bind(&ok);
  }

  // Record the actual write.
  if (ra_status == kRAHasNotBeenSaved) {
    push(ra);
  }
  RecordWriteStub stub(isolate(), object, map, dst, OMIT_REMEMBERED_SET,
                       fp_mode);
  CallStub(&stub);
  if (ra_status == kRAHasNotBeenSaved) {
    pop(ra);
  }

  bind(&done);

  // Count number of write barriers in generated code.
  isolate()->counters()->write_barriers_static()->Increment();
  IncrementCounter(isolate()->counters()->write_barriers_dynamic(), 1, at, dst);

  // Clobber clobbered registers when running with the debug-code flag
  // turned on to provoke errors.
  if (emit_debug_code()) {
    li(dst, Operand(bit_cast<int64_t>(kZapValue + 12)));
    li(map, Operand(bit_cast<int64_t>(kZapValue + 16)));
  }
}


// Will clobber 4 registers: object, address, scratch, ip.  The
// register 'object' contains a heap object pointer.  The heap object
// tag is shifted away.
void MacroAssembler::RecordWrite(
    Register object,
    Register address,
    Register value,
    RAStatus ra_status,
    SaveFPRegsMode fp_mode,
    RememberedSetAction remembered_set_action,
    SmiCheck smi_check,
    PointersToHereCheck pointers_to_here_check_for_value) {
  DCHECK(!AreAliased(object, address, value, t8));
  DCHECK(!AreAliased(object, address, value, t9));

  if (emit_debug_code()) {
    ld(at, MemOperand(address));
    Assert(
        eq, kWrongAddressOrValuePassedToRecordWrite, at, Operand(value));
  }

  if (remembered_set_action == OMIT_REMEMBERED_SET &&
      !FLAG_incremental_marking) {
    return;
  }

  // First, check if a write barrier is even needed. The tests below
  // catch stores of smis and stores into the young generation.
  Label done;

  if (smi_check == INLINE_SMI_CHECK) {
    DCHECK_EQ(0, kSmiTag);
    JumpIfSmi(value, &done);
  }

  if (pointers_to_here_check_for_value != kPointersToHereAreAlwaysInteresting) {
    CheckPageFlag(value,
                  value,  // Used as scratch.
                  MemoryChunk::kPointersToHereAreInterestingMask,
                  eq,
                  &done);
  }
  CheckPageFlag(object,
                value,  // Used as scratch.
                MemoryChunk::kPointersFromHereAreInterestingMask,
                eq,
                &done);

  // Record the actual write.
  if (ra_status == kRAHasNotBeenSaved) {
    push(ra);
  }
  RecordWriteStub stub(isolate(), object, value, address, remembered_set_action,
                       fp_mode);
  CallStub(&stub);
  if (ra_status == kRAHasNotBeenSaved) {
    pop(ra);
  }

  bind(&done);

  // Count number of write barriers in generated code.
  isolate()->counters()->write_barriers_static()->Increment();
  IncrementCounter(isolate()->counters()->write_barriers_dynamic(), 1, at,
                   value);

  // Clobber clobbered registers when running with the debug-code flag
  // turned on to provoke errors.
  if (emit_debug_code()) {
    li(address, Operand(bit_cast<int64_t>(kZapValue + 12)));
    li(value, Operand(bit_cast<int64_t>(kZapValue + 16)));
  }
}


void MacroAssembler::RememberedSetHelper(Register object,  // For debug tests.
                                         Register address,
                                         Register scratch,
                                         SaveFPRegsMode fp_mode,
                                         RememberedSetFinalAction and_then) {
  Label done;
  if (emit_debug_code()) {
    Label ok;
    JumpIfNotInNewSpace(object, scratch, &ok);
    stop("Remembered set pointer is in new space");
    bind(&ok);
  }
  // Load store buffer top.
  ExternalReference store_buffer =
      ExternalReference::store_buffer_top(isolate());
  li(t8, Operand(store_buffer));
  ld(scratch, MemOperand(t8));
  // Store pointer to buffer and increment buffer top.
  sd(address, MemOperand(scratch));
  Daddu(scratch, scratch, kPointerSize);
  // Write back new top of buffer.
  sd(scratch, MemOperand(t8));
  // Call stub on end of buffer.
  // Check for end of buffer.
  And(t8, scratch, Operand(StoreBuffer::kStoreBufferOverflowBit));
  DCHECK(!scratch.is(t8));
  if (and_then == kFallThroughAtEnd) {
    Branch(&done, eq, t8, Operand(zero_reg));
  } else {
    DCHECK(and_then == kReturnAtEnd);
    Ret(eq, t8, Operand(zero_reg));
  }
  push(ra);
  StoreBufferOverflowStub store_buffer_overflow(isolate(), fp_mode);
  CallStub(&store_buffer_overflow);
  pop(ra);
  bind(&done);
  if (and_then == kReturnAtEnd) {
    Ret();
  }
}


// -----------------------------------------------------------------------------
// Allocation support.


void MacroAssembler::CheckAccessGlobalProxy(Register holder_reg,
                                            Register scratch,
                                            Label* miss) {
  Label same_contexts;

  DCHECK(!holder_reg.is(scratch));
  DCHECK(!holder_reg.is(at));
  DCHECK(!scratch.is(at));

  // Load current lexical context from the stack frame.
  ld(scratch, MemOperand(fp, StandardFrameConstants::kContextOffset));
  // In debug mode, make sure the lexical context is set.
#ifdef DEBUG
  Check(ne, kWeShouldNotHaveAnEmptyLexicalContext,
      scratch, Operand(zero_reg));
#endif

  // Load the native context of the current context.
  int offset =
      Context::kHeaderSize + Context::GLOBAL_OBJECT_INDEX * kPointerSize;
  ld(scratch, FieldMemOperand(scratch, offset));
  ld(scratch, FieldMemOperand(scratch, GlobalObject::kNativeContextOffset));

  // Check the context is a native context.
  if (emit_debug_code()) {
    push(holder_reg);  // Temporarily save holder on the stack.
    // Read the first word and compare to the native_context_map.
    ld(holder_reg, FieldMemOperand(scratch, HeapObject::kMapOffset));
    LoadRoot(at, Heap::kNativeContextMapRootIndex);
    Check(eq, kJSGlobalObjectNativeContextShouldBeANativeContext,
          holder_reg, Operand(at));
    pop(holder_reg);  // Restore holder.
  }

  // Check if both contexts are the same.
  ld(at, FieldMemOperand(holder_reg, JSGlobalProxy::kNativeContextOffset));
  Branch(&same_contexts, eq, scratch, Operand(at));

  // Check the context is a native context.
  if (emit_debug_code()) {
    push(holder_reg);  // Temporarily save holder on the stack.
    mov(holder_reg, at);  // Move at to its holding place.
    LoadRoot(at, Heap::kNullValueRootIndex);
    Check(ne, kJSGlobalProxyContextShouldNotBeNull,
          holder_reg, Operand(at));

    ld(holder_reg, FieldMemOperand(holder_reg, HeapObject::kMapOffset));
    LoadRoot(at, Heap::kNativeContextMapRootIndex);
    Check(eq, kJSGlobalObjectNativeContextShouldBeANativeContext,
          holder_reg, Operand(at));
    // Restore at is not needed. at is reloaded below.
    pop(holder_reg);  // Restore holder.
    // Restore at to holder's context.
    ld(at, FieldMemOperand(holder_reg, JSGlobalProxy::kNativeContextOffset));
  }

  // Check that the security token in the calling global object is
  // compatible with the security token in the receiving global
  // object.
  int token_offset = Context::kHeaderSize +
                     Context::SECURITY_TOKEN_INDEX * kPointerSize;

  ld(scratch, FieldMemOperand(scratch, token_offset));
  ld(at, FieldMemOperand(at, token_offset));
  Branch(miss, ne, scratch, Operand(at));

  bind(&same_contexts);
}


// Compute the hash code from the untagged key.  This must be kept in sync with
// ComputeIntegerHash in utils.h and KeyedLoadGenericStub in
// code-stub-hydrogen.cc
void MacroAssembler::GetNumberHash(Register reg0, Register scratch) {
  // First of all we assign the hash seed to scratch.
  LoadRoot(scratch, Heap::kHashSeedRootIndex);
  SmiUntag(scratch);

  // Xor original key with a seed.
  xor_(reg0, reg0, scratch);

  // Compute the hash code from the untagged key.  This must be kept in sync
  // with ComputeIntegerHash in utils.h.
  //
  // hash = ~hash + (hash << 15);
  // The algorithm uses 32-bit integer values.
  nor(scratch, reg0, zero_reg);
  sll(at, reg0, 15);
  addu(reg0, scratch, at);

  // hash = hash ^ (hash >> 12);
  srl(at, reg0, 12);
  xor_(reg0, reg0, at);

  // hash = hash + (hash << 2);
  sll(at, reg0, 2);
  addu(reg0, reg0, at);

  // hash = hash ^ (hash >> 4);
  srl(at, reg0, 4);
  xor_(reg0, reg0, at);

  // hash = hash * 2057;
  sll(scratch, reg0, 11);
  sll(at, reg0, 3);
  addu(reg0, reg0, at);
  addu(reg0, reg0, scratch);

  // hash = hash ^ (hash >> 16);
  srl(at, reg0, 16);
  xor_(reg0, reg0, at);
}


void MacroAssembler::LoadFromNumberDictionary(Label* miss,
                                              Register elements,
                                              Register key,
                                              Register result,
                                              Register reg0,
                                              Register reg1,
                                              Register reg2) {
  // Register use:
  //
  // elements - holds the slow-case elements of the receiver on entry.
  //            Unchanged unless 'result' is the same register.
  //
  // key      - holds the smi key on entry.
  //            Unchanged unless 'result' is the same register.
  //
  //
  // result   - holds the result on exit if the load succeeded.
  //            Allowed to be the same as 'key' or 'result'.
  //            Unchanged on bailout so 'key' or 'result' can be used
  //            in further computation.
  //
  // Scratch registers:
  //
  // reg0 - holds the untagged key on entry and holds the hash once computed.
  //
  // reg1 - Used to hold the capacity mask of the dictionary.
  //
  // reg2 - Used for the index into the dictionary.
  // at   - Temporary (avoid MacroAssembler instructions also using 'at').
  Label done;

  GetNumberHash(reg0, reg1);

  // Compute the capacity mask.
  ld(reg1, FieldMemOperand(elements, SeededNumberDictionary::kCapacityOffset));
  SmiUntag(reg1, reg1);
  Dsubu(reg1, reg1, Operand(1));

  // Generate an unrolled loop that performs a few probes before giving up.
  for (int i = 0; i < kNumberDictionaryProbes; i++) {
    // Use reg2 for index calculations and keep the hash intact in reg0.
    mov(reg2, reg0);
    // Compute the masked index: (hash + i + i * i) & mask.
    if (i > 0) {
      Daddu(reg2, reg2, Operand(SeededNumberDictionary::GetProbeOffset(i)));
    }
    and_(reg2, reg2, reg1);

    // Scale the index by multiplying by the element size.
    DCHECK(SeededNumberDictionary::kEntrySize == 3);
    dsll(at, reg2, 1);  // 2x.
    daddu(reg2, reg2, at);  // reg2 = reg2 * 3.

    // Check if the key is identical to the name.
    dsll(at, reg2, kPointerSizeLog2);
    daddu(reg2, elements, at);

    ld(at, FieldMemOperand(reg2, SeededNumberDictionary::kElementsStartOffset));
    if (i != kNumberDictionaryProbes - 1) {
      Branch(&done, eq, key, Operand(at));
    } else {
      Branch(miss, ne, key, Operand(at));
    }
  }

  bind(&done);
  // Check that the value is a field property.
  // reg2: elements + (index * kPointerSize).
  const int kDetailsOffset =
      SeededNumberDictionary::kElementsStartOffset + 2 * kPointerSize;
  ld(reg1, FieldMemOperand(reg2, kDetailsOffset));
  DCHECK_EQ(FIELD, 0);
  And(at, reg1, Operand(Smi::FromInt(PropertyDetails::TypeField::kMask)));
  Branch(miss, ne, at, Operand(zero_reg));

  // Get the value at the masked, scaled index and return.
  const int kValueOffset =
      SeededNumberDictionary::kElementsStartOffset + kPointerSize;
  ld(result, FieldMemOperand(reg2, kValueOffset));
}


// ---------------------------------------------------------------------------
// Instruction macros.

void MacroAssembler::Addu(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    addu(rd, rs, rt.rm());
  } else {
    if (is_int16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      addiu(rd, rs, rt.imm64_);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      addu(rd, rs, at);
    }
  }
}


void MacroAssembler::Daddu(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    daddu(rd, rs, rt.rm());
  } else {
    if (is_int16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      daddiu(rd, rs, rt.imm64_);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      daddu(rd, rs, at);
    }
  }
}


void MacroAssembler::Subu(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    subu(rd, rs, rt.rm());
  } else {
    if (is_int16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      addiu(rd, rs, -rt.imm64_);  // No subiu instr, use addiu(x, y, -imm).
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      subu(rd, rs, at);
    }
  }
}


void MacroAssembler::Dsubu(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    dsubu(rd, rs, rt.rm());
  } else {
    if (is_int16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      daddiu(rd, rs, -rt.imm64_);  // No subiu instr, use addiu(x, y, -imm).
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      dsubu(rd, rs, at);
    }
  }
}


void MacroAssembler::Mul(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    mul(rd, rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    mul(rd, rs, at);
  }
}


void MacroAssembler::Mulh(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant != kMips64r6) {
      mult(rs, rt.rm());
      mfhi(rd);
    } else {
      muh(rd, rs, rt.rm());
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant != kMips64r6) {
      mult(rs, at);
      mfhi(rd);
    } else {
      muh(rd, rs, at);
    }
  }
}


void MacroAssembler::Mulhu(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant != kMips64r6) {
      multu(rs, rt.rm());
      mfhi(rd);
    } else {
      muhu(rd, rs, rt.rm());
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant != kMips64r6) {
      multu(rs, at);
      mfhi(rd);
    } else {
      muhu(rd, rs, at);
    }
  }
}


void MacroAssembler::Dmul(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant == kMips64r6) {
      dmul(rd, rs, rt.rm());
    } else {
      dmult(rs, rt.rm());
      mflo(rd);
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant == kMips64r6) {
      dmul(rd, rs, at);
    } else {
      dmult(rs, at);
      mflo(rd);
    }
  }
}


void MacroAssembler::Dmulh(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant == kMips64r6) {
      dmuh(rd, rs, rt.rm());
    } else {
      dmult(rs, rt.rm());
      mfhi(rd);
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant == kMips64r6) {
      dmuh(rd, rs, at);
    } else {
      dmult(rs, at);
      mfhi(rd);
    }
  }
}


void MacroAssembler::Mult(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    mult(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    mult(rs, at);
  }
}


void MacroAssembler::Dmult(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    dmult(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    dmult(rs, at);
  }
}


void MacroAssembler::Multu(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    multu(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    multu(rs, at);
  }
}


void MacroAssembler::Dmultu(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    dmultu(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    dmultu(rs, at);
  }
}


void MacroAssembler::Div(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    div(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    div(rs, at);
  }
}


void MacroAssembler::Div(Register res, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant != kMips64r6) {
      div(rs, rt.rm());
      mflo(res);
    } else {
      div(res, rs, rt.rm());
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant != kMips64r6) {
      div(rs, at);
      mflo(res);
    } else {
      div(res, rs, at);
    }
  }
}


void MacroAssembler::Mod(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant != kMips64r6) {
      div(rs, rt.rm());
      mfhi(rd);
    } else {
      mod(rd, rs, rt.rm());
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant != kMips64r6) {
      div(rs, at);
      mfhi(rd);
    } else {
      mod(rd, rs, at);
    }
  }
}


void MacroAssembler::Modu(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant != kMips64r6) {
      divu(rs, rt.rm());
      mfhi(rd);
    } else {
      modu(rd, rs, rt.rm());
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant != kMips64r6) {
      divu(rs, at);
      mfhi(rd);
    } else {
      modu(rd, rs, at);
    }
  }
}


void MacroAssembler::Ddiv(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    ddiv(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    ddiv(rs, at);
  }
}


void MacroAssembler::Ddiv(Register rd, Register rs, const Operand& rt) {
  if (kArchVariant != kMips64r6) {
    if (rt.is_reg()) {
      ddiv(rs, rt.rm());
      mflo(rd);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      ddiv(rs, at);
      mflo(rd);
    }
  } else {
    if (rt.is_reg()) {
      ddiv(rd, rs, rt.rm());
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      ddiv(rd, rs, at);
    }
  }
}


void MacroAssembler::Divu(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    divu(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    divu(rs, at);
  }
}


void MacroAssembler::Divu(Register res, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant != kMips64r6) {
      divu(rs, rt.rm());
      mflo(res);
    } else {
      divu(res, rs, rt.rm());
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant != kMips64r6) {
      divu(rs, at);
      mflo(res);
    } else {
      divu(res, rs, at);
    }
  }
}


void MacroAssembler::Ddivu(Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    ddivu(rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    ddivu(rs, at);
  }
}


void MacroAssembler::Ddivu(Register res, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    if (kArchVariant != kMips64r6) {
      ddivu(rs, rt.rm());
      mflo(res);
    } else {
      ddivu(res, rs, rt.rm());
    }
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    if (kArchVariant != kMips64r6) {
      ddivu(rs, at);
      mflo(res);
    } else {
      ddivu(res, rs, at);
    }
  }
}


void MacroAssembler::Dmod(Register rd, Register rs, const Operand& rt) {
  if (kArchVariant != kMips64r6) {
    if (rt.is_reg()) {
      ddiv(rs, rt.rm());
      mfhi(rd);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      ddiv(rs, at);
      mfhi(rd);
    }
  } else {
    if (rt.is_reg()) {
      dmod(rd, rs, rt.rm());
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      dmod(rd, rs, at);
    }
  }
}


void MacroAssembler::Dmodu(Register rd, Register rs, const Operand& rt) {
  if (kArchVariant != kMips64r6) {
    if (rt.is_reg()) {
      ddivu(rs, rt.rm());
      mfhi(rd);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      ddivu(rs, at);
      mfhi(rd);
    }
  } else {
    if (rt.is_reg()) {
      dmodu(rd, rs, rt.rm());
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      dmodu(rd, rs, at);
    }
  }
}


void MacroAssembler::And(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    and_(rd, rs, rt.rm());
  } else {
    if (is_uint16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      andi(rd, rs, rt.imm64_);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      and_(rd, rs, at);
    }
  }
}


void MacroAssembler::Or(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    or_(rd, rs, rt.rm());
  } else {
    if (is_uint16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      ori(rd, rs, rt.imm64_);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      or_(rd, rs, at);
    }
  }
}


void MacroAssembler::Xor(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    xor_(rd, rs, rt.rm());
  } else {
    if (is_uint16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      xori(rd, rs, rt.imm64_);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      xor_(rd, rs, at);
    }
  }
}


void MacroAssembler::Nor(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    nor(rd, rs, rt.rm());
  } else {
    // li handles the relocation.
    DCHECK(!rs.is(at));
    li(at, rt);
    nor(rd, rs, at);
  }
}


void MacroAssembler::Neg(Register rs, const Operand& rt) {
  DCHECK(rt.is_reg());
  DCHECK(!at.is(rs));
  DCHECK(!at.is(rt.rm()));
  li(at, -1);
  xor_(rs, rt.rm(), at);
}


void MacroAssembler::Slt(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    slt(rd, rs, rt.rm());
  } else {
    if (is_int16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      slti(rd, rs, rt.imm64_);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      slt(rd, rs, at);
    }
  }
}


void MacroAssembler::Sltu(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    sltu(rd, rs, rt.rm());
  } else {
    if (is_int16(rt.imm64_) && !MustUseReg(rt.rmode_)) {
      sltiu(rd, rs, rt.imm64_);
    } else {
      // li handles the relocation.
      DCHECK(!rs.is(at));
      li(at, rt);
      sltu(rd, rs, at);
    }
  }
}


void MacroAssembler::Ror(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    rotrv(rd, rs, rt.rm());
  } else {
    rotr(rd, rs, rt.imm64_);
  }
}


void MacroAssembler::Dror(Register rd, Register rs, const Operand& rt) {
  if (rt.is_reg()) {
    drotrv(rd, rs, rt.rm());
  } else {
    drotr(rd, rs, rt.imm64_);
  }
}


void MacroAssembler::Pref(int32_t hint, const MemOperand& rs) {
    pref(hint, rs);
}


// ------------Pseudo-instructions-------------

void MacroAssembler::Ulw(Register rd, const MemOperand& rs) {
  lwr(rd, rs);
  lwl(rd, MemOperand(rs.rm(), rs.offset() + 3));
}


void MacroAssembler::Usw(Register rd, const MemOperand& rs) {
  swr(rd, rs);
  swl(rd, MemOperand(rs.rm(), rs.offset() + 3));
}


// Do 64-bit load from unaligned address. Note this only handles
// the specific case of 32-bit aligned, but not 64-bit aligned.
void MacroAssembler::Uld(Register rd, const MemOperand& rs, Register scratch) {
  // Assert fail if the offset from start of object IS actually aligned.
  // ONLY use with known misalignment, since there is performance cost.
  DCHECK((rs.offset() + kHeapObjectTag) & (kPointerSize - 1));
  // TODO(plind): endian dependency.
  lwu(rd, rs);
  lw(scratch, MemOperand(rs.rm(), rs.offset() + kPointerSize / 2));
  dsll32(scratch, scratch, 0);
  Daddu(rd, rd, scratch);
}


// Do 64-bit store to unaligned address. Note this only handles
// the specific case of 32-bit aligned, but not 64-bit aligned.
void MacroAssembler::Usd(Register rd, const MemOperand& rs, Register scratch) {
  // Assert fail if the offset from start of object IS actually aligned.
  // ONLY use with known misalignment, since there is performance cost.
  DCHECK((rs.offset() + kHeapObjectTag) & (kPointerSize - 1));
  // TODO(plind): endian dependency.
  sw(rd, rs);
  dsrl32(scratch, rd, 0);
  sw(scratch, MemOperand(rs.rm(), rs.offset() + kPointerSize / 2));
}


void MacroAssembler::li(Register dst, Handle<Object> value, LiFlags mode) {
  AllowDeferredHandleDereference smi_check;
  if (value->IsSmi()) {
    li(dst, Operand(value), mode);
  } else {
    DCHECK(value->IsHeapObject());
    if (isolate()->heap()->InNewSpace(*value)) {
      Handle<Cell> cell = isolate()->factory()->NewCell(value);
      li(dst, Operand(cell));
      ld(dst, FieldMemOperand(dst, Cell::kValueOffset));
    } else {
      li(dst, Operand(value));
    }
  }
}


void MacroAssembler::li(Register rd, Operand j, LiFlags mode) {
  DCHECK(!j.is_reg());
  BlockTrampolinePoolScope block_trampoline_pool(this);
  if (!MustUseReg(j.rmode_) && mode == OPTIMIZE_SIZE) {
    // Normal load of an immediate value which does not need Relocation Info.
    if (is_int32(j.imm64_)) {
      if (is_int16(j.imm64_)) {
        daddiu(rd, zero_reg, (j.imm64_ & kImm16Mask));
      } else if (!(j.imm64_ & kHiMask)) {
        ori(rd, zero_reg, (j.imm64_ & kImm16Mask));
      } else if (!(j.imm64_ & kImm16Mask)) {
        lui(rd, (j.imm64_ >> kLuiShift) & kImm16Mask);
      } else {
        lui(rd, (j.imm64_ >> kLuiShift) & kImm16Mask);
        ori(rd, rd, (j.imm64_ & kImm16Mask));
      }
    } else {
      lui(rd, (j.imm64_ >> 48) & kImm16Mask);
      ori(rd, rd, (j.imm64_ >> 32) & kImm16Mask);
      dsll(rd, rd, 16);
      ori(rd, rd, (j.imm64_ >> 16) & kImm16Mask);
      dsll(rd, rd, 16);
      ori(rd, rd, j.imm64_ & kImm16Mask);
    }
  } else if (MustUseReg(j.rmode_)) {
    RecordRelocInfo(j.rmode_, j.imm64_);
    lui(rd, (j.imm64_ >> 32) & kImm16Mask);
    ori(rd, rd, (j.imm64_ >> 16) & kImm16Mask);
    dsll(rd, rd, 16);
    ori(rd, rd, j.imm64_ & kImm16Mask);
  } else if (mode == ADDRESS_LOAD)  {
    // We always need the same number of instructions as we may need to patch
    // this code to load another value which may need all 4 instructions.
    lui(rd, (j.imm64_ >> 32) & kImm16Mask);
    ori(rd, rd, (j.imm64_ >> 16) & kImm16Mask);
    dsll(rd, rd, 16);
    ori(rd, rd, j.imm64_ & kImm16Mask);
  } else {
    lui(rd, (j.imm64_ >> 48) & kImm16Mask);
    ori(rd, rd, (j.imm64_ >> 32) & kImm16Mask);
    dsll(rd, rd, 16);
    ori(rd, rd, (j.imm64_ >> 16) & kImm16Mask);
    dsll(rd, rd, 16);
    ori(rd, rd, j.imm64_ & kImm16Mask);
  }
}


void MacroAssembler::MultiPush(RegList regs) {
  int16_t num_to_push = NumberOfBitsSet(regs);
  int16_t stack_offset = num_to_push * kPointerSize;

  Dsubu(sp, sp, Operand(stack_offset));
  for (int16_t i = kNumRegisters - 1; i >= 0; i--) {
    if ((regs & (1 << i)) != 0) {
      stack_offset -= kPointerSize;
      sd(ToRegister(i), MemOperand(sp, stack_offset));
    }
  }
}


void MacroAssembler::MultiPushReversed(RegList regs) {
  int16_t num_to_push = NumberOfBitsSet(regs);
  int16_t stack_offset = num_to_push * kPointerSize;

  Dsubu(sp, sp, Operand(stack_offset));
  for (int16_t i = 0; i < kNumRegisters; i++) {
    if ((regs & (1 << i)) != 0) {
      stack_offset -= kPointerSize;
      sd(ToRegister(i), MemOperand(sp, stack_offset));
    }
  }
}


void MacroAssembler::MultiPop(RegList regs) {
  int16_t stack_offset = 0;

  for (int16_t i = 0; i < kNumRegisters; i++) {
    if ((regs & (1 << i)) != 0) {
      ld(ToRegister(i), MemOperand(sp, stack_offset));
      stack_offset += kPointerSize;
    }
  }
  daddiu(sp, sp, stack_offset);
}


void MacroAssembler::MultiPopReversed(RegList regs) {
  int16_t stack_offset = 0;

  for (int16_t i = kNumRegisters - 1; i >= 0; i--) {
    if ((regs & (1 << i)) != 0) {
      ld(ToRegister(i), MemOperand(sp, stack_offset));
      stack_offset += kPointerSize;
    }
  }
  daddiu(sp, sp, stack_offset);
}


void MacroAssembler::MultiPushFPU(RegList regs) {
  int16_t num_to_push = NumberOfBitsSet(regs);
  int16_t stack_offset = num_to_push * kDoubleSize;

  Dsubu(sp, sp, Operand(stack_offset));
  for (int16_t i = kNumRegisters - 1; i >= 0; i--) {
    if ((regs & (1 << i)) != 0) {
      stack_offset -= kDoubleSize;
      sdc1(FPURegister::from_code(i), MemOperand(sp, stack_offset));
    }
  }
}


void MacroAssembler::MultiPushReversedFPU(RegList regs) {
  int16_t num_to_push = NumberOfBitsSet(regs);
  int16_t stack_offset = num_to_push * kDoubleSize;

  Dsubu(sp, sp, Operand(stack_offset));
  for (int16_t i = 0; i < kNumRegisters; i++) {
    if ((regs & (1 << i)) != 0) {
      stack_offset -= kDoubleSize;
      sdc1(FPURegister::from_code(i), MemOperand(sp, stack_offset));
    }
  }
}


void MacroAssembler::MultiPopFPU(RegList regs) {
  int16_t stack_offset = 0;

  for (int16_t i = 0; i < kNumRegisters; i++) {
    if ((regs & (1 << i)) != 0) {
      ldc1(FPURegister::from_code(i), MemOperand(sp, stack_offset));
      stack_offset += kDoubleSize;
    }
  }
  daddiu(sp, sp, stack_offset);
}


void MacroAssembler::MultiPopReversedFPU(RegList regs) {
  int16_t stack_offset = 0;

  for (int16_t i = kNumRegisters - 1; i >= 0; i--) {
    if ((regs & (1 << i)) != 0) {
      ldc1(FPURegister::from_code(i), MemOperand(sp, stack_offset));
      stack_offset += kDoubleSize;
    }
  }
  daddiu(sp, sp, stack_offset);
}


void MacroAssembler::FlushICache(Register address, unsigned instructions) {
  RegList saved_regs = kJSCallerSaved | ra.bit();
  MultiPush(saved_regs);
  AllowExternalCallThatCantCauseGC scope(this);

  // Save to a0 in case address == a4.
  Move(a0, address);
  PrepareCallCFunction(2, a4);

  li(a1, instructions * kInstrSize);
  CallCFunction(ExternalReference::flush_icache_function(isolate()), 2);
  MultiPop(saved_regs);
}


void MacroAssembler::Ext(Register rt,
                         Register rs,
                         uint16_t pos,
                         uint16_t size) {
  DCHECK(pos < 32);
  DCHECK(pos + size < 33);
  ext_(rt, rs, pos, size);
}


void MacroAssembler::Dext(Register rt, Register rs, uint16_t pos,
                          uint16_t size) {
  DCHECK(pos < 32);
  DCHECK(pos + size < 33);
  dext_(rt, rs, pos, size);
}


void MacroAssembler::Ins(Register rt,
                         Register rs,
                         uint16_t pos,
                         uint16_t size) {
  DCHECK(pos < 32);
  DCHECK(pos + size <= 32);
  DCHECK(size != 0);
  ins_(rt, rs, pos, size);
}


void MacroAssembler::Cvt_d_uw(FPURegister fd,
                              FPURegister fs,
                              FPURegister scratch) {
  // Move the data from fs to t8.
  mfc1(t8, fs);
  Cvt_d_uw(fd, t8, scratch);
}


void MacroAssembler::Cvt_d_uw(FPURegister fd,
                              Register rs,
                              FPURegister scratch) {
  // Convert rs to a FP value in fd (and fd + 1).
  // We do this by converting rs minus the MSB to avoid sign conversion,
  // then adding 2^31 to the result (if needed).

  DCHECK(!fd.is(scratch));
  DCHECK(!rs.is(t9));
  DCHECK(!rs.is(at));

  // Save rs's MSB to t9.
  Ext(t9, rs, 31, 1);
  // Remove rs's MSB.
  Ext(at, rs, 0, 31);
  // Move the result to fd.
  mtc1(at, fd);
  mthc1(zero_reg, fd);

  // Convert fd to a real FP value.
  cvt_d_w(fd, fd);

  Label conversion_done;

  // If rs's MSB was 0, it's done.
  // Otherwise we need to add that to the FP register.
  Branch(&conversion_done, eq, t9, Operand(zero_reg));

  // Load 2^31 into f20 as its float representation.
  li(at, 0x41E00000);
  mtc1(zero_reg, scratch);
  mthc1(at, scratch);
  // Add it to fd.
  add_d(fd, fd, scratch);

  bind(&conversion_done);
}


void MacroAssembler::Round_l_d(FPURegister fd, FPURegister fs) {
  round_l_d(fd, fs);
}


void MacroAssembler::Floor_l_d(FPURegister fd, FPURegister fs) {
  floor_l_d(fd, fs);
}


void MacroAssembler::Ceil_l_d(FPURegister fd, FPURegister fs) {
  ceil_l_d(fd, fs);
}


void MacroAssembler::Trunc_l_d(FPURegister fd, FPURegister fs) {
  trunc_l_d(fd, fs);
}


void MacroAssembler::Trunc_l_ud(FPURegister fd,
                                FPURegister fs,
                                FPURegister scratch) {
  // Load to GPR.
  dmfc1(t8, fs);
  // Reset sign bit.
  li(at, 0x7fffffffffffffff);
  and_(t8, t8, at);
  dmtc1(t8, fs);
  trunc_l_d(fd, fs);
}


void MacroAssembler::Trunc_uw_d(FPURegister fd,
                                FPURegister fs,
                                FPURegister scratch) {
  Trunc_uw_d(fs, t8, scratch);
  mtc1(t8, fd);
}


void MacroAssembler::Trunc_w_d(FPURegister fd, FPURegister fs) {
  trunc_w_d(fd, fs);
}


void MacroAssembler::Round_w_d(FPURegister fd, FPURegister fs) {
  round_w_d(fd, fs);
}


void MacroAssembler::Floor_w_d(FPURegister fd, FPURegister fs) {
  floor_w_d(fd, fs);
}


void MacroAssembler::Ceil_w_d(FPURegister fd, FPURegister fs) {
  ceil_w_d(fd, fs);
}


void MacroAssembler::Trunc_uw_d(FPURegister fd,
                                Register rs,
                                FPURegister scratch) {
  DCHECK(!fd.is(scratch));
  DCHECK(!rs.is(at));

  // Load 2^31 into scratch as its float representation.
  li(at, 0x41E00000);
  mtc1(zero_reg, scratch);
  mthc1(at, scratch);
  // Test if scratch > fd.
  // If fd < 2^31 we can convert it normally.
  Label simple_convert;
  BranchF(&simple_convert, NULL, lt, fd, scratch);

  // First we subtract 2^31 from fd, then trunc it to rs
  // and add 2^31 to rs.
  sub_d(scratch, fd, scratch);
  trunc_w_d(scratch, scratch);
  mfc1(rs, scratch);
  Or(rs, rs, 1 << 31);

  Label done;
  Branch(&done);
  // Simple conversion.
  bind(&simple_convert);
  trunc_w_d(scratch, fd);
  mfc1(rs, scratch);

  bind(&done);
}


void MacroAssembler::Madd_d(FPURegister fd, FPURegister fr, FPURegister fs,
    FPURegister ft, FPURegister scratch) {
  if (0) {  // TODO(plind): find reasonable arch-variant symbol names.
    madd_d(fd, fr, fs, ft);
  } else {
    // Can not change source regs's value.
    DCHECK(!fr.is(scratch) && !fs.is(scratch) && !ft.is(scratch));
    mul_d(scratch, fs, ft);
    add_d(fd, fr, scratch);
  }
}


void MacroAssembler::BranchF(Label* target,
                             Label* nan,
                             Condition cc,
                             FPURegister cmp1,
                             FPURegister cmp2,
                             BranchDelaySlot bd) {
  BlockTrampolinePoolScope block_trampoline_pool(this);
  if (cc == al) {
    Branch(bd, target);
    return;
  }

  DCHECK(nan || target);
  // Check for unordered (NaN) cases.
  if (nan) {
    if (kArchVariant != kMips64r6) {
      c(UN, D, cmp1, cmp2);
      bc1t(nan);
    } else {
      // Use f31 for comparison result. It has to be unavailable to lithium
      // register allocator.
      DCHECK(!cmp1.is(f31) && !cmp2.is(f31));
      cmp(UN, L, f31, cmp1, cmp2);
      bc1nez(nan, f31);
    }
  }

  if (kArchVariant != kMips64r6) {
    if (target) {
      // Here NaN cases were either handled by this function or are assumed to
      // have been handled by the caller.
      switch (cc) {
        case lt:
          c(OLT, D, cmp1, cmp2);
          bc1t(target);
          break;
        case gt:
          c(ULE, D, cmp1, cmp2);
          bc1f(target);
          break;
        case ge:
          c(ULT, D, cmp1, cmp2);
          bc1f(target);
          break;
        case le:
          c(OLE, D, cmp1, cmp2);
          bc1t(target);
          break;
        case eq:
          c(EQ, D, cmp1, cmp2);
          bc1t(target);
          break;
        case ueq:
          c(UEQ, D, cmp1, cmp2);
          bc1t(target);
          break;
        case ne:
          c(EQ, D, cmp1, cmp2);
          bc1f(target);
          break;
        case nue:
          c(UEQ, D, cmp1, cmp2);
          bc1f(target);
          break;
        default:
          CHECK(0);
      }
    }
  } else {
    if (target) {
      // Here NaN cases were either handled by this function or are assumed to
      // have been handled by the caller.
      // Unsigned conditions are treated as their signed counterpart.
      // Use f31 for comparison result, it is valid in fp64 (FR = 1) mode.
      DCHECK(!cmp1.is(f31) && !cmp2.is(f31));
      switch (cc) {
        case lt:
          cmp(OLT, L, f31, cmp1, cmp2);
          bc1nez(target, f31);
          break;
        case gt:
          cmp(ULE, L, f31, cmp1, cmp2);
          bc1eqz(target, f31);
          break;
        case ge:
          cmp(ULT, L, f31, cmp1, cmp2);
          bc1eqz(target, f31);
          break;
        case le:
          cmp(OLE, L, f31, cmp1, cmp2);
          bc1nez(target, f31);
          break;
        case eq:
          cmp(EQ, L, f31, cmp1, cmp2);
          bc1nez(target, f31);
          break;
        case ueq:
          cmp(UEQ, L, f31, cmp1, cmp2);
          bc1nez(target, f31);
          break;
        case ne:
          cmp(EQ, L, f31, cmp1, cmp2);
          bc1eqz(target, f31);
          break;
        case nue:
          cmp(UEQ, L, f31, cmp1, cmp2);
          bc1eqz(target, f31);
          break;
        default:
          CHECK(0);
      }
    }
  }

  if (bd == PROTECT) {
    nop();
  }
}


void MacroAssembler::Move(FPURegister dst, float imm) {
  li(at, Operand(bit_cast<int32_t>(imm)));
  mtc1(at, dst);
}


void MacroAssembler::Move(FPURegister dst, double imm) {
  static const DoubleRepresentation minus_zero(-0.0);
  static const DoubleRepresentation zero(0.0);
  DoubleRepresentation value_rep(imm);
  // Handle special values first.
  if (value_rep == zero && has_double_zero_reg_set_) {
    mov_d(dst, kDoubleRegZero);
  } else if (value_rep == minus_zero && has_double_zero_reg_set_) {
    neg_d(dst, kDoubleRegZero);
  } else {
    uint32_t lo, hi;
    DoubleAsTwoUInt32(imm, &lo, &hi);
    // Move the low part of the double into the lower bits of the corresponding
    // FPU register.
    if (lo != 0) {
      li(at, Operand(lo));
      mtc1(at, dst);
    } else {
      mtc1(zero_reg, dst);
    }
    // Move the high part of the double into the high bits of the corresponding
    // FPU register.
    if (hi != 0) {
      li(at, Operand(hi));
      mthc1(at, dst);
    } else {
      mthc1(zero_reg, dst);
    }
    if (dst.is(kDoubleRegZero)) has_double_zero_reg_set_ = true;
  }
}


void MacroAssembler::Movz(Register rd, Register rs, Register rt) {
  if (kArchVariant == kMips64r6) {
    Label done;
    Branch(&done, ne, rt, Operand(zero_reg));
    mov(rd, rs);
    bind(&done);
  } else {
    movz(rd, rs, rt);
  }
}


void MacroAssembler::Movn(Register rd, Register rs, Register rt) {
  if (kArchVariant == kMips64r6) {
    Label done;
    Branch(&done, eq, rt, Operand(zero_reg));
    mov(rd, rs);
    bind(&done);
  } else {
    movn(rd, rs, rt);
  }
}


void MacroAssembler::Movt(Register rd, Register rs, uint16_t cc) {
  movt(rd, rs, cc);
}


void MacroAssembler::Movf(Register rd, Register rs, uint16_t cc) {
  movf(rd, rs, cc);
}


void MacroAssembler::Clz(Register rd, Register rs) {
  clz(rd, rs);
}


void MacroAssembler::EmitFPUTruncate(FPURoundingMode rounding_mode,
                                     Register result,
                                     DoubleRegister double_input,
                                     Register scratch,
                                     DoubleRegister double_scratch,
                                     Register except_flag,
                                     CheckForInexactConversion check_inexact) {
  DCHECK(!result.is(scratch));
  DCHECK(!double_input.is(double_scratch));
  DCHECK(!except_flag.is(scratch));

  Label done;

  // Clear the except flag (0 = no exception)
  mov(except_flag, zero_reg);

  // Test for values that can be exactly represented as a signed 32-bit integer.
  cvt_w_d(double_scratch, double_input);
  mfc1(result, double_scratch);
  cvt_d_w(double_scratch, double_scratch);
  BranchF(&done, NULL, eq, double_input, double_scratch);

  int32_t except_mask = kFCSRFlagMask;  // Assume interested in all exceptions.

  if (check_inexact == kDontCheckForInexactConversion) {
    // Ignore inexact exceptions.
    except_mask &= ~kFCSRInexactFlagMask;
  }

  // Save FCSR.
  cfc1(scratch, FCSR);
  // Disable FPU exceptions.
  ctc1(zero_reg, FCSR);

  // Do operation based on rounding mode.
  switch (rounding_mode) {
    case kRoundToNearest:
      Round_w_d(double_scratch, double_input);
      break;
    case kRoundToZero:
      Trunc_w_d(double_scratch, double_input);
      break;
    case kRoundToPlusInf:
      Ceil_w_d(double_scratch, double_input);
      break;
    case kRoundToMinusInf:
      Floor_w_d(double_scratch, double_input);
      break;
  }  // End of switch-statement.

  // Retrieve FCSR.
  cfc1(except_flag, FCSR);
  // Restore FCSR.
  ctc1(scratch, FCSR);
  // Move the converted value into the result register.
  mfc1(result, double_scratch);

  // Check for fpu exceptions.
  And(except_flag, except_flag, Operand(except_mask));

  bind(&done);
}


void MacroAssembler::TryInlineTruncateDoubleToI(Register result,
                                                DoubleRegister double_input,
                                                Label* done) {
  DoubleRegister single_scratch = kLithiumScratchDouble.low();
  Register scratch = at;
  Register scratch2 = t9;

  // Clear cumulative exception flags and save the FCSR.
  cfc1(scratch2, FCSR);
  ctc1(zero_reg, FCSR);
  // Try a conversion to a signed integer.
  trunc_w_d(single_scratch, double_input);
  mfc1(result, single_scratch);
  // Retrieve and restore the FCSR.
  cfc1(scratch, FCSR);
  ctc1(scratch2, FCSR);
  // Check for overflow and NaNs.
  And(scratch,
      scratch,
      kFCSROverflowFlagMask | kFCSRUnderflowFlagMask | kFCSRInvalidOpFlagMask);
  // If we had no exceptions we are done.
  Branch(done, eq, scratch, Operand(zero_reg));
}


void MacroAssembler::TruncateDoubleToI(Register result,
                                       DoubleRegister double_input) {
  Label done;

  TryInlineTruncateDoubleToI(result, double_input, &done);

  // If we fell through then inline version didn't succeed - call stub instead.
  push(ra);
  Dsubu(sp, sp, Operand(kDoubleSize));  // Put input on stack.
  sdc1(double_input, MemOperand(sp, 0));

  DoubleToIStub stub(isolate(), sp, result, 0, true, true);
  CallStub(&stub);

  Daddu(sp, sp, Operand(kDoubleSize));
  pop(ra);

  bind(&done);
}


void MacroAssembler::TruncateHeapNumberToI(Register result, Register object) {
  Label done;
  DoubleRegister double_scratch = f12;
  DCHECK(!result.is(object));

  ldc1(double_scratch,
       MemOperand(object, HeapNumber::kValueOffset - kHeapObjectTag));
  TryInlineTruncateDoubleToI(result, double_scratch, &done);

  // If we fell through then inline version didn't succeed - call stub instead.
  push(ra);
  DoubleToIStub stub(isolate(),
                     object,
                     result,
                     HeapNumber::kValueOffset - kHeapObjectTag,
                     true,
                     true);
  CallStub(&stub);
  pop(ra);

  bind(&done);
}


void MacroAssembler::TruncateNumberToI(Register object,
                                       Register result,
                                       Register heap_number_map,
                                       Register scratch,
                                       Label* not_number) {
  Label done;
  DCHECK(!result.is(object));

  UntagAndJumpIfSmi(result, object, &done);
  JumpIfNotHeapNumber(object, heap_number_map, scratch, not_number);
  TruncateHeapNumberToI(result, object);

  bind(&done);
}


void MacroAssembler::GetLeastBitsFromSmi(Register dst,
                                         Register src,
                                         int num_least_bits) {
  // Ext(dst, src, kSmiTagSize, num_least_bits);
  SmiUntag(dst, src);
  And(dst, dst, Operand((1 << num_least_bits) - 1));
}


void MacroAssembler::GetLeastBitsFromInt32(Register dst,
                                           Register src,
                                           int num_least_bits) {
  DCHECK(!src.is(dst));
  And(dst, src, Operand((1 << num_least_bits) - 1));
}


// Emulated condtional branches do not emit a nop in the branch delay slot.
//
// BRANCH_ARGS_CHECK checks that conditional jump arguments are correct.
#define BRANCH_ARGS_CHECK(cond, rs, rt) DCHECK(                                \
    (cond == cc_always && rs.is(zero_reg) && rt.rm().is(zero_reg)) ||          \
    (cond != cc_always && (!rs.is(zero_reg) || !rt.rm().is(zero_reg))))


void MacroAssembler::Branch(int16_t offset, BranchDelaySlot bdslot) {
  BranchShort(offset, bdslot);
}


void MacroAssembler::Branch(int16_t offset, Condition cond, Register rs,
                            const Operand& rt,
                            BranchDelaySlot bdslot) {
  BranchShort(offset, cond, rs, rt, bdslot);
}


void MacroAssembler::Branch(Label* L, BranchDelaySlot bdslot) {
  if (L->is_bound()) {
    if (is_near(L)) {
      BranchShort(L, bdslot);
    } else {
      Jr(L, bdslot);
    }
  } else {
    if (is_trampoline_emitted()) {
      Jr(L, bdslot);
    } else {
      BranchShort(L, bdslot);
    }
  }
}


void MacroAssembler::Branch(Label* L, Condition cond, Register rs,
                            const Operand& rt,
                            BranchDelaySlot bdslot) {
  if (L->is_bound()) {
    if (is_near(L)) {
      BranchShort(L, cond, rs, rt, bdslot);
    } else {
      if (cond != cc_always) {
        Label skip;
        Condition neg_cond = NegateCondition(cond);
        BranchShort(&skip, neg_cond, rs, rt);
        Jr(L, bdslot);
        bind(&skip);
      } else {
        Jr(L, bdslot);
      }
    }
  } else {
    if (is_trampoline_emitted()) {
      if (cond != cc_always) {
        Label skip;
        Condition neg_cond = NegateCondition(cond);
        BranchShort(&skip, neg_cond, rs, rt);
        Jr(L, bdslot);
        bind(&skip);
      } else {
        Jr(L, bdslot);
      }
    } else {
      BranchShort(L, cond, rs, rt, bdslot);
    }
  }
}


void MacroAssembler::Branch(Label* L,
                            Condition cond,
                            Register rs,
                            Heap::RootListIndex index,
                            BranchDelaySlot bdslot) {
  LoadRoot(at, index);
  Branch(L, cond, rs, Operand(at), bdslot);
}


void MacroAssembler::BranchShort(int16_t offset, BranchDelaySlot bdslot) {
  b(offset);

  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::BranchShort(int16_t offset, Condition cond, Register rs,
                                 const Operand& rt,
                                 BranchDelaySlot bdslot) {
  BRANCH_ARGS_CHECK(cond, rs, rt);
  DCHECK(!rs.is(zero_reg));
  Register r2 = no_reg;
  Register scratch = at;

  if (rt.is_reg()) {
    // NOTE: 'at' can be clobbered by Branch but it is legal to use it as rs or
    // rt.
    BlockTrampolinePoolScope block_trampoline_pool(this);
    r2 = rt.rm_;
    switch (cond) {
      case cc_always:
        b(offset);
        break;
      case eq:
        beq(rs, r2, offset);
        break;
      case ne:
        bne(rs, r2, offset);
        break;
      // Signed comparison.
      case greater:
        if (r2.is(zero_reg)) {
          bgtz(rs, offset);
        } else {
          slt(scratch, r2, rs);
          bne(scratch, zero_reg, offset);
        }
        break;
      case greater_equal:
        if (r2.is(zero_reg)) {
          bgez(rs, offset);
        } else {
          slt(scratch, rs, r2);
          beq(scratch, zero_reg, offset);
        }
        break;
      case less:
        if (r2.is(zero_reg)) {
          bltz(rs, offset);
        } else {
          slt(scratch, rs, r2);
          bne(scratch, zero_reg, offset);
        }
        break;
      case less_equal:
        if (r2.is(zero_reg)) {
          blez(rs, offset);
        } else {
          slt(scratch, r2, rs);
          beq(scratch, zero_reg, offset);
        }
        break;
      // Unsigned comparison.
      case Ugreater:
        if (r2.is(zero_reg)) {
          bgtz(rs, offset);
        } else {
          sltu(scratch, r2, rs);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Ugreater_equal:
        if (r2.is(zero_reg)) {
          bgez(rs, offset);
        } else {
          sltu(scratch, rs, r2);
          beq(scratch, zero_reg, offset);
        }
        break;
      case Uless:
        if (r2.is(zero_reg)) {
          // No code needs to be emitted.
          return;
        } else {
          sltu(scratch, rs, r2);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Uless_equal:
        if (r2.is(zero_reg)) {
          b(offset);
        } else {
          sltu(scratch, r2, rs);
          beq(scratch, zero_reg, offset);
        }
        break;
      default:
        UNREACHABLE();
    }
  } else {
    // Be careful to always use shifted_branch_offset only just before the
    // branch instruction, as the location will be remember for patching the
    // target.
    BlockTrampolinePoolScope block_trampoline_pool(this);
    switch (cond) {
      case cc_always:
        b(offset);
        break;
      case eq:
        if (rt.imm64_ == 0) {
          beq(rs, zero_reg, offset);
        } else {
          // We don't want any other register but scratch clobbered.
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          beq(rs, r2, offset);
        }
        break;
      case ne:
        if (rt.imm64_ == 0) {
          bne(rs, zero_reg, offset);
        } else {
          // We don't want any other register but scratch clobbered.
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          bne(rs, r2, offset);
        }
        break;
      // Signed comparison.
      case greater:
        if (rt.imm64_ == 0) {
          bgtz(rs, offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          slt(scratch, r2, rs);
          bne(scratch, zero_reg, offset);
        }
        break;
      case greater_equal:
        if (rt.imm64_ == 0) {
          bgez(rs, offset);
        } else if (is_int16(rt.imm64_)) {
          slti(scratch, rs, rt.imm64_);
          beq(scratch, zero_reg, offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          slt(scratch, rs, r2);
          beq(scratch, zero_reg, offset);
        }
        break;
      case less:
        if (rt.imm64_ == 0) {
          bltz(rs, offset);
        } else if (is_int16(rt.imm64_)) {
          slti(scratch, rs, rt.imm64_);
          bne(scratch, zero_reg, offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          slt(scratch, rs, r2);
          bne(scratch, zero_reg, offset);
        }
        break;
      case less_equal:
        if (rt.imm64_ == 0) {
          blez(rs, offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          slt(scratch, r2, rs);
          beq(scratch, zero_reg, offset);
       }
       break;
      // Unsigned comparison.
      case Ugreater:
        if (rt.imm64_ == 0) {
          bgtz(rs, offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, r2, rs);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Ugreater_equal:
        if (rt.imm64_ == 0) {
          bgez(rs, offset);
        } else if (is_int16(rt.imm64_)) {
          sltiu(scratch, rs, rt.imm64_);
          beq(scratch, zero_reg, offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, rs, r2);
          beq(scratch, zero_reg, offset);
        }
        break;
      case Uless:
        if (rt.imm64_ == 0) {
          // No code needs to be emitted.
          return;
        } else if (is_int16(rt.imm64_)) {
          sltiu(scratch, rs, rt.imm64_);
          bne(scratch, zero_reg, offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, rs, r2);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Uless_equal:
        if (rt.imm64_ == 0) {
          b(offset);
        } else {
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, r2, rs);
          beq(scratch, zero_reg, offset);
        }
        break;
      default:
        UNREACHABLE();
    }
  }
  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::BranchShort(Label* L, BranchDelaySlot bdslot) {
  // We use branch_offset as an argument for the branch instructions to be sure
  // it is called just before generating the branch instruction, as needed.

  b(shifted_branch_offset(L, false));

  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::BranchShort(Label* L, Condition cond, Register rs,
                                 const Operand& rt,
                                 BranchDelaySlot bdslot) {
  BRANCH_ARGS_CHECK(cond, rs, rt);

  int32_t offset = 0;
  Register r2 = no_reg;
  Register scratch = at;
  if (rt.is_reg()) {
    BlockTrampolinePoolScope block_trampoline_pool(this);
    r2 = rt.rm_;
    // Be careful to always use shifted_branch_offset only just before the
    // branch instruction, as the location will be remember for patching the
    // target.
    switch (cond) {
      case cc_always:
        offset = shifted_branch_offset(L, false);
        b(offset);
        break;
      case eq:
        offset = shifted_branch_offset(L, false);
        beq(rs, r2, offset);
        break;
      case ne:
        offset = shifted_branch_offset(L, false);
        bne(rs, r2, offset);
        break;
      // Signed comparison.
      case greater:
        if (r2.is(zero_reg)) {
          offset = shifted_branch_offset(L, false);
          bgtz(rs, offset);
        } else {
          slt(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case greater_equal:
        if (r2.is(zero_reg)) {
          offset = shifted_branch_offset(L, false);
          bgez(rs, offset);
        } else {
          slt(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
      case less:
        if (r2.is(zero_reg)) {
          offset = shifted_branch_offset(L, false);
          bltz(rs, offset);
        } else {
          slt(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case less_equal:
        if (r2.is(zero_reg)) {
          offset = shifted_branch_offset(L, false);
          blez(rs, offset);
        } else {
          slt(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
      // Unsigned comparison.
      case Ugreater:
        if (r2.is(zero_reg)) {
          offset = shifted_branch_offset(L, false);
           bgtz(rs, offset);
        } else {
          sltu(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Ugreater_equal:
        if (r2.is(zero_reg)) {
          offset = shifted_branch_offset(L, false);
          bgez(rs, offset);
        } else {
          sltu(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
      case Uless:
        if (r2.is(zero_reg)) {
          // No code needs to be emitted.
          return;
        } else {
          sltu(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Uless_equal:
        if (r2.is(zero_reg)) {
          offset = shifted_branch_offset(L, false);
          b(offset);
        } else {
          sltu(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
      default:
        UNREACHABLE();
    }
  } else {
    // Be careful to always use shifted_branch_offset only just before the
    // branch instruction, as the location will be remember for patching the
    // target.
    BlockTrampolinePoolScope block_trampoline_pool(this);
    switch (cond) {
      case cc_always:
        offset = shifted_branch_offset(L, false);
        b(offset);
        break;
      case eq:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          beq(rs, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          offset = shifted_branch_offset(L, false);
          beq(rs, r2, offset);
        }
        break;
      case ne:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          bne(rs, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          offset = shifted_branch_offset(L, false);
          bne(rs, r2, offset);
        }
        break;
      // Signed comparison.
      case greater:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          bgtz(rs, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          slt(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case greater_equal:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          bgez(rs, offset);
        } else if (is_int16(rt.imm64_)) {
          slti(scratch, rs, rt.imm64_);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          slt(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
      case less:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          bltz(rs, offset);
        } else if (is_int16(rt.imm64_)) {
          slti(scratch, rs, rt.imm64_);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          slt(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case less_equal:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          blez(rs, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          slt(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
      // Unsigned comparison.
      case Ugreater:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          bne(rs, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Ugreater_equal:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          bgez(rs, offset);
        } else if (is_int16(rt.imm64_)) {
          sltiu(scratch, rs, rt.imm64_);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
     case Uless:
        if (rt.imm64_ == 0) {
          // No code needs to be emitted.
          return;
        } else if (is_int16(rt.imm64_)) {
          sltiu(scratch, rs, rt.imm64_);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, rs, r2);
          offset = shifted_branch_offset(L, false);
          bne(scratch, zero_reg, offset);
        }
        break;
      case Uless_equal:
        if (rt.imm64_ == 0) {
          offset = shifted_branch_offset(L, false);
          beq(rs, zero_reg, offset);
        } else {
          DCHECK(!scratch.is(rs));
          r2 = scratch;
          li(r2, rt);
          sltu(scratch, r2, rs);
          offset = shifted_branch_offset(L, false);
          beq(scratch, zero_reg, offset);
        }
        break;
      default:
        UNREACHABLE();
    }
  }
  // Check that offset could actually hold on an int16_t.
  DCHECK(is_int16(offset));
  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::BranchAndLink(int16_t offset, BranchDelaySlot bdslot) {
  BranchAndLinkShort(offset, bdslot);
}


void MacroAssembler::BranchAndLink(int16_t offset, Condition cond, Register rs,
                                   const Operand& rt,
                                   BranchDelaySlot bdslot) {
  BranchAndLinkShort(offset, cond, rs, rt, bdslot);
}


void MacroAssembler::BranchAndLink(Label* L, BranchDelaySlot bdslot) {
  if (L->is_bound()) {
    if (is_near(L)) {
      BranchAndLinkShort(L, bdslot);
    } else {
      Jalr(L, bdslot);
    }
  } else {
    if (is_trampoline_emitted()) {
      Jalr(L, bdslot);
    } else {
      BranchAndLinkShort(L, bdslot);
    }
  }
}


void MacroAssembler::BranchAndLink(Label* L, Condition cond, Register rs,
                                   const Operand& rt,
                                   BranchDelaySlot bdslot) {
  if (L->is_bound()) {
    if (is_near(L)) {
      BranchAndLinkShort(L, cond, rs, rt, bdslot);
    } else {
      Label skip;
      Condition neg_cond = NegateCondition(cond);
      BranchShort(&skip, neg_cond, rs, rt);
      Jalr(L, bdslot);
      bind(&skip);
    }
  } else {
    if (is_trampoline_emitted()) {
      Label skip;
      Condition neg_cond = NegateCondition(cond);
      BranchShort(&skip, neg_cond, rs, rt);
      Jalr(L, bdslot);
      bind(&skip);
    } else {
      BranchAndLinkShort(L, cond, rs, rt, bdslot);
    }
  }
}


// We need to use a bgezal or bltzal, but they can't be used directly with the
// slt instructions. We could use sub or add instead but we would miss overflow
// cases, so we keep slt and add an intermediate third instruction.
void MacroAssembler::BranchAndLinkShort(int16_t offset,
                                        BranchDelaySlot bdslot) {
  bal(offset);

  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::BranchAndLinkShort(int16_t offset, Condition cond,
                                        Register rs, const Operand& rt,
                                        BranchDelaySlot bdslot) {
  BRANCH_ARGS_CHECK(cond, rs, rt);
  Register r2 = no_reg;
  Register scratch = at;

  if (rt.is_reg()) {
    r2 = rt.rm_;
  } else if (cond != cc_always) {
    r2 = scratch;
    li(r2, rt);
  }

  {
    BlockTrampolinePoolScope block_trampoline_pool(this);
    switch (cond) {
      case cc_always:
        bal(offset);
        break;
      case eq:
        bne(rs, r2, 2);
        nop();
        bal(offset);
        break;
      case ne:
        beq(rs, r2, 2);
        nop();
        bal(offset);
        break;

      // Signed comparison.
      case greater:
        // rs > rt
        slt(scratch, r2, rs);
        beq(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;
      case greater_equal:
        // rs >= rt
        slt(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;
      case less:
        // rs < r2
        slt(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;
      case less_equal:
        // rs <= r2
        slt(scratch, r2, rs);
        bne(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;


      // Unsigned comparison.
      case Ugreater:
        // rs > rt
        sltu(scratch, r2, rs);
        beq(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;
      case Ugreater_equal:
        // rs >= rt
        sltu(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;
      case Uless:
        // rs < r2
        sltu(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;
      case Uless_equal:
        // rs <= r2
        sltu(scratch, r2, rs);
        bne(scratch, zero_reg, 2);
        nop();
        bal(offset);
        break;
      default:
        UNREACHABLE();
    }
  }
  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::BranchAndLinkShort(Label* L, BranchDelaySlot bdslot) {
  bal(shifted_branch_offset(L, false));

  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::BranchAndLinkShort(Label* L, Condition cond, Register rs,
                                        const Operand& rt,
                                        BranchDelaySlot bdslot) {
  BRANCH_ARGS_CHECK(cond, rs, rt);

  int32_t offset = 0;
  Register r2 = no_reg;
  Register scratch = at;
  if (rt.is_reg()) {
    r2 = rt.rm_;
  } else if (cond != cc_always) {
    r2 = scratch;
    li(r2, rt);
  }

  {
    BlockTrampolinePoolScope block_trampoline_pool(this);
    switch (cond) {
      case cc_always:
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case eq:
        bne(rs, r2, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case ne:
        beq(rs, r2, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;

      // Signed comparison.
      case greater:
        // rs > rt
        slt(scratch, r2, rs);
        beq(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case greater_equal:
        // rs >= rt
        slt(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case less:
        // rs < r2
        slt(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case less_equal:
        // rs <= r2
        slt(scratch, r2, rs);
        bne(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;


      // Unsigned comparison.
      case Ugreater:
        // rs > rt
        sltu(scratch, r2, rs);
        beq(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case Ugreater_equal:
        // rs >= rt
        sltu(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case Uless:
        // rs < r2
        sltu(scratch, rs, r2);
        bne(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;
      case Uless_equal:
        // rs <= r2
        sltu(scratch, r2, rs);
        bne(scratch, zero_reg, 2);
        nop();
        offset = shifted_branch_offset(L, false);
        bal(offset);
        break;

      default:
        UNREACHABLE();
    }
  }
  // Check that offset could actually hold on an int16_t.
  DCHECK(is_int16(offset));

  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::Jump(Register target,
                          Condition cond,
                          Register rs,
                          const Operand& rt,
                          BranchDelaySlot bd) {
  BlockTrampolinePoolScope block_trampoline_pool(this);
  if (cond == cc_always) {
    jr(target);
  } else {
    BRANCH_ARGS_CHECK(cond, rs, rt);
    Branch(2, NegateCondition(cond), rs, rt);
    jr(target);
  }
  // Emit a nop in the branch delay slot if required.
  if (bd == PROTECT)
    nop();
}


void MacroAssembler::Jump(intptr_t target,
                          RelocInfo::Mode rmode,
                          Condition cond,
                          Register rs,
                          const Operand& rt,
                          BranchDelaySlot bd) {
  Label skip;
  if (cond != cc_always) {
    Branch(USE_DELAY_SLOT, &skip, NegateCondition(cond), rs, rt);
  }
  // The first instruction of 'li' may be placed in the delay slot.
  // This is not an issue, t9 is expected to be clobbered anyway.
  li(t9, Operand(target, rmode));
  Jump(t9, al, zero_reg, Operand(zero_reg), bd);
  bind(&skip);
}


void MacroAssembler::Jump(Address target,
                          RelocInfo::Mode rmode,
                          Condition cond,
                          Register rs,
                          const Operand& rt,
                          BranchDelaySlot bd) {
  DCHECK(!RelocInfo::IsCodeTarget(rmode));
  Jump(reinterpret_cast<intptr_t>(target), rmode, cond, rs, rt, bd);
}


void MacroAssembler::Jump(Handle<Code> code,
                          RelocInfo::Mode rmode,
                          Condition cond,
                          Register rs,
                          const Operand& rt,
                          BranchDelaySlot bd) {
  DCHECK(RelocInfo::IsCodeTarget(rmode));
  AllowDeferredHandleDereference embedding_raw_address;
  Jump(reinterpret_cast<intptr_t>(code.location()), rmode, cond, rs, rt, bd);
}


int MacroAssembler::CallSize(Register target,
                             Condition cond,
                             Register rs,
                             const Operand& rt,
                             BranchDelaySlot bd) {
  int size = 0;

  if (cond == cc_always) {
    size += 1;
  } else {
    size += 3;
  }

  if (bd == PROTECT)
    size += 1;

  return size * kInstrSize;
}


// Note: To call gcc-compiled C code on mips, you must call thru t9.
void MacroAssembler::Call(Register target,
                          Condition cond,
                          Register rs,
                          const Operand& rt,
                          BranchDelaySlot bd) {
  BlockTrampolinePoolScope block_trampoline_pool(this);
  Label start;
  bind(&start);
  if (cond == cc_always) {
    jalr(target);
  } else {
    BRANCH_ARGS_CHECK(cond, rs, rt);
    Branch(2, NegateCondition(cond), rs, rt);
    jalr(target);
  }
  // Emit a nop in the branch delay slot if required.
  if (bd == PROTECT)
    nop();

  DCHECK_EQ(CallSize(target, cond, rs, rt, bd),
            SizeOfCodeGeneratedSince(&start));
}


int MacroAssembler::CallSize(Address target,
                             RelocInfo::Mode rmode,
                             Condition cond,
                             Register rs,
                             const Operand& rt,
                             BranchDelaySlot bd) {
  int size = CallSize(t9, cond, rs, rt, bd);
  return size + 4 * kInstrSize;
}


void MacroAssembler::Call(Address target,
                          RelocInfo::Mode rmode,
                          Condition cond,
                          Register rs,
                          const Operand& rt,
                          BranchDelaySlot bd) {
  BlockTrampolinePoolScope block_trampoline_pool(this);
  Label start;
  bind(&start);
  int64_t target_int = reinterpret_cast<int64_t>(target);
  // Must record previous source positions before the
  // li() generates a new code target.
  positions_recorder()->WriteRecordedPositions();
  li(t9, Operand(target_int, rmode), ADDRESS_LOAD);
  Call(t9, cond, rs, rt, bd);
  DCHECK_EQ(CallSize(target, rmode, cond, rs, rt, bd),
            SizeOfCodeGeneratedSince(&start));
}


int MacroAssembler::CallSize(Handle<Code> code,
                             RelocInfo::Mode rmode,
                             TypeFeedbackId ast_id,
                             Condition cond,
                             Register rs,
                             const Operand& rt,
                             BranchDelaySlot bd) {
  AllowDeferredHandleDereference using_raw_address;
  return CallSize(reinterpret_cast<Address>(code.location()),
      rmode, cond, rs, rt, bd);
}


void MacroAssembler::Call(Handle<Code> code,
                          RelocInfo::Mode rmode,
                          TypeFeedbackId ast_id,
                          Condition cond,
                          Register rs,
                          const Operand& rt,
                          BranchDelaySlot bd) {
  BlockTrampolinePoolScope block_trampoline_pool(this);
  Label start;
  bind(&start);
  DCHECK(RelocInfo::IsCodeTarget(rmode));
  if (rmode == RelocInfo::CODE_TARGET && !ast_id.IsNone()) {
    SetRecordedAstId(ast_id);
    rmode = RelocInfo::CODE_TARGET_WITH_ID;
  }
  AllowDeferredHandleDereference embedding_raw_address;
  Call(reinterpret_cast<Address>(code.location()), rmode, cond, rs, rt, bd);
  DCHECK_EQ(CallSize(code, rmode, ast_id, cond, rs, rt, bd),
            SizeOfCodeGeneratedSince(&start));
}


void MacroAssembler::Ret(Condition cond,
                         Register rs,
                         const Operand& rt,
                         BranchDelaySlot bd) {
  Jump(ra, cond, rs, rt, bd);
}


void MacroAssembler::J(Label* L, BranchDelaySlot bdslot) {
  BlockTrampolinePoolScope block_trampoline_pool(this);

  uint64_t imm28;
  imm28 = jump_address(L);
  imm28 &= kImm28Mask;
  { BlockGrowBufferScope block_buf_growth(this);
    // Buffer growth (and relocation) must be blocked for internal references
    // until associated instructions are emitted and available to be patched.
    RecordRelocInfo(RelocInfo::INTERNAL_REFERENCE);
    j(imm28);
  }
  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::Jr(Label* L, BranchDelaySlot bdslot) {
  BlockTrampolinePoolScope block_trampoline_pool(this);

  uint64_t imm64;
  imm64 = jump_address(L);
  { BlockGrowBufferScope block_buf_growth(this);
    // Buffer growth (and relocation) must be blocked for internal references
    // until associated instructions are emitted and available to be patched.
    RecordRelocInfo(RelocInfo::INTERNAL_REFERENCE);
    li(at, Operand(imm64), ADDRESS_LOAD);
  }
  jr(at);

  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::Jalr(Label* L, BranchDelaySlot bdslot) {
  BlockTrampolinePoolScope block_trampoline_pool(this);

  uint64_t imm64;
  imm64 = jump_address(L);
  { BlockGrowBufferScope block_buf_growth(this);
    // Buffer growth (and relocation) must be blocked for internal references
    // until associated instructions are emitted and available to be patched.
    RecordRelocInfo(RelocInfo::INTERNAL_REFERENCE);
    li(at, Operand(imm64), ADDRESS_LOAD);
  }
  jalr(at);

  // Emit a nop in the branch delay slot if required.
  if (bdslot == PROTECT)
    nop();
}


void MacroAssembler::DropAndRet(int drop) {
  Ret(USE_DELAY_SLOT);
  daddiu(sp, sp, drop * kPointerSize);
}

void MacroAssembler::DropAndRet(int drop,
                                Condition cond,
                                Register r1,
                                const Operand& r2) {
  // Both Drop and Ret need to be conditional.
  Label skip;
  if (cond != cc_always) {
    Branch(&skip, NegateCondition(cond), r1, r2);
  }

  Drop(drop);
  Ret();

  if (cond != cc_always) {
    bind(&skip);
  }
}


void MacroAssembler::Drop(int count,
                          Condition cond,
                          Register reg,
                          const Operand& op) {
  if (count <= 0) {
    return;
  }

  Label skip;

  if (cond != al) {
     Branch(&skip, NegateCondition(cond), reg, op);
  }

  daddiu(sp, sp, count * kPointerSize);

  if (cond != al) {
    bind(&skip);
  }
}



void MacroAssembler::Swap(Register reg1,
                          Register reg2,
                          Register scratch) {
  if (scratch.is(no_reg)) {
    Xor(reg1, reg1, Operand(reg2));
    Xor(reg2, reg2, Operand(reg1));
    Xor(reg1, reg1, Operand(reg2));
  } else {
    mov(scratch, reg1);
    mov(reg1, reg2);
    mov(reg2, scratch);
  }
}


void MacroAssembler::Call(Label* target) {
  BranchAndLink(target);
}


void MacroAssembler::Push(Handle<Object> handle) {
  li(at, Operand(handle));
  push(at);
}


void MacroAssembler::PushRegisterAsTwoSmis(Register src, Register scratch) {
  DCHECK(!src.is(scratch));
  mov(scratch, src);
  dsrl32(src, src, 0);
  dsll32(src, src, 0);
  push(src);
  dsll32(scratch, scratch, 0);
  push(scratch);
}


void MacroAssembler::PopRegisterAsTwoSmis(Register dst, Register scratch) {
  DCHECK(!dst.is(scratch));
  pop(scratch);
  dsrl32(scratch, scratch, 0);
  pop(dst);
  dsrl32(dst, dst, 0);
  dsll32(dst, dst, 0);
  or_(dst, dst, scratch);
}


void MacroAssembler::DebugBreak() {
  PrepareCEntryArgs(0);
  PrepareCEntryFunction(ExternalReference(Runtime::kDebugBreak, isolate()));
  CEntryStub ces(isolate(), 1);
  DCHECK(AllowThisStubCall(&ces));
  Call(ces.GetCode(), RelocInfo::DEBUG_BREAK);
}


// ---------------------------------------------------------------------------
// Exception handling.

void MacroAssembler::PushTryHandler(StackHandler::Kind kind,
                                    int handler_index) {
  // Adjust this code if not the case.
  STATIC_ASSERT(StackHandlerConstants::kSize == 5 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kNextOffset == 0 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kCodeOffset == 1 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kStateOffset == 2 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kContextOffset == 3 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kFPOffset == 4 * kPointerSize);

  // For the JSEntry handler, we must preserve a0-a3 and s0.
  // a5-a7 are available. We will build up the handler from the bottom by
  // pushing on the stack.
  // Set up the code object (a5) and the state (a6) for pushing.
  unsigned state =
      StackHandler::IndexField::encode(handler_index) |
      StackHandler::KindField::encode(kind);
  li(a5, Operand(CodeObject()), CONSTANT_SIZE);
  li(a6, Operand(state));

  // Push the frame pointer, context, state, and code object.
  if (kind == StackHandler::JS_ENTRY) {
    DCHECK_EQ(Smi::FromInt(0), 0);
    // The second zero_reg indicates no context.
    // The first zero_reg is the NULL frame pointer.
    // The operands are reversed to match the order of MultiPush/Pop.
    Push(zero_reg, zero_reg, a6, a5);
  } else {
    MultiPush(a5.bit() | a6.bit() | cp.bit() | fp.bit());
  }

  // Link the current handler as the next handler.
  li(a6, Operand(ExternalReference(Isolate::kHandlerAddress, isolate())));
  ld(a5, MemOperand(a6));
  push(a5);
  // Set this new handler as the current one.
  sd(sp, MemOperand(a6));
}


void MacroAssembler::PopTryHandler() {
  STATIC_ASSERT(StackHandlerConstants::kNextOffset == 0);
  pop(a1);
  Daddu(sp, sp, Operand(StackHandlerConstants::kSize - kPointerSize));
  li(at, Operand(ExternalReference(Isolate::kHandlerAddress, isolate())));
  sd(a1, MemOperand(at));
}


void MacroAssembler::JumpToHandlerEntry() {
  // Compute the handler entry address and jump to it.  The handler table is
  // a fixed array of (smi-tagged) code offsets.
  // v0 = exception, a1 = code object, a2 = state.
  ld(a3, FieldMemOperand(a1, Code::kHandlerTableOffset));
  Daddu(a3, a3, Operand(FixedArray::kHeaderSize - kHeapObjectTag));
  dsrl(a2, a2, StackHandler::kKindWidth);  // Handler index.
  dsll(a2, a2, kPointerSizeLog2);
  Daddu(a2, a3, a2);
  ld(a2, MemOperand(a2));  // Smi-tagged offset.
  Daddu(a1, a1, Operand(Code::kHeaderSize - kHeapObjectTag));  // Code start.
  dsra32(t9, a2, 0);
  Daddu(t9, t9, a1);
  Jump(t9);  // Jump.
}


void MacroAssembler::Throw(Register value) {
  // Adjust this code if not the case.
  STATIC_ASSERT(StackHandlerConstants::kSize == 5 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kNextOffset == 0);
  STATIC_ASSERT(StackHandlerConstants::kCodeOffset == 1 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kStateOffset == 2 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kContextOffset == 3 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kFPOffset == 4 * kPointerSize);

  // The exception is expected in v0.
  Move(v0, value);

  // Drop the stack pointer to the top of the top handler.
  li(a3, Operand(ExternalReference(Isolate::kHandlerAddress,
                                   isolate())));
  ld(sp, MemOperand(a3));

  // Restore the next handler.
  pop(a2);
  sd(a2, MemOperand(a3));

  // Get the code object (a1) and state (a2).  Restore the context and frame
  // pointer.
  MultiPop(a1.bit() | a2.bit() | cp.bit() | fp.bit());

  // If the handler is a JS frame, restore the context to the frame.
  // (kind == ENTRY) == (fp == 0) == (cp == 0), so we could test either fp
  // or cp.
  Label done;
  Branch(&done, eq, cp, Operand(zero_reg));
  sd(cp, MemOperand(fp, StandardFrameConstants::kContextOffset));
  bind(&done);

  JumpToHandlerEntry();
}


void MacroAssembler::ThrowUncatchable(Register value) {
  // Adjust this code if not the case.
  STATIC_ASSERT(StackHandlerConstants::kSize == 5 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kNextOffset == 0 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kCodeOffset == 1 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kStateOffset == 2 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kContextOffset == 3 * kPointerSize);
  STATIC_ASSERT(StackHandlerConstants::kFPOffset == 4 * kPointerSize);

  // The exception is expected in v0.
  if (!value.is(v0)) {
    mov(v0, value);
  }
  // Drop the stack pointer to the top of the top stack handler.
  li(a3, Operand(ExternalReference(Isolate::kHandlerAddress, isolate())));
  ld(sp, MemOperand(a3));

  // Unwind the handlers until the ENTRY handler is found.
  Label fetch_next, check_kind;
  jmp(&check_kind);
  bind(&fetch_next);
  ld(sp, MemOperand(sp, StackHandlerConstants::kNextOffset));

  bind(&check_kind);
  STATIC_ASSERT(StackHandler::JS_ENTRY == 0);
  ld(a2, MemOperand(sp, StackHandlerConstants::kStateOffset));
  And(a2, a2, Operand(StackHandler::KindField::kMask));
  Branch(&fetch_next, ne, a2, Operand(zero_reg));

  // Set the top handler address to next handler past the top ENTRY handler.
  pop(a2);
  sd(a2, MemOperand(a3));

  // Get the code object (a1) and state (a2).  Clear the context and frame
  // pointer (0 was saved in the handler).
  MultiPop(a1.bit() | a2.bit() | cp.bit() | fp.bit());

  JumpToHandlerEntry();
}


void MacroAssembler::Allocate(int object_size,
                              Register result,
                              Register scratch1,
                              Register scratch2,
                              Label* gc_required,
                              AllocationFlags flags) {
  DCHECK(object_size <= Page::kMaxRegularHeapObjectSize);
  if (!FLAG_inline_new) {
    if (emit_debug_code()) {
      // Trash the registers to simulate an allocation failure.
      li(result, 0x7091);
      li(scratch1, 0x7191);
      li(scratch2, 0x7291);
    }
    jmp(gc_required);
    return;
  }

  DCHECK(!result.is(scratch1));
  DCHECK(!result.is(scratch2));
  DCHECK(!scratch1.is(scratch2));
  DCHECK(!scratch1.is(t9));
  DCHECK(!scratch2.is(t9));
  DCHECK(!result.is(t9));

  // Make object size into bytes.
  if ((flags & SIZE_IN_WORDS) != 0) {
    object_size *= kPointerSize;
  }
  DCHECK(0 == (object_size & kObjectAlignmentMask));

  // Check relative positions of allocation top and limit addresses.
  // ARM adds additional checks to make sure the ldm instruction can be
  // used. On MIPS we don't have ldm so we don't need additional checks either.
  ExternalReference allocation_top =
      AllocationUtils::GetAllocationTopReference(isolate(), flags);
  ExternalReference allocation_limit =
      AllocationUtils::GetAllocationLimitReference(isolate(), flags);

  intptr_t top   =
      reinterpret_cast<intptr_t>(allocation_top.address());
  intptr_t limit =
      reinterpret_cast<intptr_t>(allocation_limit.address());
  DCHECK((limit - top) == kPointerSize);

  // Set up allocation top address and object size registers.
  Register topaddr = scratch1;
  li(topaddr, Operand(allocation_top));

  // This code stores a temporary value in t9.
  if ((flags & RESULT_CONTAINS_TOP) == 0) {
    // Load allocation top into result and allocation limit into t9.
    ld(result, MemOperand(topaddr));
    ld(t9, MemOperand(topaddr, kPointerSize));
  } else {
    if (emit_debug_code()) {
      // Assert that result actually contains top on entry. t9 is used
      // immediately below so this use of t9 does not cause difference with
      // respect to register content between debug and release mode.
      ld(t9, MemOperand(topaddr));
      Check(eq, kUnexpectedAllocationTop, result, Operand(t9));
    }
    // Load allocation limit into t9. Result already contains allocation top.
    ld(t9, MemOperand(topaddr, limit - top));
  }

  DCHECK(kPointerSize == kDoubleSize);
  if (emit_debug_code()) {
    And(at, result, Operand(kDoubleAlignmentMask));
    Check(eq, kAllocationIsNotDoubleAligned, at, Operand(zero_reg));
  }

  // Calculate new top and bail out if new space is exhausted. Use result
  // to calculate the new top.
  Daddu(scratch2, result, Operand(object_size));
  Branch(gc_required, Ugreater, scratch2, Operand(t9));
  sd(scratch2, MemOperand(topaddr));

  // Tag object if requested.
  if ((flags & TAG_OBJECT) != 0) {
    Daddu(result, result, Operand(kHeapObjectTag));
  }
}


void MacroAssembler::Allocate(Register object_size,
                              Register result,
                              Register scratch1,
                              Register scratch2,
                              Label* gc_required,
                              AllocationFlags flags) {
  if (!FLAG_inline_new) {
    if (emit_debug_code()) {
      // Trash the registers to simulate an allocation failure.
      li(result, 0x7091);
      li(scratch1, 0x7191);
      li(scratch2, 0x7291);
    }
    jmp(gc_required);
    return;
  }

  DCHECK(!result.is(scratch1));
  DCHECK(!result.is(scratch2));
  DCHECK(!scratch1.is(scratch2));
  DCHECK(!object_size.is(t9));
  DCHECK(!scratch1.is(t9) && !scratch2.is(t9) && !result.is(t9));

  // Check relative positions of allocation top and limit addresses.
  // ARM adds additional checks to make sure the ldm instruction can be
  // used. On MIPS we don't have ldm so we don't need additional checks either.
  ExternalReference allocation_top =
      AllocationUtils::GetAllocationTopReference(isolate(), flags);
  ExternalReference allocation_limit =
      AllocationUtils::GetAllocationLimitReference(isolate(), flags);
  intptr_t top   =
      reinterpret_cast<intptr_t>(allocation_top.address());
  intptr_t limit =
      reinterpret_cast<intptr_t>(allocation_limit.address());
  DCHECK((limit - top) == kPointerSize);

  // Set up allocation top address and object size registers.
  Register topaddr = scratch1;
  li(topaddr, Operand(allocation_top));

  // This code stores a temporary value in t9.
  if ((flags & RESULT_CONTAINS_TOP) == 0) {
    // Load allocation top into result and allocation limit into t9.
    ld(result, MemOperand(topaddr));
    ld(t9, MemOperand(topaddr, kPointerSize));
  } else {
    if (emit_debug_code()) {
      // Assert that result actually contains top on entry. t9 is used
      // immediately below so this use of t9 does not cause difference with
      // respect to register content between debug and release mode.
      ld(t9, MemOperand(topaddr));
      Check(eq, kUnexpectedAllocationTop, result, Operand(t9));
    }
    // Load allocation limit into t9. Result already contains allocation top.
    ld(t9, MemOperand(topaddr, limit - top));
  }

  DCHECK(kPointerSize == kDoubleSize);
  if (emit_debug_code()) {
    And(at, result, Operand(kDoubleAlignmentMask));
    Check(eq, kAllocationIsNotDoubleAligned, at, Operand(zero_reg));
  }

  // Calculate new top and bail out if new space is exhausted. Use result
  // to calculate the new top. Object size may be in words so a shift is
  // required to get the number of bytes.
  if ((flags & SIZE_IN_WORDS) != 0) {
    dsll(scratch2, object_size, kPointerSizeLog2);
    Daddu(scratch2, result, scratch2);
  } else {
    Daddu(scratch2, result, Operand(object_size));
  }
  Branch(gc_required, Ugreater, scratch2, Operand(t9));

  // Update allocation top. result temporarily holds the new top.
  if (emit_debug_code()) {
    And(t9, scratch2, Operand(kObjectAlignmentMask));
    Check(eq, kUnalignedAllocationInNewSpace, t9, Operand(zero_reg));
  }
  sd(scratch2, MemOperand(topaddr));

  // Tag object if requested.
  if ((flags & TAG_OBJECT) != 0) {
    Daddu(result, result, Operand(kHeapObjectTag));
  }
}


void MacroAssembler::UndoAllocationInNewSpace(Register object,
                                              Register scratch) {
  ExternalReference new_space_allocation_top =
      ExternalReference::new_space_allocation_top_address(isolate());

  // Make sure the object has no tag before resetting top.
  And(object, object, Operand(~kHeapObjectTagMask));
#ifdef DEBUG
  // Check that the object un-allocated is below the current top.
  li(scratch, Operand(new_space_allocation_top));
  ld(scratch, MemOperand(scratch));
  Check(less, kUndoAllocationOfNonAllocatedMemory,
      object, Operand(scratch));
#endif
  // Write the address of the object to un-allocate as the current top.
  li(scratch, Operand(new_space_allocation_top));
  sd(object, MemOperand(scratch));
}


void MacroAssembler::AllocateTwoByteString(Register result,
                                           Register length,
                                           Register scratch1,
                                           Register scratch2,
                                           Register scratch3,
                                           Label* gc_required) {
  // Calculate the number of bytes needed for the characters in the string while
  // observing object alignment.
  DCHECK((SeqTwoByteString::kHeaderSize & kObjectAlignmentMask) == 0);
  dsll(scratch1, length, 1);  // Length in bytes, not chars.
  daddiu(scratch1, scratch1,
       kObjectAlignmentMask + SeqTwoByteString::kHeaderSize);
  And(scratch1, scratch1, Operand(~kObjectAlignmentMask));

  // Allocate two-byte string in new space.
  Allocate(scratch1,
           result,
           scratch2,
           scratch3,
           gc_required,
           TAG_OBJECT);

  // Set the map, length and hash field.
  InitializeNewString(result,
                      length,
                      Heap::kStringMapRootIndex,
                      scratch1,
                      scratch2);
}


void MacroAssembler::AllocateOneByteString(Register result, Register length,
                                           Register scratch1, Register scratch2,
                                           Register scratch3,
                                           Label* gc_required) {
  // Calculate the number of bytes needed for the characters in the string
  // while observing object alignment.
  DCHECK((SeqOneByteString::kHeaderSize & kObjectAlignmentMask) == 0);
  DCHECK(kCharSize == 1);
  daddiu(scratch1, length,
      kObjectAlignmentMask + SeqOneByteString::kHeaderSize);
  And(scratch1, scratch1, Operand(~kObjectAlignmentMask));

  // Allocate one-byte string in new space.
  Allocate(scratch1,
           result,
           scratch2,
           scratch3,
           gc_required,
           TAG_OBJECT);

  // Set the map, length and hash field.
  InitializeNewString(result, length, Heap::kOneByteStringMapRootIndex,
                      scratch1, scratch2);
}


void MacroAssembler::AllocateTwoByteConsString(Register result,
                                               Register length,
                                               Register scratch1,
                                               Register scratch2,
                                               Label* gc_required) {
  Allocate(ConsString::kSize, result, scratch1, scratch2, gc_required,
           TAG_OBJECT);
  InitializeNewString(result,
                      length,
                      Heap::kConsStringMapRootIndex,
                      scratch1,
                      scratch2);
}


void MacroAssembler::AllocateOneByteConsString(Register result, Register length,
                                               Register scratch1,
                                               Register scratch2,
                                               Label* gc_required) {
  Allocate(ConsString::kSize,
           result,
           scratch1,
           scratch2,
           gc_required,
           TAG_OBJECT);

  InitializeNewString(result, length, Heap::kConsOneByteStringMapRootIndex,
                      scratch1, scratch2);
}


void MacroAssembler::AllocateTwoByteSlicedString(Register result,
                                                 Register length,
                                                 Register scratch1,
                                                 Register scratch2,
                                                 Label* gc_required) {
  Allocate(SlicedString::kSize, result, scratch1, scratch2, gc_required,
           TAG_OBJECT);

  InitializeNewString(result,
                      length,
                      Heap::kSlicedStringMapRootIndex,
                      scratch1,
                      scratch2);
}


void MacroAssembler::AllocateOneByteSlicedString(Register result,
                                                 Register length,
                                                 Register scratch1,
                                                 Register scratch2,
                                                 Label* gc_required) {
  Allocate(SlicedString::kSize, result, scratch1, scratch2, gc_required,
           TAG_OBJECT);

  InitializeNewString(result, length, Heap::kSlicedOneByteStringMapRootIndex,
                      scratch1, scratch2);
}


void MacroAssembler::JumpIfNotUniqueNameInstanceType(Register reg,
                                                     Label* not_unique_name) {
  STATIC_ASSERT(kInternalizedTag == 0 && kStringTag == 0);
  Label succeed;
  And(at, reg, Operand(kIsNotStringMask | kIsNotInternalizedMask));
  Branch(&succeed, eq, at, Operand(zero_reg));
  Branch(not_unique_name, ne, reg, Operand(SYMBOL_TYPE));

  bind(&succeed);
}


// Allocates a heap number or jumps to the label if the young space is full and
// a scavenge is needed.
void MacroAssembler::AllocateHeapNumber(Register result,
                                        Register scratch1,
                                        Register scratch2,
                                        Register heap_number_map,
                                        Label* need_gc,
                                        TaggingMode tagging_mode,
                                        MutableMode mode) {
  // Allocate an object in the heap for the heap number and tag it as a heap
  // object.
  Allocate(HeapNumber::kSize, result, scratch1, scratch2, need_gc,
           tagging_mode == TAG_RESULT ? TAG_OBJECT : NO_ALLOCATION_FLAGS);

  Heap::RootListIndex map_index = mode == MUTABLE
      ? Heap::kMutableHeapNumberMapRootIndex
      : Heap::kHeapNumberMapRootIndex;
  AssertIsRoot(heap_number_map, map_index);

  // Store heap number map in the allocated object.
  if (tagging_mode == TAG_RESULT) {
    sd(heap_number_map, FieldMemOperand(result, HeapObject::kMapOffset));
  } else {
    sd(heap_number_map, MemOperand(result, HeapObject::kMapOffset));
  }
}


void MacroAssembler::AllocateHeapNumberWithValue(Register result,
                                                 FPURegister value,
                                                 Register scratch1,
                                                 Register scratch2,
                                                 Label* gc_required) {
  LoadRoot(t8, Heap::kHeapNumberMapRootIndex);
  AllocateHeapNumber(result, scratch1, scratch2, t8, gc_required);
  sdc1(value, FieldMemOperand(result, HeapNumber::kValueOffset));
}


// Copies a fixed number of fields of heap objects from src to dst.
void MacroAssembler::CopyFields(Register dst,
                                Register src,
                                RegList temps,
                                int field_count) {
  DCHECK((temps & dst.bit()) == 0);
  DCHECK((temps & src.bit()) == 0);
  // Primitive implementation using only one temporary register.

  Register tmp = no_reg;
  // Find a temp register in temps list.
  for (int i = 0; i < kNumRegisters; i++) {
    if ((temps & (1 << i)) != 0) {
      tmp.code_ = i;
      break;
    }
  }
  DCHECK(!tmp.is(no_reg));

  for (int i = 0; i < field_count; i++) {
    ld(tmp, FieldMemOperand(src, i * kPointerSize));
    sd(tmp, FieldMemOperand(dst, i * kPointerSize));
  }
}


void MacroAssembler::CopyBytes(Register src,
                               Register dst,
                               Register length,
                               Register scratch) {
  Label align_loop_1, word_loop, byte_loop, byte_loop_1, done;

  // Align src before copying in word size chunks.
  Branch(&byte_loop, le, length, Operand(kPointerSize));
  bind(&align_loop_1);
  And(scratch, src, kPointerSize - 1);
  Branch(&word_loop, eq, scratch, Operand(zero_reg));
  lbu(scratch, MemOperand(src));
  Daddu(src, src, 1);
  sb(scratch, MemOperand(dst));
  Daddu(dst, dst, 1);
  Dsubu(length, length, Operand(1));
  Branch(&align_loop_1, ne, length, Operand(zero_reg));

  // Copy bytes in word size chunks.
  bind(&word_loop);
  if (emit_debug_code()) {
    And(scratch, src, kPointerSize - 1);
    Assert(eq, kExpectingAlignmentForCopyBytes,
        scratch, Operand(zero_reg));
  }
  Branch(&byte_loop, lt, length, Operand(kPointerSize));
  ld(scratch, MemOperand(src));
  Daddu(src, src, kPointerSize);

  // TODO(kalmard) check if this can be optimized to use sw in most cases.
  // Can't use unaligned access - copy byte by byte.
  sb(scratch, MemOperand(dst, 0));
  dsrl(scratch, scratch, 8);
  sb(scratch, MemOperand(dst, 1));
  dsrl(scratch, scratch, 8);
  sb(scratch, MemOperand(dst, 2));
  dsrl(scratch, scratch, 8);
  sb(scratch, MemOperand(dst, 3));
  dsrl(scratch, scratch, 8);
  sb(scratch, MemOperand(dst, 4));
  dsrl(scratch, scratch, 8);
  sb(scratch, MemOperand(dst, 5));
  dsrl(scratch, scratch, 8);
  sb(scratch, MemOperand(dst, 6));
  dsrl(scratch, scratch, 8);
  sb(scratch, MemOperand(dst, 7));
  Daddu(dst, dst, 8);

  Dsubu(length, length, Operand(kPointerSize));
  Branch(&word_loop);

  // Copy the last bytes if any left.
  bind(&byte_loop);
  Branch(&done, eq, length, Operand(zero_reg));
  bind(&byte_loop_1);
  lbu(scratch, MemOperand(src));
  Daddu(src, src, 1);
  sb(scratch, MemOperand(dst));
  Daddu(dst, dst, 1);
  Dsubu(length, length, Operand(1));
  Branch(&byte_loop_1, ne, length, Operand(zero_reg));
  bind(&done);
}


void MacroAssembler::InitializeFieldsWithFiller(Register start_offset,
                                                Register end_offset,
                                                Register filler) {
  Label loop, entry;
  Branch(&entry);
  bind(&loop);
  sd(filler, MemOperand(start_offset));
  Daddu(start_offset, start_offset, kPointerSize);
  bind(&entry);
  Branch(&loop, lt, start_offset, Operand(end_offset));
}


void MacroAssembler::CheckFastElements(Register map,
                                       Register scratch,
                                       Label* fail) {
  STATIC_ASSERT(FAST_SMI_ELEMENTS == 0);
  STATIC_ASSERT(FAST_HOLEY_SMI_ELEMENTS == 1);
  STATIC_ASSERT(FAST_ELEMENTS == 2);
  STATIC_ASSERT(FAST_HOLEY_ELEMENTS == 3);
  lbu(scratch, FieldMemOperand(map, Map::kBitField2Offset));
  Branch(fail, hi, scratch,
         Operand(Map::kMaximumBitField2FastHoleyElementValue));
}


void MacroAssembler::CheckFastObjectElements(Register map,
                                             Register scratch,
                                             Label* fail) {
  STATIC_ASSERT(FAST_SMI_ELEMENTS == 0);
  STATIC_ASSERT(FAST_HOLEY_SMI_ELEMENTS == 1);
  STATIC_ASSERT(FAST_ELEMENTS == 2);
  STATIC_ASSERT(FAST_HOLEY_ELEMENTS == 3);
  lbu(scratch, FieldMemOperand(map, Map::kBitField2Offset));
  Branch(fail, ls, scratch,
         Operand(Map::kMaximumBitField2FastHoleySmiElementValue));
  Branch(fail, hi, scratch,
         Operand(Map::kMaximumBitField2FastHoleyElementValue));
}


void MacroAssembler::CheckFastSmiElements(Register map,
                                          Register scratch,
                                          Label* fail) {
  STATIC_ASSERT(FAST_SMI_ELEMENTS == 0);
  STATIC_ASSERT(FAST_HOLEY_SMI_ELEMENTS == 1);
  lbu(scratch, FieldMemOperand(map, Map::kBitField2Offset));
  Branch(fail, hi, scratch,
         Operand(Map::kMaximumBitField2FastHoleySmiElementValue));
}


void MacroAssembler::StoreNumberToDoubleElements(Register value_reg,
                                                 Register key_reg,
                                                 Register elements_reg,
                                                 Register scratch1,
                                                 Register scratch2,
                                                 Register scratch3,
                                                 Label* fail,
                                                 int elements_offset) {
  Label smi_value, maybe_nan, have_double_value, is_nan, done;
  Register mantissa_reg = scratch2;
  Register exponent_reg = scratch3;

  // Handle smi values specially.
  JumpIfSmi(value_reg, &smi_value);

  // Ensure that the object is a heap number
  CheckMap(value_reg,
           scratch1,
           Heap::kHeapNumberMapRootIndex,
           fail,
           DONT_DO_SMI_CHECK);

  // Check for nan: all NaN values have a value greater (signed) than 0x7ff00000
  // in the exponent.
  li(scratch1, Operand(kNaNOrInfinityLowerBoundUpper32));
  lw(exponent_reg, FieldMemOperand(value_reg, HeapNumber::kExponentOffset));
  Branch(&maybe_nan, ge, exponent_reg, Operand(scratch1));

  lwu(mantissa_reg, FieldMemOperand(value_reg, HeapNumber::kMantissaOffset));

  bind(&have_double_value);
  // dsll(scratch1, key_reg, kDoubleSizeLog2 - kSmiTagSize);
  dsra(scratch1, key_reg, 32 - kDoubleSizeLog2);
  Daddu(scratch1, scratch1, elements_reg);
  sw(mantissa_reg, FieldMemOperand(
     scratch1, FixedDoubleArray::kHeaderSize - elements_offset));
  uint32_t offset = FixedDoubleArray::kHeaderSize - elements_offset +
      sizeof(kHoleNanLower32);
  sw(exponent_reg, FieldMemOperand(scratch1, offset));
  jmp(&done);

  bind(&maybe_nan);
  // Could be NaN, Infinity or -Infinity. If fraction is not zero, it's NaN,
  // otherwise it's Infinity or -Infinity, and the non-NaN code path applies.
  lw(mantissa_reg, FieldMemOperand(value_reg, HeapNumber::kMantissaOffset));
  Branch(&have_double_value, eq, mantissa_reg, Operand(zero_reg));
  bind(&is_nan);
  // Load canonical NaN for storing into the double array.
  LoadRoot(at, Heap::kNanValueRootIndex);
  lw(mantissa_reg, FieldMemOperand(at, HeapNumber::kMantissaOffset));
  lw(exponent_reg, FieldMemOperand(at, HeapNumber::kExponentOffset));
  jmp(&have_double_value);

  bind(&smi_value);
  Daddu(scratch1, elements_reg,
      Operand(FixedDoubleArray::kHeaderSize - kHeapObjectTag -
              elements_offset));
  // dsll(scratch2, key_reg, kDoubleSizeLog2 - kSmiTagSize);
  dsra(scratch2, key_reg, 32 - kDoubleSizeLog2);
  Daddu(scratch1, scratch1, scratch2);
  // scratch1 is now effective address of the double element

  Register untagged_value = elements_reg;
  SmiUntag(untagged_value, value_reg);
  mtc1(untagged_value, f2);
  cvt_d_w(f0, f2);
  sdc1(f0, MemOperand(scratch1, 0));
  bind(&done);
}


void MacroAssembler::CompareMapAndBranch(Register obj,
                                         Register scratch,
                                         Handle<Map> map,
                                         Label* early_success,
                                         Condition cond,
                                         Label* branch_to) {
  ld(scratch, FieldMemOperand(obj, HeapObject::kMapOffset));
  CompareMapAndBranch(scratch, map, early_success, cond, branch_to);
}


void MacroAssembler::CompareMapAndBranch(Register obj_map,
                                         Handle<Map> map,
                                         Label* early_success,
                                         Condition cond,
                                         Label* branch_to) {
  Branch(branch_to, cond, obj_map, Operand(map));
}


void MacroAssembler::CheckMap(Register obj,
                              Register scratch,
                              Handle<Map> map,
                              Label* fail,
                              SmiCheckType smi_check_type) {
  if (smi_check_type == DO_SMI_CHECK) {
    JumpIfSmi(obj, fail);
  }
  Label success;
  CompareMapAndBranch(obj, scratch, map, &success, ne, fail);
  bind(&success);
}


void MacroAssembler::DispatchWeakMap(Register obj, Register scratch1,
                                     Register scratch2, Handle<WeakCell> cell,
                                     Handle<Code> success,
                                     SmiCheckType smi_check_type) {
  Label fail;
  if (smi_check_type == DO_SMI_CHECK) {
    JumpIfSmi(obj, &fail);
  }
  ld(scratch1, FieldMemOperand(obj, HeapObject::kMapOffset));
  GetWeakValue(scratch2, cell);
  Jump(success, RelocInfo::CODE_TARGET, eq, scratch1, Operand(scratch2));
  bind(&fail);
}


void MacroAssembler::CheckMap(Register obj,
                              Register scratch,
                              Heap::RootListIndex index,
                              Label* fail,
                              SmiCheckType smi_check_type) {
  if (smi_check_type == DO_SMI_CHECK) {
    JumpIfSmi(obj, fail);
  }
  ld(scratch, FieldMemOperand(obj, HeapObject::kMapOffset));
  LoadRoot(at, index);
  Branch(fail, ne, scratch, Operand(at));
}


void MacroAssembler::GetWeakValue(Register value, Handle<WeakCell> cell) {
  li(value, Operand(cell));
  ld(value, FieldMemOperand(value, WeakCell::kValueOffset));
}


void MacroAssembler::LoadWeakValue(Register value, Handle<WeakCell> cell,
                                   Label* miss) {
  GetWeakValue(value, cell);
  JumpIfSmi(value, miss);
}


void MacroAssembler::MovFromFloatResult(const DoubleRegister dst) {
  if (IsMipsSoftFloatABI) {
    Move(dst, v0, v1);
  } else {
    Move(dst, f0);  // Reg f0 is o32 ABI FP return value.
  }
}


void MacroAssembler::MovFromFloatParameter(const DoubleRegister dst) {
  if (IsMipsSoftFloatABI) {
    Move(dst, a0, a1);
  } else {
    Move(dst, f12);  // Reg f12 is o32 ABI FP first argument value.
  }
}


void MacroAssembler::MovToFloatParameter(DoubleRegister src) {
  if (!IsMipsSoftFloatABI) {
    Move(f12, src);
  } else {
    Move(a0, a1, src);
  }
}


void MacroAssembler::MovToFloatResult(DoubleRegister src) {
  if (!IsMipsSoftFloatABI) {
    Move(f0, src);
  } else {
    Move(v0, v1, src);
  }
}


void MacroAssembler::MovToFloatParameters(DoubleRegister src1,
                                          DoubleRegister src2) {
  if (!IsMipsSoftFloatABI) {
    const DoubleRegister fparg2 = (kMipsAbi == kN64) ? f13 : f14;
    if (src2.is(f12)) {
      DCHECK(!src1.is(fparg2));
      Move(fparg2, src2);
      Move(f12, src1);
    } else {
      Move(f12, src1);
      Move(fparg2, src2);
    }
  } else {
    Move(a0, a1, src1);
    Move(a2, a3, src2);
  }
}


// -----------------------------------------------------------------------------
// JavaScript invokes.

void MacroAssembler::InvokePrologue(const ParameterCount& expected,
                                    const ParameterCount& actual,
                                    Handle<Code> code_constant,
                                    Register code_reg,
                                    Label* done,
                                    bool* definitely_mismatches,
                                    InvokeFlag flag,
                                    const CallWrapper& call_wrapper) {
  bool definitely_matches = false;
  *definitely_mismatches = false;
  Label regular_invoke;

  // Check whether the expected and actual arguments count match. If not,
  // setup registers according to contract with ArgumentsAdaptorTrampoline:
  //  a0: actual arguments count
  //  a1: function (passed through to callee)
  //  a2: expected arguments count

  // The code below is made a lot easier because the calling code already sets
  // up actual and expected registers according to the contract if values are
  // passed in registers.
  DCHECK(actual.is_immediate() || actual.reg().is(a0));
  DCHECK(expected.is_immediate() || expected.reg().is(a2));
  DCHECK((!code_constant.is_null() && code_reg.is(no_reg)) || code_reg.is(a3));

  if (expected.is_immediate()) {
    DCHECK(actual.is_immediate());
    if (expected.immediate() == actual.immediate()) {
      definitely_matches = true;
    } else {
      li(a0, Operand(actual.immediate()));
      const int sentinel = SharedFunctionInfo::kDontAdaptArgumentsSentinel;
      if (expected.immediate() == sentinel) {
        // Don't worry about adapting arguments for builtins that
        // don't want that done. Skip adaption code by making it look
        // like we have a match between expected and actual number of
        // arguments.
        definitely_matches = true;
      } else {
        *definitely_mismatches = true;
        li(a2, Operand(expected.immediate()));
      }
    }
  } else if (actual.is_immediate()) {
    Branch(&regular_invoke, eq, expected.reg(), Operand(actual.immediate()));
    li(a0, Operand(actual.immediate()));
  } else {
    Branch(&regular_invoke, eq, expected.reg(), Operand(actual.reg()));
  }

  if (!definitely_matches) {
    if (!code_constant.is_null()) {
      li(a3, Operand(code_constant));
      daddiu(a3, a3, Code::kHeaderSize - kHeapObjectTag);
    }

    Handle<Code> adaptor =
        isolate()->builtins()->ArgumentsAdaptorTrampoline();
    if (flag == CALL_FUNCTION) {
      call_wrapper.BeforeCall(CallSize(adaptor));
      Call(adaptor);
      call_wrapper.AfterCall();
      if (!*definitely_mismatches) {
        Branch(done);
      }
    } else {
      Jump(adaptor, RelocInfo::CODE_TARGET);
    }
    bind(&regular_invoke);
  }
}


void MacroAssembler::InvokeCode(Register code,
                                const ParameterCount& expected,
                                const ParameterCount& actual,
                                InvokeFlag flag,
                                const CallWrapper& call_wrapper) {
  // You can't call a function without a valid frame.
  DCHECK(flag == JUMP_FUNCTION || has_frame());

  Label done;

  bool definitely_mismatches = false;
  InvokePrologue(expected, actual, Handle<Code>::null(), code,
                 &done, &definitely_mismatches, flag,
                 call_wrapper);
  if (!definitely_mismatches) {
    if (flag == CALL_FUNCTION) {
      call_wrapper.BeforeCall(CallSize(code));
      Call(code);
      call_wrapper.AfterCall();
    } else {
      DCHECK(flag == JUMP_FUNCTION);
      Jump(code);
    }
    // Continue here if InvokePrologue does handle the invocation due to
    // mismatched parameter counts.
    bind(&done);
  }
}


void MacroAssembler::InvokeFunction(Register function,
                                    const ParameterCount& actual,
                                    InvokeFlag flag,
                                    const CallWrapper& call_wrapper) {
  // You can't call a function without a valid frame.
  DCHECK(flag == JUMP_FUNCTION || has_frame());

  // Contract with called JS functions requires that function is passed in a1.
  DCHECK(function.is(a1));
  Register expected_reg = a2;
  Register code_reg = a3;
  ld(code_reg, FieldMemOperand(a1, JSFunction::kSharedFunctionInfoOffset));
  ld(cp, FieldMemOperand(a1, JSFunction::kContextOffset));
  // The argument count is stored as int32_t on 64-bit platforms.
  // TODO(plind): Smi on 32-bit platforms.
  lw(expected_reg,
      FieldMemOperand(code_reg,
                      SharedFunctionInfo::kFormalParameterCountOffset));
  ld(code_reg, FieldMemOperand(a1, JSFunction::kCodeEntryOffset));
  ParameterCount expected(expected_reg);
  InvokeCode(code_reg, expected, actual, flag, call_wrapper);
}


void MacroAssembler::InvokeFunction(Register function,
                                    const ParameterCount& expected,
                                    const ParameterCount& actual,
                                    InvokeFlag flag,
                                    const CallWrapper& call_wrapper) {
  // You can't call a function without a valid frame.
  DCHECK(flag == JUMP_FUNCTION || has_frame());

  // Contract with called JS functions requires that function is passed in a1.
  DCHECK(function.is(a1));

  // Get the function and setup the context.
  ld(cp, FieldMemOperand(a1, JSFunction::kContextOffset));

  // We call indirectly through the code field in the function to
  // allow recompilation to take effect without changing any of the
  // call sites.
  ld(a3, FieldMemOperand(a1, JSFunction::kCodeEntryOffset));
  InvokeCode(a3, expected, actual, flag, call_wrapper);
}


void MacroAssembler::InvokeFunction(Handle<JSFunction> function,
                                    const ParameterCount& expected,
                                    const ParameterCount& actual,
                                    InvokeFlag flag,
                                    const CallWrapper& call_wrapper) {
  li(a1, function);
  InvokeFunction(a1, expected, actual, flag, call_wrapper);
}


void MacroAssembler::IsObjectJSObjectType(Register heap_object,
                                          Register map,
                                          Register scratch,
                                          Label* fail) {
  ld(map, FieldMemOperand(heap_object, HeapObject::kMapOffset));
  IsInstanceJSObjectType(map, scratch, fail);
}


void MacroAssembler::IsInstanceJSObjectType(Register map,
                                            Register scratch,
                                            Label* fail) {
  lbu(scratch, FieldMemOperand(map, Map::kInstanceTypeOffset));
  Branch(fail, lt, scratch, Operand(FIRST_NONCALLABLE_SPEC_OBJECT_TYPE));
  Branch(fail, gt, scratch, Operand(LAST_NONCALLABLE_SPEC_OBJECT_TYPE));
}


void MacroAssembler::IsObjectJSStringType(Register object,
                                          Register scratch,
                                          Label* fail) {
  DCHECK(kNotStringTag != 0);

  ld(scratch, FieldMemOperand(object, HeapObject::kMapOffset));
  lbu(scratch, FieldMemOperand(scratch, Map::kInstanceTypeOffset));
  And(scratch, scratch, Operand(kIsNotStringMask));
  Branch(fail, ne, scratch, Operand(zero_reg));
}


void MacroAssembler::IsObjectNameType(Register object,
                                      Register scratch,
                                      Label* fail) {
  ld(scratch, FieldMemOperand(object, HeapObject::kMapOffset));
  lbu(scratch, FieldMemOperand(scratch, Map::kInstanceTypeOffset));
  Branch(fail, hi, scratch, Operand(LAST_NAME_TYPE));
}


// ---------------------------------------------------------------------------
// Support functions.


void MacroAssembler::TryGetFunctionPrototype(Register function,
                                             Register result,
                                             Register scratch,
                                             Label* miss,
                                             bool miss_on_bound_function) {
  Label non_instance;
  if (miss_on_bound_function) {
    // Check that the receiver isn't a smi.
    JumpIfSmi(function, miss);

    // Check that the function really is a function.  Load map into result reg.
    GetObjectType(function, result, scratch);
    Branch(miss, ne, scratch, Operand(JS_FUNCTION_TYPE));

    ld(scratch,
       FieldMemOperand(function, JSFunction::kSharedFunctionInfoOffset));
    lwu(scratch,
        FieldMemOperand(scratch, SharedFunctionInfo::kCompilerHintsOffset));
    And(scratch, scratch,
        Operand(1 << SharedFunctionInfo::kBoundFunction));
    Branch(miss, ne, scratch, Operand(zero_reg));

    // Make sure that the function has an instance prototype.
    lbu(scratch, FieldMemOperand(result, Map::kBitFieldOffset));
    And(scratch, scratch, Operand(1 << Map::kHasNonInstancePrototype));
    Branch(&non_instance, ne, scratch, Operand(zero_reg));
  }

  // Get the prototype or initial map from the function.
  ld(result,
     FieldMemOperand(function, JSFunction::kPrototypeOrInitialMapOffset));

  // If the prototype or initial map is the hole, don't return it and
  // simply miss the cache instead. This will allow us to allocate a
  // prototype object on-demand in the runtime system.
  LoadRoot(t8, Heap::kTheHoleValueRootIndex);
  Branch(miss, eq, result, Operand(t8));

  // If the function does not have an initial map, we're done.
  Label done;
  GetObjectType(result, scratch, scratch);
  Branch(&done, ne, scratch, Operand(MAP_TYPE));

  // Get the prototype from the initial map.
  ld(result, FieldMemOperand(result, Map::kPrototypeOffset));

  if (miss_on_bound_function) {
    jmp(&done);

    // Non-instance prototype: Fetch prototype from constructor field
    // in initial map.
    bind(&non_instance);
    ld(result, FieldMemOperand(result, Map::kConstructorOffset));
  }

  // All done.
  bind(&done);
}


void MacroAssembler::GetObjectType(Register object,
                                   Register map,
                                   Register type_reg) {
  ld(map, FieldMemOperand(object, HeapObject::kMapOffset));
  lbu(type_reg, FieldMemOperand(map, Map::kInstanceTypeOffset));
}


// -----------------------------------------------------------------------------
// Runtime calls.

void MacroAssembler::CallStub(CodeStub* stub,
                              TypeFeedbackId ast_id,
                              Condition cond,
                              Register r1,
                              const Operand& r2,
                              BranchDelaySlot bd) {
  DCHECK(AllowThisStubCall(stub));  // Stub calls are not allowed in some stubs.
  Call(stub->GetCode(), RelocInfo::CODE_TARGET, ast_id,
       cond, r1, r2, bd);
}


void MacroAssembler::TailCallStub(CodeStub* stub,
                                  Condition cond,
                                  Register r1,
                                  const Operand& r2,
                                  BranchDelaySlot bd) {
  Jump(stub->GetCode(), RelocInfo::CODE_TARGET, cond, r1, r2, bd);
}


static int AddressOffset(ExternalReference ref0, ExternalReference ref1) {
  int64_t offset = (ref0.address() - ref1.address());
  DCHECK(static_cast<int>(offset) == offset);
  return static_cast<int>(offset);
}


void MacroAssembler::CallApiFunctionAndReturn(
    Register function_address,
    ExternalReference thunk_ref,
    int stack_space,
    MemOperand return_value_operand,
    MemOperand* context_restore_operand) {
  ExternalReference next_address =
      ExternalReference::handle_scope_next_address(isolate());
  const int kNextOffset = 0;
  const int kLimitOffset = AddressOffset(
      ExternalReference::handle_scope_limit_address(isolate()),
      next_address);
  const int kLevelOffset = AddressOffset(
      ExternalReference::handle_scope_level_address(isolate()),
      next_address);

  DCHECK(function_address.is(a1) || function_address.is(a2));

  Label profiler_disabled;
  Label end_profiler_check;
  li(t9, Operand(ExternalReference::is_profiling_address(isolate())));
  lb(t9, MemOperand(t9, 0));
  Branch(&profiler_disabled, eq, t9, Operand(zero_reg));

  // Additional parameter is the address of the actual callback.
  li(t9, Operand(thunk_ref));
  jmp(&end_profiler_check);

  bind(&profiler_disabled);
  mov(t9, function_address);
  bind(&end_profiler_check);

  // Allocate HandleScope in callee-save registers.
  li(s3, Operand(next_address));
  ld(s0, MemOperand(s3, kNextOffset));
  ld(s1, MemOperand(s3, kLimitOffset));
  ld(s2, MemOperand(s3, kLevelOffset));
  Daddu(s2, s2, Operand(1));
  sd(s2, MemOperand(s3, kLevelOffset));

  if (FLAG_log_timer_events) {
    FrameScope frame(this, StackFrame::MANUAL);
    PushSafepointRegisters();
    PrepareCallCFunction(1, a0);
    li(a0, Operand(ExternalReference::isolate_address(isolate())));
    CallCFunction(ExternalReference::log_enter_external_function(isolate()), 1);
    PopSafepointRegisters();
  }

  // Native call returns to the DirectCEntry stub which redirects to the
  // return address pushed on stack (could have moved after GC).
  // DirectCEntry stub itself is generated early and never moves.
  DirectCEntryStub stub(isolate());
  stub.GenerateCall(this, t9);

  if (FLAG_log_timer_events) {
    FrameScope frame(this, StackFrame::MANUAL);
    PushSafepointRegisters();
    PrepareCallCFunction(1, a0);
    li(a0, Operand(ExternalReference::isolate_address(isolate())));
    CallCFunction(ExternalReference::log_leave_external_function(isolate()), 1);
    PopSafepointRegisters();
  }

  Label promote_scheduled_exception;
  Label exception_handled;
  Label delete_allocated_handles;
  Label leave_exit_frame;
  Label return_value_loaded;

  // Load value from ReturnValue.
  ld(v0, return_value_operand);
  bind(&return_value_loaded);

  // No more valid handles (the result handle was the last one). Restore
  // previous handle scope.
  sd(s0, MemOperand(s3, kNextOffset));
  if (emit_debug_code()) {
    ld(a1, MemOperand(s3, kLevelOffset));
    Check(eq, kUnexpectedLevelAfterReturnFromApiCall, a1, Operand(s2));
  }
  Dsubu(s2, s2, Operand(1));
  sd(s2, MemOperand(s3, kLevelOffset));
  ld(at, MemOperand(s3, kLimitOffset));
  Branch(&delete_allocated_handles, ne, s1, Operand(at));

  // Check if the function scheduled an exception.
  bind(&leave_exit_frame);
  LoadRoot(a4, Heap::kTheHoleValueRootIndex);
  li(at, Operand(ExternalReference::scheduled_exception_address(isolate())));
  ld(a5, MemOperand(at));
  Branch(&promote_scheduled_exception, ne, a4, Operand(a5));
  bind(&exception_handled);

  bool restore_context = context_restore_operand != NULL;
  if (restore_context) {
    ld(cp, *context_restore_operand);
  }
  li(s0, Operand(stack_space));
  LeaveExitFrame(false, s0, !restore_context, EMIT_RETURN);

  bind(&promote_scheduled_exception);
  {
    FrameScope frame(this, StackFrame::INTERNAL);
    CallExternalReference(
        ExternalReference(Runtime::kPromoteScheduledException, isolate()),
        0);
  }
  jmp(&exception_handled);

  // HandleScope limit has changed. Delete allocated extensions.
  bind(&delete_allocated_handles);
  sd(s1, MemOperand(s3, kLimitOffset));
  mov(s0, v0);
  mov(a0, v0);
  PrepareCallCFunction(1, s1);
  li(a0, Operand(ExternalReference::isolate_address(isolate())));
  CallCFunction(ExternalReference::delete_handle_scope_extensions(isolate()),
      1);
  mov(v0, s0);
  jmp(&leave_exit_frame);
}


bool MacroAssembler::AllowThisStubCall(CodeStub* stub) {
  return has_frame_ || !stub->SometimesSetsUpAFrame();
}


void MacroAssembler::IndexFromHash(Register hash, Register index) {
  // If the hash field contains an array index pick it out. The assert checks
  // that the constants for the maximum number of digits for an array index
  // cached in the hash field and the number of bits reserved for it does not
  // conflict.
  DCHECK(TenToThe(String::kMaxCachedArrayIndexLength) <
         (1 << String::kArrayIndexValueBits));
  DecodeFieldToSmi<String::ArrayIndexValueBits>(index, hash);
}


void MacroAssembler::ObjectToDoubleFPURegister(Register object,
                                               FPURegister result,
                                               Register scratch1,
                                               Register scratch2,
                                               Register heap_number_map,
                                               Label* not_number,
                                               ObjectToDoubleFlags flags) {
  Label done;
  if ((flags & OBJECT_NOT_SMI) == 0) {
    Label not_smi;
    JumpIfNotSmi(object, &not_smi);
    // Remove smi tag and convert to double.
    // dsra(scratch1, object, kSmiTagSize);
    dsra32(scratch1, object, 0);
    mtc1(scratch1, result);
    cvt_d_w(result, result);
    Branch(&done);
    bind(&not_smi);
  }
  // Check for heap number and load double value from it.
  ld(scratch1, FieldMemOperand(object, HeapObject::kMapOffset));
  Branch(not_number, ne, scratch1, Operand(heap_number_map));

  if ((flags & AVOID_NANS_AND_INFINITIES) != 0) {
    // If exponent is all ones the number is either a NaN or +/-Infinity.
    Register exponent = scratch1;
    Register mask_reg = scratch2;
    lwu(exponent, FieldMemOperand(object, HeapNumber::kExponentOffset));
    li(mask_reg, HeapNumber::kExponentMask);

    And(exponent, exponent, mask_reg);
    Branch(not_number, eq, exponent, Operand(mask_reg));
  }
  ldc1(result, FieldMemOperand(object, HeapNumber::kValueOffset));
  bind(&done);
}


void MacroAssembler::SmiToDoubleFPURegister(Register smi,
                                            FPURegister value,
                                            Register scratch1) {
  // dsra(scratch1, smi, kSmiTagSize);
  dsra32(scratch1, smi, 0);
  mtc1(scratch1, value);
  cvt_d_w(value, value);
}


void MacroAssembler::AdduAndCheckForOverflow(Register dst, Register left,
                                             const Operand& right,
                                             Register overflow_dst,
                                             Register scratch) {
  if (right.is_reg()) {
    AdduAndCheckForOverflow(dst, left, right.rm(), overflow_dst, scratch);
  } else {
    if (dst.is(left)) {
      mov(scratch, left);                    // Preserve left.
      daddiu(dst, left, right.immediate());  // Left is overwritten.
      xor_(scratch, dst, scratch);           // Original left.
      // Load right since xori takes uint16 as immediate.
      daddiu(t9, zero_reg, right.immediate());
      xor_(overflow_dst, dst, t9);
      and_(overflow_dst, overflow_dst, scratch);
    } else {
      daddiu(dst, left, right.immediate());
      xor_(overflow_dst, dst, left);
      // Load right since xori takes uint16 as immediate.
      daddiu(t9, zero_reg, right.immediate());
      xor_(scratch, dst, t9);
      and_(overflow_dst, scratch, overflow_dst);
    }
  }
}


void MacroAssembler::AdduAndCheckForOverflow(Register dst,
                                             Register left,
                                             Register right,
                                             Register overflow_dst,
                                             Register scratch) {
  DCHECK(!dst.is(overflow_dst));
  DCHECK(!dst.is(scratch));
  DCHECK(!overflow_dst.is(scratch));
  DCHECK(!overflow_dst.is(left));
  DCHECK(!overflow_dst.is(right));

  if (left.is(right) && dst.is(left)) {
    DCHECK(!dst.is(t9));
    DCHECK(!scratch.is(t9));
    DCHECK(!left.is(t9));
    DCHECK(!right.is(t9));
    DCHECK(!overflow_dst.is(t9));
    mov(t9, right);
    right = t9;
  }

  if (dst.is(left)) {
    mov(scratch, left);  // Preserve left.
    daddu(dst, left, right);  // Left is overwritten.
    xor_(scratch, dst, scratch);  // Original left.
    xor_(overflow_dst, dst, right);
    and_(overflow_dst, overflow_dst, scratch);
  } else if (dst.is(right)) {
    mov(scratch, right);  // Preserve right.
    daddu(dst, left, right);  // Right is overwritten.
    xor_(scratch, dst, scratch);  // Original right.
    xor_(overflow_dst, dst, left);
    and_(overflow_dst, overflow_dst, scratch);
  } else {
    daddu(dst, left, right);
    xor_(overflow_dst, dst, left);
    xor_(scratch, dst, right);
    and_(overflow_dst, scratch, overflow_dst);
  }
}


void MacroAssembler::SubuAndCheckForOverflow(Register dst, Register left,
                                             const Operand& right,
                                             Register overflow_dst,
                                             Register scratch) {
  if (right.is_reg()) {
    SubuAndCheckForOverflow(dst, left, right.rm(), overflow_dst, scratch);
  } else {
    if (dst.is(left)) {
      mov(scratch, left);                       // Preserve left.
      daddiu(dst, left, -(right.immediate()));  // Left is overwritten.
      xor_(overflow_dst, dst, scratch);         // scratch is original left.
      // Load right since xori takes uint16 as immediate.
      daddiu(t9, zero_reg, right.immediate());
      xor_(scratch, scratch, t9);  // scratch is original left.
      and_(overflow_dst, scratch, overflow_dst);
    } else {
      daddiu(dst, left, -(right.immediate()));
      xor_(overflow_dst, dst, left);
      // Load right since xori takes uint16 as immediate.
      daddiu(t9, zero_reg, right.immediate());
      xor_(scratch, left, t9);
      and_(overflow_dst, scratch, overflow_dst);
    }
  }
}


void MacroAssembler::SubuAndCheckForOverflow(Register dst,
                                             Register left,
                                             Register right,
                                             Register overflow_dst,
                                             Register scratch) {
  DCHECK(!dst.is(overflow_dst));
  DCHECK(!dst.is(scratch));
  DCHECK(!overflow_dst.is(scratch));
  DCHECK(!overflow_dst.is(left));
  DCHECK(!overflow_dst.is(right));
  DCHECK(!scratch.is(left));
  DCHECK(!scratch.is(right));

  // This happens with some crankshaft code. Since Subu works fine if
  // left == right, let's not make that restriction here.
  if (left.is(right)) {
    mov(dst, zero_reg);
    mov(overflow_dst, zero_reg);
    return;
  }

  if (dst.is(left)) {
    mov(scratch, left);  // Preserve left.
    dsubu(dst, left, right);  // Left is overwritten.
    xor_(overflow_dst, dst, scratch);  // scratch is original left.
    xor_(scratch, scratch, right);  // scratch is original left.
    and_(overflow_dst, scratch, overflow_dst);
  } else if (dst.is(right)) {
    mov(scratch, right);  // Preserve right.
    dsubu(dst, left, right);  // Right is overwritten.
    xor_(overflow_dst, dst, left);
    xor_(scratch, left, scratch);  // Original right.
    and_(overflow_dst, scratch, overflow_dst);
  } else {
    dsubu(dst, left, right);
    xor_(overflow_dst, dst, left);
    xor_(scratch, left, right);
    and_(overflow_dst, scratch, overflow_dst);
  }
}


void MacroAssembler::CallRuntime(const Runtime::Function* f,
                                 int num_arguments,
                                 SaveFPRegsMode save_doubles) {
  // All parameters are on the stack. v0 has the return value after call.

  // If the expected number of arguments of the runtime function is
  // constant, we check that the actual number of arguments match the
  // expectation.
  CHECK(f->nargs < 0 || f->nargs == num_arguments);

  // TODO(1236192): Most runtime routines don't need the number of
  // arguments passed in because it is constant. At some point we
  // should remove this need and make the runtime routine entry code
  // smarter.
  PrepareCEntryArgs(num_arguments);
  PrepareCEntryFunction(ExternalReference(f, isolate()));
  CEntryStub stub(isolate(), 1, save_doubles);
  CallStub(&stub);
}


void MacroAssembler::CallExternalReference(const ExternalReference& ext,
                                           int num_arguments,
                                           BranchDelaySlot bd) {
  PrepareCEntryArgs(num_arguments);
  PrepareCEntryFunction(ext);

  CEntryStub stub(isolate(), 1);
  CallStub(&stub, TypeFeedbackId::None(), al, zero_reg, Operand(zero_reg), bd);
}


void MacroAssembler::TailCallExternalReference(const ExternalReference& ext,
                                               int num_arguments,
                                               int result_size) {
  // TODO(1236192): Most runtime routines don't need the number of
  // arguments passed in because it is constant. At some point we
  // should remove this need and make the runtime routine entry code
  // smarter.
  PrepareCEntryArgs(num_arguments);
  JumpToExternalReference(ext);
}


void MacroAssembler::TailCallRuntime(Runtime::FunctionId fid,
                                     int num_arguments,
                                     int result_size) {
  TailCallExternalReference(ExternalReference(fid, isolate()),
                            num_arguments,
                            result_size);
}


void MacroAssembler::JumpToExternalReference(const ExternalReference& builtin,
                                             BranchDelaySlot bd) {
  PrepareCEntryFunction(builtin);
  CEntryStub stub(isolate(), 1);
  Jump(stub.GetCode(),
       RelocInfo::CODE_TARGET,
       al,
       zero_reg,
       Operand(zero_reg),
       bd);
}


void MacroAssembler::InvokeBuiltin(Builtins::JavaScript id,
                                   InvokeFlag flag,
                                   const CallWrapper& call_wrapper) {
  // You can't call a builtin without a valid frame.
  DCHECK(flag == JUMP_FUNCTION || has_frame());

  GetBuiltinEntry(t9, id);
  if (flag == CALL_FUNCTION) {
    call_wrapper.BeforeCall(CallSize(t9));
    Call(t9);
    call_wrapper.AfterCall();
  } else {
    DCHECK(flag == JUMP_FUNCTION);
    Jump(t9);
  }
}


void MacroAssembler::GetBuiltinFunction(Register target,
                                        Builtins::JavaScript id) {
  // Load the builtins object into target register.
  ld(target, MemOperand(cp, Context::SlotOffset(Context::GLOBAL_OBJECT_INDEX)));
  ld(target, FieldMemOperand(target, GlobalObject::kBuiltinsOffset));
  // Load the JavaScript builtin function from the builtins object.
  ld(target, FieldMemOperand(target,
                          JSBuiltinsObject::OffsetOfFunctionWithId(id)));
}


void MacroAssembler::GetBuiltinEntry(Register target, Builtins::JavaScript id) {
  DCHECK(!target.is(a1));
  GetBuiltinFunction(a1, id);
  // Load the code entry point from the builtins object.
  ld(target, FieldMemOperand(a1, JSFunction::kCodeEntryOffset));
}


void MacroAssembler::SetCounter(StatsCounter* counter, int value,
                                Register scratch1, Register scratch2) {
  if (FLAG_native_code_counters && counter->Enabled()) {
    li(scratch1, Operand(value));
    li(scratch2, Operand(ExternalReference(counter)));
    sd(scratch1, MemOperand(scratch2));
  }
}


void MacroAssembler::IncrementCounter(StatsCounter* counter, int value,
                                      Register scratch1, Register scratch2) {
  DCHECK(value > 0);
  if (FLAG_native_code_counters && counter->Enabled()) {
    li(scratch2, Operand(ExternalReference(counter)));
    ld(scratch1, MemOperand(scratch2));
    Daddu(scratch1, scratch1, Operand(value));
    sd(scratch1, MemOperand(scratch2));
  }
}


void MacroAssembler::DecrementCounter(StatsCounter* counter, int value,
                                      Register scratch1, Register scratch2) {
  DCHECK(value > 0);
  if (FLAG_native_code_counters && counter->Enabled()) {
    li(scratch2, Operand(ExternalReference(counter)));
    ld(scratch1, MemOperand(scratch2));
    Dsubu(scratch1, scratch1, Operand(value));
    sd(scratch1, MemOperand(scratch2));
  }
}


// -----------------------------------------------------------------------------
// Debugging.

void MacroAssembler::Assert(Condition cc, BailoutReason reason,
                            Register rs, Operand rt) {
  if (emit_debug_code())
    Check(cc, reason, rs, rt);
}


void MacroAssembler::AssertFastElements(Register elements) {
  if (emit_debug_code()) {
    DCHECK(!elements.is(at));
    Label ok;
    push(elements);
    ld(elements, FieldMemOperand(elements, HeapObject::kMapOffset));
    LoadRoot(at, Heap::kFixedArrayMapRootIndex);
    Branch(&ok, eq, elements, Operand(at));
    LoadRoot(at, Heap::kFixedDoubleArrayMapRootIndex);
    Branch(&ok, eq, elements, Operand(at));
    LoadRoot(at, Heap::kFixedCOWArrayMapRootIndex);
    Branch(&ok, eq, elements, Operand(at));
    Abort(kJSObjectWithFastElementsMapHasSlowElements);
    bind(&ok);
    pop(elements);
  }
}


void MacroAssembler::Check(Condition cc, BailoutReason reason,
                           Register rs, Operand rt) {
  Label L;
  Branch(&L, cc, rs, rt);
  Abort(reason);
  // Will not return here.
  bind(&L);
}


void MacroAssembler::Abort(BailoutReason reason) {
  Label abort_start;
  bind(&abort_start);
#ifdef DEBUG
  const char* msg = GetBailoutReason(reason);
  if (msg != NULL) {
    RecordComment("Abort message: ");
    RecordComment(msg);
  }

  if (FLAG_trap_on_abort) {
    stop(msg);
    return;
  }
#endif

  li(a0, Operand(Smi::FromInt(reason)));
  push(a0);
  // Disable stub call restrictions to always allow calls to abort.
  if (!has_frame_) {
    // We don't actually want to generate a pile of code for this, so just
    // claim there is a stack frame, without generating one.
    FrameScope scope(this, StackFrame::NONE);
    CallRuntime(Runtime::kAbort, 1);
  } else {
    CallRuntime(Runtime::kAbort, 1);
  }
  // Will not return here.
  if (is_trampoline_pool_blocked()) {
    // If the calling code cares about the exact number of
    // instructions generated, we insert padding here to keep the size
    // of the Abort macro constant.
    // Currently in debug mode with debug_code enabled the number of
    // generated instructions is 10, so we use this as a maximum value.
    static const int kExpectedAbortInstructions = 10;
    int abort_instructions = InstructionsGeneratedSince(&abort_start);
    DCHECK(abort_instructions <= kExpectedAbortInstructions);
    while (abort_instructions++ < kExpectedAbortInstructions) {
      nop();
    }
  }
}


void MacroAssembler::LoadContext(Register dst, int context_chain_length) {
  if (context_chain_length > 0) {
    // Move up the chain of contexts to the context containing the slot.
    ld(dst, MemOperand(cp, Context::SlotOffset(Context::PREVIOUS_INDEX)));
    for (int i = 1; i < context_chain_length; i++) {
      ld(dst, MemOperand(dst, Context::SlotOffset(Context::PREVIOUS_INDEX)));
    }
  } else {
    // Slot is in the current function context.  Move it into the
    // destination register in case we store into it (the write barrier
    // cannot be allowed to destroy the context in esi).
    Move(dst, cp);
  }
}


void MacroAssembler::LoadTransitionedArrayMapConditional(
    ElementsKind expected_kind,
    ElementsKind transitioned_kind,
    Register map_in_out,
    Register scratch,
    Label* no_map_match) {
  // Load the global or builtins object from the current context.
  ld(scratch,
     MemOperand(cp, Context::SlotOffset(Context::GLOBAL_OBJECT_INDEX)));
  ld(scratch, FieldMemOperand(scratch, GlobalObject::kNativeContextOffset));

  // Check that the function's map is the same as the expected cached map.
  ld(scratch,
     MemOperand(scratch,
                Context::SlotOffset(Context::JS_ARRAY_MAPS_INDEX)));
  size_t offset = expected_kind * kPointerSize +
      FixedArrayBase::kHeaderSize;
  ld(at, FieldMemOperand(scratch, offset));
  Branch(no_map_match, ne, map_in_out, Operand(at));

  // Use the transitioned cached map.
  offset = transitioned_kind * kPointerSize +
      FixedArrayBase::kHeaderSize;
  ld(map_in_out, FieldMemOperand(scratch, offset));
}


void MacroAssembler::LoadGlobalFunction(int index, Register function) {
  // Load the global or builtins object from the current context.
  ld(function,
     MemOperand(cp, Context::SlotOffset(Context::GLOBAL_OBJECT_INDEX)));
  // Load the native context from the global or builtins object.
  ld(function, FieldMemOperand(function,
                               GlobalObject::kNativeContextOffset));
  // Load the function from the native context.
  ld(function, MemOperand(function, Context::SlotOffset(index)));
}


void MacroAssembler::LoadGlobalFunctionInitialMap(Register function,
                                                  Register map,
                                                  Register scratch) {
  // Load the initial map. The global functions all have initial maps.
  ld(map, FieldMemOperand(function, JSFunction::kPrototypeOrInitialMapOffset));
  if (emit_debug_code()) {
    Label ok, fail;
    CheckMap(map, scratch, Heap::kMetaMapRootIndex, &fail, DO_SMI_CHECK);
    Branch(&ok);
    bind(&fail);
    Abort(kGlobalFunctionsMustHaveInitialMap);
    bind(&ok);
  }
}


void MacroAssembler::StubPrologue() {
    Push(ra, fp, cp);
    Push(Smi::FromInt(StackFrame::STUB));
    // Adjust FP to point to saved FP.
    Daddu(fp, sp, Operand(StandardFrameConstants::kFixedFrameSizeFromFp));
}


void MacroAssembler::Prologue(bool code_pre_aging) {
  PredictableCodeSizeScope predictible_code_size_scope(
      this, kNoCodeAgeSequenceLength);
  // The following three instructions must remain together and unmodified
  // for code aging to work properly.
  if (code_pre_aging) {
    // Pre-age the code.
    Code* stub = Code::GetPreAgedCodeAgeStub(isolate());
    nop(Assembler::CODE_AGE_MARKER_NOP);
    // Load the stub address to t9 and call it,
    // GetCodeAgeAndParity() extracts the stub address from this instruction.
    li(t9,
       Operand(reinterpret_cast<uint64_t>(stub->instruction_start())),
       ADDRESS_LOAD);
    nop();  // Prevent jalr to jal optimization.
    jalr(t9, a0);
    nop();  // Branch delay slot nop.
    nop();  // Pad the empty space.
  } else {
    Push(ra, fp, cp, a1);
    nop(Assembler::CODE_AGE_SEQUENCE_NOP);
    nop(Assembler::CODE_AGE_SEQUENCE_NOP);
    nop(Assembler::CODE_AGE_SEQUENCE_NOP);
    // Adjust fp to point to caller's fp.
    Daddu(fp, sp, Operand(StandardFrameConstants::kFixedFrameSizeFromFp));
  }
}


void MacroAssembler::EnterFrame(StackFrame::Type type,
                                bool load_constant_pool_pointer_reg) {
  // Out-of-line constant pool not implemented on mips64.
  UNREACHABLE();
}


void MacroAssembler::EnterFrame(StackFrame::Type type) {
  daddiu(sp, sp, -5 * kPointerSize);
  li(t8, Operand(Smi::FromInt(type)));
  li(t9, Operand(CodeObject()), CONSTANT_SIZE);
  sd(ra, MemOperand(sp, 4 * kPointerSize));
  sd(fp, MemOperand(sp, 3 * kPointerSize));
  sd(cp, MemOperand(sp, 2 * kPointerSize));
  sd(t8, MemOperand(sp, 1 * kPointerSize));
  sd(t9, MemOperand(sp, 0 * kPointerSize));
  // Adjust FP to point to saved FP.
  Daddu(fp, sp,
       Operand(StandardFrameConstants::kFixedFrameSizeFromFp + kPointerSize));
}


void MacroAssembler::LeaveFrame(StackFrame::Type type) {
  mov(sp, fp);
  ld(fp, MemOperand(sp, 0 * kPointerSize));
  ld(ra, MemOperand(sp, 1 * kPointerSize));
  daddiu(sp, sp, 2 * kPointerSize);
}


void MacroAssembler::EnterExitFrame(bool save_doubles,
                                    int stack_space) {
  // Set up the frame structure on the stack.
  STATIC_ASSERT(2 * kPointerSize == ExitFrameConstants::kCallerSPDisplacement);
  STATIC_ASSERT(1 * kPointerSize == ExitFrameConstants::kCallerPCOffset);
  STATIC_ASSERT(0 * kPointerSize == ExitFrameConstants::kCallerFPOffset);

  // This is how the stack will look:
  // fp + 2 (==kCallerSPDisplacement) - old stack's end
  // [fp + 1 (==kCallerPCOffset)] - saved old ra
  // [fp + 0 (==kCallerFPOffset)] - saved old fp
  // [fp - 1 (==kSPOffset)] - sp of the called function
  // [fp - 2 (==kCodeOffset)] - CodeObject
  // fp - (2 + stack_space + alignment) == sp == [fp - kSPOffset] - top of the
  //   new stack (will contain saved ra)

  // Save registers.
  daddiu(sp, sp, -4 * kPointerSize);
  sd(ra, MemOperand(sp, 3 * kPointerSize));
  sd(fp, MemOperand(sp, 2 * kPointerSize));
  daddiu(fp, sp, 2 * kPointerSize);  // Set up new frame pointer.

  if (emit_debug_code()) {
    sd(zero_reg, MemOperand(fp, ExitFrameConstants::kSPOffset));
  }

  // Accessed from ExitFrame::code_slot.
  li(t8, Operand(CodeObject()), CONSTANT_SIZE);
  sd(t8, MemOperand(fp, ExitFrameConstants::kCodeOffset));

  // Save the frame pointer and the context in top.
  li(t8, Operand(ExternalReference(Isolate::kCEntryFPAddress, isolate())));
  sd(fp, MemOperand(t8));
  li(t8, Operand(ExternalReference(Isolate::kContextAddress, isolate())));
  sd(cp, MemOperand(t8));

  const int frame_alignment = MacroAssembler::ActivationFrameAlignment();
  if (save_doubles) {
    // The stack is already aligned to 0 modulo 8 for stores with sdc1.
    int kNumOfSavedRegisters = FPURegister::kMaxNumRegisters / 2;
    int space = kNumOfSavedRegisters * kDoubleSize ;
    Dsubu(sp, sp, Operand(space));
    // Remember: we only need to save every 2nd double FPU value.
    for (int i = 0; i < kNumOfSavedRegisters; i++) {
      FPURegister reg = FPURegister::from_code(2 * i);
      sdc1(reg, MemOperand(sp, i * kDoubleSize));
    }
  }

  // Reserve place for the return address, stack space and an optional slot
  // (used by the DirectCEntryStub to hold the return value if a struct is
  // returned) and align the frame preparing for calling the runtime function.
  DCHECK(stack_space >= 0);
  Dsubu(sp, sp, Operand((stack_space + 2) * kPointerSize));
  if (frame_alignment > 0) {
    DCHECK(base::bits::IsPowerOfTwo32(frame_alignment));
    And(sp, sp, Operand(-frame_alignment));  // Align stack.
  }

  // Set the exit frame sp value to point just before the return address
  // location.
  daddiu(at, sp, kPointerSize);
  sd(at, MemOperand(fp, ExitFrameConstants::kSPOffset));
}


void MacroAssembler::LeaveExitFrame(bool save_doubles,
                                    Register argument_count,
                                    bool restore_context,
                                    bool do_return) {
  // Optionally restore all double registers.
  if (save_doubles) {
    // Remember: we only need to restore every 2nd double FPU value.
    int kNumOfSavedRegisters = FPURegister::kMaxNumRegisters / 2;
    Dsubu(t8, fp, Operand(ExitFrameConstants::kFrameSize +
        kNumOfSavedRegisters * kDoubleSize));
    for (int i = 0; i < kNumOfSavedRegisters; i++) {
      FPURegister reg = FPURegister::from_code(2 * i);
      ldc1(reg, MemOperand(t8, i  * kDoubleSize));
    }
  }

  // Clear top frame.
  li(t8, Operand(ExternalReference(Isolate::kCEntryFPAddress, isolate())));
  sd(zero_reg, MemOperand(t8));

  // Restore current context from top and clear it in debug mode.
  if (restore_context) {
    li(t8, Operand(ExternalReference(Isolate::kContextAddress, isolate())));
    ld(cp, MemOperand(t8));
  }
#ifdef DEBUG
  li(t8, Operand(ExternalReference(Isolate::kContextAddress, isolate())));
  sd(a3, MemOperand(t8));
#endif

  // Pop the arguments, restore registers, and return.
  mov(sp, fp);  // Respect ABI stack constraint.
  ld(fp, MemOperand(sp, ExitFrameConstants::kCallerFPOffset));
  ld(ra, MemOperand(sp, ExitFrameConstants::kCallerPCOffset));

  if (argument_count.is_valid()) {
    dsll(t8, argument_count, kPointerSizeLog2);
    daddu(sp, sp, t8);
  }

  if (do_return) {
    Ret(USE_DELAY_SLOT);
    // If returning, the instruction in the delay slot will be the addiu below.
  }
  daddiu(sp, sp, 2 * kPointerSize);
}


void MacroAssembler::InitializeNewString(Register string,
                                         Register length,
                                         Heap::RootListIndex map_index,
                                         Register scratch1,
                                         Register scratch2) {
  // dsll(scratch1, length, kSmiTagSize);
  dsll32(scratch1, length, 0);
  LoadRoot(scratch2, map_index);
  sd(scratch1, FieldMemOperand(string, String::kLengthOffset));
  li(scratch1, Operand(String::kEmptyHashField));
  sd(scratch2, FieldMemOperand(string, HeapObject::kMapOffset));
  sd(scratch1, FieldMemOperand(string, String::kHashFieldOffset));
}


int MacroAssembler::ActivationFrameAlignment() {
#if V8_HOST_ARCH_MIPS || V8_HOST_ARCH_MIPS64
  // Running on the real platform. Use the alignment as mandated by the local
  // environment.
  // Note: This will break if we ever start generating snapshots on one Mips
  // platform for another Mips platform with a different alignment.
  return base::OS::ActivationFrameAlignment();
#else  // V8_HOST_ARCH_MIPS
  // If we are using the simulator then we should always align to the expected
  // alignment. As the simulator is used to generate snapshots we do not know
  // if the target platform will need alignment, so this is controlled from a
  // flag.
  return FLAG_sim_stack_alignment;
#endif  // V8_HOST_ARCH_MIPS
}


void MacroAssembler::AssertStackIsAligned() {
  if (emit_debug_code()) {
      const int frame_alignment = ActivationFrameAlignment();
      const int frame_alignment_mask = frame_alignment - 1;

      if (frame_alignment > kPointerSize) {
        Label alignment_as_expected;
        DCHECK(base::bits::IsPowerOfTwo32(frame_alignment));
        andi(at, sp, frame_alignment_mask);
        Branch(&alignment_as_expected, eq, at, Operand(zero_reg));
        // Don't use Check here, as it will call Runtime_Abort re-entering here.
        stop("Unexpected stack alignment");
        bind(&alignment_as_expected);
      }
    }
}


void MacroAssembler::JumpIfNotPowerOfTwoOrZero(
    Register reg,
    Register scratch,
    Label* not_power_of_two_or_zero) {
  Dsubu(scratch, reg, Operand(1));
  Branch(USE_DELAY_SLOT, not_power_of_two_or_zero, lt,
         scratch, Operand(zero_reg));
  and_(at, scratch, reg);  // In the delay slot.
  Branch(not_power_of_two_or_zero, ne, at, Operand(zero_reg));
}


void MacroAssembler::SmiTagCheckOverflow(Register reg, Register overflow) {
  DCHECK(!reg.is(overflow));
  mov(overflow, reg);  // Save original value.
  SmiTag(reg);
  xor_(overflow, overflow, reg);  // Overflow if (value ^ 2 * value) < 0.
}


void MacroAssembler::SmiTagCheckOverflow(Register dst,
                                         Register src,
                                         Register overflow) {
  if (dst.is(src)) {
    // Fall back to slower case.
    SmiTagCheckOverflow(dst, overflow);
  } else {
    DCHECK(!dst.is(src));
    DCHECK(!dst.is(overflow));
    DCHECK(!src.is(overflow));
    SmiTag(dst, src);
    xor_(overflow, dst, src);  // Overflow if (value ^ 2 * value) < 0.
  }
}


void MacroAssembler::SmiLoadUntag(Register dst, MemOperand src) {
  if (SmiValuesAre32Bits()) {
    lw(dst, UntagSmiMemOperand(src.rm(), src.offset()));
  } else {
    lw(dst, src);
    SmiUntag(dst);
  }
}


void MacroAssembler::SmiLoadScale(Register dst, MemOperand src, int scale) {
  if (SmiValuesAre32Bits()) {
    // TODO(plind): not clear if lw or ld faster here, need micro-benchmark.
    lw(dst, UntagSmiMemOperand(src.rm(), src.offset()));
    dsll(dst, dst, scale);
  } else {
    lw(dst, src);
    DCHECK(scale >= kSmiTagSize);
    sll(dst, dst, scale - kSmiTagSize);
  }
}


// Returns 2 values: the Smi and a scaled version of the int within the Smi.
void MacroAssembler::SmiLoadWithScale(Register d_smi,
                                      Register d_scaled,
                                      MemOperand src,
                                      int scale) {
  if (SmiValuesAre32Bits()) {
    ld(d_smi, src);
    dsra(d_scaled, d_smi, kSmiShift - scale);
  } else {
    lw(d_smi, src);
    DCHECK(scale >= kSmiTagSize);
    sll(d_scaled, d_smi, scale - kSmiTagSize);
  }
}


// Returns 2 values: the untagged Smi (int32) and scaled version of that int.
void MacroAssembler::SmiLoadUntagWithScale(Register d_int,
                                           Register d_scaled,
                                           MemOperand src,
                                           int scale) {
  if (SmiValuesAre32Bits()) {
    lw(d_int, UntagSmiMemOperand(src.rm(), src.offset()));
    dsll(d_scaled, d_int, scale);
  } else {
    lw(d_int, src);
    // Need both the int and the scaled in, so use two instructions.
    SmiUntag(d_int);
    sll(d_scaled, d_int, scale);
  }
}


void MacroAssembler::UntagAndJumpIfSmi(Register dst,
                                       Register src,
                                       Label* smi_case) {
  // DCHECK(!dst.is(src));
  JumpIfSmi(src, smi_case, at, USE_DELAY_SLOT);
  SmiUntag(dst, src);
}


void MacroAssembler::UntagAndJumpIfNotSmi(Register dst,
                                          Register src,
                                          Label* non_smi_case) {
  // DCHECK(!dst.is(src));
  JumpIfNotSmi(src, non_smi_case, at, USE_DELAY_SLOT);
  SmiUntag(dst, src);
}

void MacroAssembler::JumpIfSmi(Register value,
                               Label* smi_label,
                               Register scratch,
                               BranchDelaySlot bd) {
  DCHECK_EQ(0, kSmiTag);
  andi(scratch, value, kSmiTagMask);
  Branch(bd, smi_label, eq, scratch, Operand(zero_reg));
}

void MacroAssembler::JumpIfNotSmi(Register value,
                                  Label* not_smi_label,
                                  Register scratch,
                                  BranchDelaySlot bd) {
  DCHECK_EQ(0, kSmiTag);
  andi(scratch, value, kSmiTagMask);
  Branch(bd, not_smi_label, ne, scratch, Operand(zero_reg));
}


void MacroAssembler::JumpIfNotBothSmi(Register reg1,
                                      Register reg2,
                                      Label* on_not_both_smi) {
  STATIC_ASSERT(kSmiTag == 0);
  // TODO(plind): Find some better to fix this assert issue.
#if defined(__APPLE__)
  DCHECK_EQ(1, kSmiTagMask);
#else
  DCHECK_EQ((uint64_t)1, kSmiTagMask);
#endif
  or_(at, reg1, reg2);
  JumpIfNotSmi(at, on_not_both_smi);
}


void MacroAssembler::JumpIfEitherSmi(Register reg1,
                                     Register reg2,
                                     Label* on_either_smi) {
  STATIC_ASSERT(kSmiTag == 0);
  // TODO(plind): Find some better to fix this assert issue.
#if defined(__APPLE__)
  DCHECK_EQ(1, kSmiTagMask);
#else
  DCHECK_EQ((uint64_t)1, kSmiTagMask);
#endif
  // Both Smi tags must be 1 (not Smi).
  and_(at, reg1, reg2);
  JumpIfSmi(at, on_either_smi);
}


void MacroAssembler::AssertNotSmi(Register object) {
  if (emit_debug_code()) {
    STATIC_ASSERT(kSmiTag == 0);
    andi(at, object, kSmiTagMask);
    Check(ne, kOperandIsASmi, at, Operand(zero_reg));
  }
}


void MacroAssembler::AssertSmi(Register object) {
  if (emit_debug_code()) {
    STATIC_ASSERT(kSmiTag == 0);
    andi(at, object, kSmiTagMask);
    Check(eq, kOperandIsASmi, at, Operand(zero_reg));
  }
}


void MacroAssembler::AssertString(Register object) {
  if (emit_debug_code()) {
    STATIC_ASSERT(kSmiTag == 0);
    SmiTst(object, a4);
    Check(ne, kOperandIsASmiAndNotAString, a4, Operand(zero_reg));
    push(object);
    ld(object, FieldMemOperand(object, HeapObject::kMapOffset));
    lbu(object, FieldMemOperand(object, Map::kInstanceTypeOffset));
    Check(lo, kOperandIsNotAString, object, Operand(FIRST_NONSTRING_TYPE));
    pop(object);
  }
}


void MacroAssembler::AssertName(Register object) {
  if (emit_debug_code()) {
    STATIC_ASSERT(kSmiTag == 0);
    SmiTst(object, a4);
    Check(ne, kOperandIsASmiAndNotAName, a4, Operand(zero_reg));
    push(object);
    ld(object, FieldMemOperand(object, HeapObject::kMapOffset));
    lbu(object, FieldMemOperand(object, Map::kInstanceTypeOffset));
    Check(le, kOperandIsNotAName, object, Operand(LAST_NAME_TYPE));
    pop(object);
  }
}


void MacroAssembler::AssertUndefinedOrAllocationSite(Register object,
                                                     Register scratch) {
  if (emit_debug_code()) {
    Label done_checking;
    AssertNotSmi(object);
    LoadRoot(scratch, Heap::kUndefinedValueRootIndex);
    Branch(&done_checking, eq, object, Operand(scratch));
    push(object);
    ld(object, FieldMemOperand(object, HeapObject::kMapOffset));
    LoadRoot(scratch, Heap::kAllocationSiteMapRootIndex);
    Assert(eq, kExpectedUndefinedOrCell, object, Operand(scratch));
    pop(object);
    bind(&done_checking);
  }
}


void MacroAssembler::AssertIsRoot(Register reg, Heap::RootListIndex index) {
  if (emit_debug_code()) {
    DCHECK(!reg.is(at));
    LoadRoot(at, index);
    Check(eq, kHeapNumberMapRegisterClobbered, reg, Operand(at));
  }
}


void MacroAssembler::JumpIfNotHeapNumber(Register object,
                                         Register heap_number_map,
                                         Register scratch,
                                         Label* on_not_heap_number) {
  ld(scratch, FieldMemOperand(object, HeapObject::kMapOffset));
  AssertIsRoot(heap_number_map, Heap::kHeapNumberMapRootIndex);
  Branch(on_not_heap_number, ne, scratch, Operand(heap_number_map));
}


void MacroAssembler::LookupNumberStringCache(Register object,
                                             Register result,
                                             Register scratch1,
                                             Register scratch2,
                                             Register scratch3,
                                             Label* not_found) {
  // Use of registers. Register result is used as a temporary.
  Register number_string_cache = result;
  Register mask = scratch3;

  // Load the number string cache.
  LoadRoot(number_string_cache, Heap::kNumberStringCacheRootIndex);

  // Make the hash mask from the length of the number string cache. It
  // contains two elements (number and string) for each cache entry.
  ld(mask, FieldMemOperand(number_string_cache, FixedArray::kLengthOffset));
  // Divide length by two (length is a smi).
  // dsra(mask, mask, kSmiTagSize + 1);
  dsra32(mask, mask, 1);
  Daddu(mask, mask, -1);  // Make mask.

  // Calculate the entry in the number string cache. The hash value in the
  // number string cache for smis is just the smi value, and the hash for
  // doubles is the xor of the upper and lower words. See
  // Heap::GetNumberStringCache.
  Label is_smi;
  Label load_result_from_cache;
  JumpIfSmi(object, &is_smi);
  CheckMap(object,
           scratch1,
           Heap::kHeapNumberMapRootIndex,
           not_found,
           DONT_DO_SMI_CHECK);

  STATIC_ASSERT(8 == kDoubleSize);
  Daddu(scratch1,
       object,
       Operand(HeapNumber::kValueOffset - kHeapObjectTag));
  ld(scratch2, MemOperand(scratch1, kPointerSize));
  ld(scratch1, MemOperand(scratch1, 0));
  Xor(scratch1, scratch1, Operand(scratch2));
  And(scratch1, scratch1, Operand(mask));

  // Calculate address of entry in string cache: each entry consists
  // of two pointer sized fields.
  dsll(scratch1, scratch1, kPointerSizeLog2 + 1);
  Daddu(scratch1, number_string_cache, scratch1);

  Register probe = mask;
  ld(probe, FieldMemOperand(scratch1, FixedArray::kHeaderSize));
  JumpIfSmi(probe, not_found);
  ldc1(f12, FieldMemOperand(object, HeapNumber::kValueOffset));
  ldc1(f14, FieldMemOperand(probe, HeapNumber::kValueOffset));
  BranchF(&load_result_from_cache, NULL, eq, f12, f14);
  Branch(not_found);

  bind(&is_smi);
  Register scratch = scratch1;
  // dsra(scratch, object, 1);   // Shift away the tag.
  dsra32(scratch, scratch, 0);
  And(scratch, mask, Operand(scratch));

  // Calculate address of entry in string cache: each entry consists
  // of two pointer sized fields.
  dsll(scratch, scratch, kPointerSizeLog2 + 1);
  Daddu(scratch, number_string_cache, scratch);

  // Check if the entry is the smi we are looking for.
  ld(probe, FieldMemOperand(scratch, FixedArray::kHeaderSize));
  Branch(not_found, ne, object, Operand(probe));

  // Get the result from the cache.
  bind(&load_result_from_cache);
  ld(result, FieldMemOperand(scratch, FixedArray::kHeaderSize + kPointerSize));

  IncrementCounter(isolate()->counters()->number_to_string_native(),
                   1,
                   scratch1,
                   scratch2);
}


void MacroAssembler::JumpIfNonSmisNotBothSequentialOneByteStrings(
    Register first, Register second, Register scratch1, Register scratch2,
    Label* failure) {
  // Test that both first and second are sequential one-byte strings.
  // Assume that they are non-smis.
  ld(scratch1, FieldMemOperand(first, HeapObject::kMapOffset));
  ld(scratch2, FieldMemOperand(second, HeapObject::kMapOffset));
  lbu(scratch1, FieldMemOperand(scratch1, Map::kInstanceTypeOffset));
  lbu(scratch2, FieldMemOperand(scratch2, Map::kInstanceTypeOffset));

  JumpIfBothInstanceTypesAreNotSequentialOneByte(scratch1, scratch2, scratch1,
                                                 scratch2, failure);
}


void MacroAssembler::JumpIfNotBothSequentialOneByteStrings(Register first,
                                                           Register second,
                                                           Register scratch1,
                                                           Register scratch2,
                                                           Label* failure) {
  // Check that neither is a smi.
  STATIC_ASSERT(kSmiTag == 0);
  And(scratch1, first, Operand(second));
  JumpIfSmi(scratch1, failure);
  JumpIfNonSmisNotBothSequentialOneByteStrings(first, second, scratch1,
                                               scratch2, failure);
}


void MacroAssembler::JumpIfBothInstanceTypesAreNotSequentialOneByte(
    Register first, Register second, Register scratch1, Register scratch2,
    Label* failure) {
  const int kFlatOneByteStringMask =
      kIsNotStringMask | kStringEncodingMask | kStringRepresentationMask;
  const int kFlatOneByteStringTag =
      kStringTag | kOneByteStringTag | kSeqStringTag;
  DCHECK(kFlatOneByteStringTag <= 0xffff);  // Ensure this fits 16-bit immed.
  andi(scratch1, first, kFlatOneByteStringMask);
  Branch(failure, ne, scratch1, Operand(kFlatOneByteStringTag));
  andi(scratch2, second, kFlatOneByteStringMask);
  Branch(failure, ne, scratch2, Operand(kFlatOneByteStringTag));
}


void MacroAssembler::JumpIfInstanceTypeIsNotSequentialOneByte(Register type,
                                                              Register scratch,
                                                              Label* failure) {
  const int kFlatOneByteStringMask =
      kIsNotStringMask | kStringEncodingMask | kStringRepresentationMask;
  const int kFlatOneByteStringTag =
      kStringTag | kOneByteStringTag | kSeqStringTag;
  And(scratch, type, Operand(kFlatOneByteStringMask));
  Branch(failure, ne, scratch, Operand(kFlatOneByteStringTag));
}


static const int kRegisterPassedArguments = (kMipsAbi == kN64) ? 8 : 4;

int MacroAssembler::CalculateStackPassedWords(int num_reg_arguments,
                                              int num_double_arguments) {
  int stack_passed_words = 0;
  num_reg_arguments += 2 * num_double_arguments;

  // O32: Up to four simple arguments are passed in registers a0..a3.
  // N64: Up to eight simple arguments are passed in registers a0..a7.
  if (num_reg_arguments > kRegisterPassedArguments) {
    stack_passed_words += num_reg_arguments - kRegisterPassedArguments;
  }
  stack_passed_words += kCArgSlotCount;
  return stack_passed_words;
}


void MacroAssembler::EmitSeqStringSetCharCheck(Register string,
                                               Register index,
                                               Register value,
                                               Register scratch,
                                               uint32_t encoding_mask) {
  Label is_object;
  SmiTst(string, at);
  Check(ne, kNonObject, at, Operand(zero_reg));

  ld(at, FieldMemOperand(string, HeapObject::kMapOffset));
  lbu(at, FieldMemOperand(at, Map::kInstanceTypeOffset));

  andi(at, at, kStringRepresentationMask | kStringEncodingMask);
  li(scratch, Operand(encoding_mask));
  Check(eq, kUnexpectedStringType, at, Operand(scratch));

  // TODO(plind): requires Smi size check code for mips32.

  ld(at, FieldMemOperand(string, String::kLengthOffset));
  Check(lt, kIndexIsTooLarge, index, Operand(at));

  DCHECK(Smi::FromInt(0) == 0);
  Check(ge, kIndexIsNegative, index, Operand(zero_reg));
}


void MacroAssembler::PrepareCallCFunction(int num_reg_arguments,
                                          int num_double_arguments,
                                          Register scratch) {
  int frame_alignment = ActivationFrameAlignment();

  // n64: Up to eight simple arguments in a0..a3, a4..a7, No argument slots.
  // O32: Up to four simple arguments are passed in registers a0..a3.
  // Those four arguments must have reserved argument slots on the stack for
  // mips, even though those argument slots are not normally used.
  // Both ABIs: Remaining arguments are pushed on the stack, above (higher
  // address than) the (O32) argument slots. (arg slot calculation handled by
  // CalculateStackPassedWords()).
  int stack_passed_arguments = CalculateStackPassedWords(
      num_reg_arguments, num_double_arguments);
  if (frame_alignment > kPointerSize) {
    // Make stack end at alignment and make room for num_arguments - 4 words
    // and the original value of sp.
    mov(scratch, sp);
    Dsubu(sp, sp, Operand((stack_passed_arguments + 1) * kPointerSize));
    DCHECK(base::bits::IsPowerOfTwo32(frame_alignment));
    And(sp, sp, Operand(-frame_alignment));
    sd(scratch, MemOperand(sp, stack_passed_arguments * kPointerSize));
  } else {
    Dsubu(sp, sp, Operand(stack_passed_arguments * kPointerSize));
  }
}


void MacroAssembler::PrepareCallCFunction(int num_reg_arguments,
                                          Register scratch) {
  PrepareCallCFunction(num_reg_arguments, 0, scratch);
}


void MacroAssembler::CallCFunction(ExternalReference function,
                                   int num_reg_arguments,
                                   int num_double_arguments) {
  li(t8, Operand(function));
  CallCFunctionHelper(t8, num_reg_arguments, num_double_arguments);
}


void MacroAssembler::CallCFunction(Register function,
                                   int num_reg_arguments,
                                   int num_double_arguments) {
  CallCFunctionHelper(function, num_reg_arguments, num_double_arguments);
}


void MacroAssembler::CallCFunction(ExternalReference function,
                                   int num_arguments) {
  CallCFunction(function, num_arguments, 0);
}


void MacroAssembler::CallCFunction(Register function,
                                   int num_arguments) {
  CallCFunction(function, num_arguments, 0);
}


void MacroAssembler::CallCFunctionHelper(Register function,
                                         int num_reg_arguments,
                                         int num_double_arguments) {
  DCHECK(has_frame());
  // Make sure that the stack is aligned before calling a C function unless
  // running in the simulator. The simulator has its own alignment check which
  // provides more information.
  // The argument stots are presumed to have been set up by
  // PrepareCallCFunction. The C function must be called via t9, for mips ABI.

#if V8_HOST_ARCH_MIPS || V8_HOST_ARCH_MIPS64
  if (emit_debug_code()) {
    int frame_alignment = base::OS::ActivationFrameAlignment();
    int frame_alignment_mask = frame_alignment - 1;
    if (frame_alignment > kPointerSize) {
      DCHECK(base::bits::IsPowerOfTwo32(frame_alignment));
      Label alignment_as_expected;
      And(at, sp, Operand(frame_alignment_mask));
      Branch(&alignment_as_expected, eq, at, Operand(zero_reg));
      // Don't use Check here, as it will call Runtime_Abort possibly
      // re-entering here.
      stop("Unexpected alignment in CallCFunction");
      bind(&alignment_as_expected);
    }
  }
#endif  // V8_HOST_ARCH_MIPS

  // Just call directly. The function called cannot cause a GC, or
  // allow preemption, so the return address in the link register
  // stays correct.

  if (!function.is(t9)) {
    mov(t9, function);
    function = t9;
  }

  Call(function);

  int stack_passed_arguments = CalculateStackPassedWords(
      num_reg_arguments, num_double_arguments);

  if (base::OS::ActivationFrameAlignment() > kPointerSize) {
    ld(sp, MemOperand(sp, stack_passed_arguments * kPointerSize));
  } else {
    Daddu(sp, sp, Operand(stack_passed_arguments * kPointerSize));
  }
}


#undef BRANCH_ARGS_CHECK


void MacroAssembler::PatchRelocatedValue(Register li_location,
                                         Register scratch,
                                         Register new_value) {
  lwu(scratch, MemOperand(li_location));
  // At this point scratch is a lui(at, ...) instruction.
  if (emit_debug_code()) {
    And(scratch, scratch, kOpcodeMask);
    Check(eq, kTheInstructionToPatchShouldBeALui,
        scratch, Operand(LUI));
    lwu(scratch, MemOperand(li_location));
  }
  dsrl32(t9, new_value, 0);
  Ins(scratch, t9, 0, kImm16Bits);
  sw(scratch, MemOperand(li_location));

  lwu(scratch, MemOperand(li_location, kInstrSize));
  // scratch is now ori(at, ...).
  if (emit_debug_code()) {
    And(scratch, scratch, kOpcodeMask);
    Check(eq, kTheInstructionToPatchShouldBeAnOri,
        scratch, Operand(ORI));
    lwu(scratch, MemOperand(li_location, kInstrSize));
  }
  dsrl(t9, new_value, kImm16Bits);
  Ins(scratch, t9, 0, kImm16Bits);
  sw(scratch, MemOperand(li_location, kInstrSize));

  lwu(scratch, MemOperand(li_location, kInstrSize * 3));
  // scratch is now ori(at, ...).
  if (emit_debug_code()) {
    And(scratch, scratch, kOpcodeMask);
    Check(eq, kTheInstructionToPatchShouldBeAnOri,
        scratch, Operand(ORI));
    lwu(scratch, MemOperand(li_location, kInstrSize * 3));
  }

  Ins(scratch, new_value, 0, kImm16Bits);
  sw(scratch, MemOperand(li_location, kInstrSize * 3));

  // Update the I-cache so the new lui and ori can be executed.
  FlushICache(li_location, 4);
}

void MacroAssembler::GetRelocatedValue(Register li_location,
                                       Register value,
                                       Register scratch) {
  lwu(value, MemOperand(li_location));
  if (emit_debug_code()) {
    And(value, value, kOpcodeMask);
    Check(eq, kTheInstructionShouldBeALui,
        value, Operand(LUI));
    lwu(value, MemOperand(li_location));
  }

  // value now holds a lui instruction. Extract the immediate.
  andi(value, value, kImm16Mask);
  dsll32(value, value, kImm16Bits);

  lwu(scratch, MemOperand(li_location, kInstrSize));
  if (emit_debug_code()) {
    And(scratch, scratch, kOpcodeMask);
    Check(eq, kTheInstructionShouldBeAnOri,
        scratch, Operand(ORI));
    lwu(scratch, MemOperand(li_location, kInstrSize));
  }
  // "scratch" now holds an ori instruction. Extract the immediate.
  andi(scratch, scratch, kImm16Mask);
  dsll32(scratch, scratch, 0);

  or_(value, value, scratch);

  lwu(scratch, MemOperand(li_location, kInstrSize * 3));
  if (emit_debug_code()) {
    And(scratch, scratch, kOpcodeMask);
    Check(eq, kTheInstructionShouldBeAnOri,
        scratch, Operand(ORI));
    lwu(scratch, MemOperand(li_location, kInstrSize * 3));
  }
  // "scratch" now holds an ori instruction. Extract the immediate.
  andi(scratch, scratch, kImm16Mask);
  dsll(scratch, scratch, kImm16Bits);

  or_(value, value, scratch);
  // Sign extend extracted address.
  dsra(value, value, kImm16Bits);
}


void MacroAssembler::CheckPageFlag(
    Register object,
    Register scratch,
    int mask,
    Condition cc,
    Label* condition_met) {
  And(scratch, object, Operand(~Page::kPageAlignmentMask));
  ld(scratch, MemOperand(scratch, MemoryChunk::kFlagsOffset));
  And(scratch, scratch, Operand(mask));
  Branch(condition_met, cc, scratch, Operand(zero_reg));
}


void MacroAssembler::JumpIfBlack(Register object,
                                 Register scratch0,
                                 Register scratch1,
                                 Label* on_black) {
  HasColor(object, scratch0, scratch1, on_black, 1, 0);  // kBlackBitPattern.
  DCHECK(strcmp(Marking::kBlackBitPattern, "10") == 0);
}


void MacroAssembler::HasColor(Register object,
                              Register bitmap_scratch,
                              Register mask_scratch,
                              Label* has_color,
                              int first_bit,
                              int second_bit) {
  DCHECK(!AreAliased(object, bitmap_scratch, mask_scratch, t8));
  DCHECK(!AreAliased(object, bitmap_scratch, mask_scratch, t9));

  GetMarkBits(object, bitmap_scratch, mask_scratch);

  Label other_color;
  // Note that we are using a 4-byte aligned 8-byte load.
  Uld(t9, MemOperand(bitmap_scratch, MemoryChunk::kHeaderSize));
  And(t8, t9, Operand(mask_scratch));
  Branch(&other_color, first_bit == 1 ? eq : ne, t8, Operand(zero_reg));
  // Shift left 1 by adding.
  Daddu(mask_scratch, mask_scratch, Operand(mask_scratch));
  And(t8, t9, Operand(mask_scratch));
  Branch(has_color, second_bit == 1 ? ne : eq, t8, Operand(zero_reg));

  bind(&other_color);
}


// Detect some, but not all, common pointer-free objects.  This is used by the
// incremental write barrier which doesn't care about oddballs (they are always
// marked black immediately so this code is not hit).
void MacroAssembler::JumpIfDataObject(Register value,
                                      Register scratch,
                                      Label* not_data_object) {
  DCHECK(!AreAliased(value, scratch, t8, no_reg));
  Label is_data_object;
  ld(scratch, FieldMemOperand(value, HeapObject::kMapOffset));
  LoadRoot(t8, Heap::kHeapNumberMapRootIndex);
  Branch(&is_data_object, eq, t8, Operand(scratch));
  DCHECK(kIsIndirectStringTag == 1 && kIsIndirectStringMask == 1);
  DCHECK(kNotStringTag == 0x80 && kIsNotStringMask == 0x80);
  // If it's a string and it's not a cons string then it's an object containing
  // no GC pointers.
  lbu(scratch, FieldMemOperand(scratch, Map::kInstanceTypeOffset));
  And(t8, scratch, Operand(kIsIndirectStringMask | kIsNotStringMask));
  Branch(not_data_object, ne, t8, Operand(zero_reg));
  bind(&is_data_object);
}


void MacroAssembler::GetMarkBits(Register addr_reg,
                                 Register bitmap_reg,
                                 Register mask_reg) {
  DCHECK(!AreAliased(addr_reg, bitmap_reg, mask_reg, no_reg));
  // addr_reg is divided into fields:
  // |63        page base        20|19    high      8|7   shift   3|2  0|
  // 'high' gives the index of the cell holding color bits for the object.
  // 'shift' gives the offset in the cell for this object's color.
  And(bitmap_reg, addr_reg, Operand(~Page::kPageAlignmentMask));
  Ext(mask_reg, addr_reg, kPointerSizeLog2, Bitmap::kBitsPerCellLog2);
  const int kLowBits = kPointerSizeLog2 + Bitmap::kBitsPerCellLog2;
  Ext(t8, addr_reg, kLowBits, kPageSizeBits - kLowBits);
  dsll(t8, t8, Bitmap::kBytesPerCellLog2);
  Daddu(bitmap_reg, bitmap_reg, t8);
  li(t8, Operand(1));
  dsllv(mask_reg, t8, mask_reg);
}


void MacroAssembler::EnsureNotWhite(
    Register value,
    Register bitmap_scratch,
    Register mask_scratch,
    Register load_scratch,
    Label* value_is_white_and_not_data) {
  DCHECK(!AreAliased(value, bitmap_scratch, mask_scratch, t8));
  GetMarkBits(value, bitmap_scratch, mask_scratch);

  // If the value is black or grey we don't need to do anything.
  DCHECK(strcmp(Marking::kWhiteBitPattern, "00") == 0);
  DCHECK(strcmp(Marking::kBlackBitPattern, "10") == 0);
  DCHECK(strcmp(Marking::kGreyBitPattern, "11") == 0);
  DCHECK(strcmp(Marking::kImpossibleBitPattern, "01") == 0);

  Label done;

  // Since both black and grey have a 1 in the first position and white does
  // not have a 1 there we only need to check one bit.
  // Note that we are using a 4-byte aligned 8-byte load.
  Uld(load_scratch, MemOperand(bitmap_scratch, MemoryChunk::kHeaderSize));
  And(t8, mask_scratch, load_scratch);
  Branch(&done, ne, t8, Operand(zero_reg));

  if (emit_debug_code()) {
    // Check for impossible bit pattern.
    Label ok;
    // sll may overflow, making the check conservative.
    dsll(t8, mask_scratch, 1);
    And(t8, load_scratch, t8);
    Branch(&ok, eq, t8, Operand(zero_reg));
    stop("Impossible marking bit pattern");
    bind(&ok);
  }

  // Value is white.  We check whether it is data that doesn't need scanning.
  // Currently only checks for HeapNumber and non-cons strings.
  Register map = load_scratch;  // Holds map while checking type.
  Register length = load_scratch;  // Holds length of object after testing type.
  Label is_data_object;

  // Check for heap-number
  ld(map, FieldMemOperand(value, HeapObject::kMapOffset));
  LoadRoot(t8, Heap::kHeapNumberMapRootIndex);
  {
    Label skip;
    Branch(&skip, ne, t8, Operand(map));
    li(length, HeapNumber::kSize);
    Branch(&is_data_object);
    bind(&skip);
  }

  // Check for strings.
  DCHECK(kIsIndirectStringTag == 1 && kIsIndirectStringMask == 1);
  DCHECK(kNotStringTag == 0x80 && kIsNotStringMask == 0x80);
  // If it's a string and it's not a cons string then it's an object containing
  // no GC pointers.
  Register instance_type = load_scratch;
  lbu(instance_type, FieldMemOperand(map, Map::kInstanceTypeOffset));
  And(t8, instance_type, Operand(kIsIndirectStringMask | kIsNotStringMask));
  Branch(value_is_white_and_not_data, ne, t8, Operand(zero_reg));
  // It's a non-indirect (non-cons and non-slice) string.
  // If it's external, the length is just ExternalString::kSize.
  // Otherwise it's String::kHeaderSize + string->length() * (1 or 2).
  // External strings are the only ones with the kExternalStringTag bit
  // set.
  DCHECK_EQ(0, kSeqStringTag & kExternalStringTag);
  DCHECK_EQ(0, kConsStringTag & kExternalStringTag);
  And(t8, instance_type, Operand(kExternalStringTag));
  {
    Label skip;
    Branch(&skip, eq, t8, Operand(zero_reg));
    li(length, ExternalString::kSize);
    Branch(&is_data_object);
    bind(&skip);
  }

  // Sequential string, either Latin1 or UC16.
  // For Latin1 (char-size of 1) we shift the smi tag away to get the length.
  // For UC16 (char-size of 2) we just leave the smi tag in place, thereby
  // getting the length multiplied by 2.
  DCHECK(kOneByteStringTag == 4 && kStringEncodingMask == 4);
  DCHECK(kSmiTag == 0 && kSmiTagSize == 1);
  lw(t9, UntagSmiFieldMemOperand(value, String::kLengthOffset));
  And(t8, instance_type, Operand(kStringEncodingMask));
  {
    Label skip;
    Branch(&skip, ne, t8, Operand(zero_reg));
    // Adjust length for UC16.
    dsll(t9, t9, 1);
    bind(&skip);
  }
  Daddu(length, t9, Operand(SeqString::kHeaderSize + kObjectAlignmentMask));
  DCHECK(!length.is(t8));
  And(length, length, Operand(~kObjectAlignmentMask));

  bind(&is_data_object);
  // Value is a data object, and it is white.  Mark it black.  Since we know
  // that the object is white we can make it black by flipping one bit.
  Uld(t8, MemOperand(bitmap_scratch, MemoryChunk::kHeaderSize));
  Or(t8, t8, Operand(mask_scratch));
  Usd(t8, MemOperand(bitmap_scratch, MemoryChunk::kHeaderSize));

  And(bitmap_scratch, bitmap_scratch, Operand(~Page::kPageAlignmentMask));
  Uld(t8, MemOperand(bitmap_scratch, MemoryChunk::kLiveBytesOffset));
  Daddu(t8, t8, Operand(length));
  Usd(t8, MemOperand(bitmap_scratch, MemoryChunk::kLiveBytesOffset));

  bind(&done);
}


void MacroAssembler::LoadInstanceDescriptors(Register map,
                                             Register descriptors) {
  ld(descriptors, FieldMemOperand(map, Map::kDescriptorsOffset));
}


void MacroAssembler::NumberOfOwnDescriptors(Register dst, Register map) {
  ld(dst, FieldMemOperand(map, Map::kBitField3Offset));
  DecodeField<Map::NumberOfOwnDescriptorsBits>(dst);
}


void MacroAssembler::EnumLength(Register dst, Register map) {
  STATIC_ASSERT(Map::EnumLengthBits::kShift == 0);
  ld(dst, FieldMemOperand(map, Map::kBitField3Offset));
  And(dst, dst, Operand(Map::EnumLengthBits::kMask));
  SmiTag(dst);
}


void MacroAssembler::CheckEnumCache(Register null_value, Label* call_runtime) {
  Register  empty_fixed_array_value = a6;
  LoadRoot(empty_fixed_array_value, Heap::kEmptyFixedArrayRootIndex);
  Label next, start;
  mov(a2, a0);

  // Check if the enum length field is properly initialized, indicating that
  // there is an enum cache.
  ld(a1, FieldMemOperand(a2, HeapObject::kMapOffset));

  EnumLength(a3, a1);
  Branch(
      call_runtime, eq, a3, Operand(Smi::FromInt(kInvalidEnumCacheSentinel)));

  jmp(&start);

  bind(&next);
  ld(a1, FieldMemOperand(a2, HeapObject::kMapOffset));

  // For all objects but the receiver, check that the cache is empty.
  EnumLength(a3, a1);
  Branch(call_runtime, ne, a3, Operand(Smi::FromInt(0)));

  bind(&start);

  // Check that there are no elements. Register a2 contains the current JS
  // object we've reached through the prototype chain.
  Label no_elements;
  ld(a2, FieldMemOperand(a2, JSObject::kElementsOffset));
  Branch(&no_elements, eq, a2, Operand(empty_fixed_array_value));

  // Second chance, the object may be using the empty slow element dictionary.
  LoadRoot(at, Heap::kEmptySlowElementDictionaryRootIndex);
  Branch(call_runtime, ne, a2, Operand(at));

  bind(&no_elements);
  ld(a2, FieldMemOperand(a1, Map::kPrototypeOffset));
  Branch(&next, ne, a2, Operand(null_value));
}


void MacroAssembler::ClampUint8(Register output_reg, Register input_reg) {
  DCHECK(!output_reg.is(input_reg));
  Label done;
  li(output_reg, Operand(255));
  // Normal branch: nop in delay slot.
  Branch(&done, gt, input_reg, Operand(output_reg));
  // Use delay slot in this branch.
  Branch(USE_DELAY_SLOT, &done, lt, input_reg, Operand(zero_reg));
  mov(output_reg, zero_reg);  // In delay slot.
  mov(output_reg, input_reg);  // Value is in range 0..255.
  bind(&done);
}


void MacroAssembler::ClampDoubleToUint8(Register result_reg,
                                        DoubleRegister input_reg,
                                        DoubleRegister temp_double_reg) {
  Label above_zero;
  Label done;
  Label in_bounds;

  Move(temp_double_reg, 0.0);
  BranchF(&above_zero, NULL, gt, input_reg, temp_double_reg);

  // Double value is less than zero, NaN or Inf, return 0.
  mov(result_reg, zero_reg);
  Branch(&done);

  // Double value is >= 255, return 255.
  bind(&above_zero);
  Move(temp_double_reg, 255.0);
  BranchF(&in_bounds, NULL, le, input_reg, temp_double_reg);
  li(result_reg, Operand(255));
  Branch(&done);

  // In 0-255 range, round and truncate.
  bind(&in_bounds);
  cvt_w_d(temp_double_reg, input_reg);
  mfc1(result_reg, temp_double_reg);
  bind(&done);
}


void MacroAssembler::TestJSArrayForAllocationMemento(
    Register receiver_reg,
    Register scratch_reg,
    Label* no_memento_found,
    Condition cond,
    Label* allocation_memento_present) {
  ExternalReference new_space_start =
      ExternalReference::new_space_start(isolate());
  ExternalReference new_space_allocation_top =
      ExternalReference::new_space_allocation_top_address(isolate());
  Daddu(scratch_reg, receiver_reg,
       Operand(JSArray::kSize + AllocationMemento::kSize - kHeapObjectTag));
  Branch(no_memento_found, lt, scratch_reg, Operand(new_space_start));
  li(at, Operand(new_space_allocation_top));
  ld(at, MemOperand(at));
  Branch(no_memento_found, gt, scratch_reg, Operand(at));
  ld(scratch_reg, MemOperand(scratch_reg, -AllocationMemento::kSize));
  if (allocation_memento_present) {
    Branch(allocation_memento_present, cond, scratch_reg,
           Operand(isolate()->factory()->allocation_memento_map()));
  }
}


Register GetRegisterThatIsNotOneOf(Register reg1,
                                   Register reg2,
                                   Register reg3,
                                   Register reg4,
                                   Register reg5,
                                   Register reg6) {
  RegList regs = 0;
  if (reg1.is_valid()) regs |= reg1.bit();
  if (reg2.is_valid()) regs |= reg2.bit();
  if (reg3.is_valid()) regs |= reg3.bit();
  if (reg4.is_valid()) regs |= reg4.bit();
  if (reg5.is_valid()) regs |= reg5.bit();
  if (reg6.is_valid()) regs |= reg6.bit();

  for (int i = 0; i < Register::NumAllocatableRegisters(); i++) {
    Register candidate = Register::FromAllocationIndex(i);
    if (regs & candidate.bit()) continue;
    return candidate;
  }
  UNREACHABLE();
  return no_reg;
}


void MacroAssembler::JumpIfDictionaryInPrototypeChain(
    Register object,
    Register scratch0,
    Register scratch1,
    Label* found) {
  DCHECK(!scratch1.is(scratch0));
  Factory* factory = isolate()->factory();
  Register current = scratch0;
  Label loop_again;

  // Scratch contained elements pointer.
  Move(current, object);

  // Loop based on the map going up the prototype chain.
  bind(&loop_again);
  ld(current, FieldMemOperand(current, HeapObject::kMapOffset));
  lb(scratch1, FieldMemOperand(current, Map::kBitField2Offset));
  DecodeField<Map::ElementsKindBits>(scratch1);
  Branch(found, eq, scratch1, Operand(DICTIONARY_ELEMENTS));
  ld(current, FieldMemOperand(current, Map::kPrototypeOffset));
  Branch(&loop_again, ne, current, Operand(factory->null_value()));
}


bool AreAliased(Register reg1,
                Register reg2,
                Register reg3,
                Register reg4,
                Register reg5,
                Register reg6,
                Register reg7,
                Register reg8) {
  int n_of_valid_regs = reg1.is_valid() + reg2.is_valid() +
      reg3.is_valid() + reg4.is_valid() + reg5.is_valid() + reg6.is_valid() +
      reg7.is_valid() + reg8.is_valid();

  RegList regs = 0;
  if (reg1.is_valid()) regs |= reg1.bit();
  if (reg2.is_valid()) regs |= reg2.bit();
  if (reg3.is_valid()) regs |= reg3.bit();
  if (reg4.is_valid()) regs |= reg4.bit();
  if (reg5.is_valid()) regs |= reg5.bit();
  if (reg6.is_valid()) regs |= reg6.bit();
  if (reg7.is_valid()) regs |= reg7.bit();
  if (reg8.is_valid()) regs |= reg8.bit();
  int n_of_non_aliasing_regs = NumRegs(regs);

  return n_of_valid_regs != n_of_non_aliasing_regs;
}


CodePatcher::CodePatcher(byte* address,
                         int instructions,
                         FlushICache flush_cache)
    : address_(address),
      size_(instructions * Assembler::kInstrSize),
      masm_(NULL, address, size_ + Assembler::kGap),
      flush_cache_(flush_cache) {
  // Create a new macro assembler pointing to the address of the code to patch.
  // The size is adjusted with kGap on order for the assembler to generate size
  // bytes of instructions without failing with buffer size constraints.
  DCHECK(masm_.reloc_info_writer.pos() == address_ + size_ + Assembler::kGap);
}


CodePatcher::~CodePatcher() {
  // Indicate that code has changed.
  if (flush_cache_ == FLUSH) {
    CpuFeatures::FlushICache(address_, size_);
  }
  // Check that the code was patched as expected.
  DCHECK(masm_.pc_ == address_ + size_);
  DCHECK(masm_.reloc_info_writer.pos() == address_ + size_ + Assembler::kGap);
}


void CodePatcher::Emit(Instr instr) {
  masm()->emit(instr);
}


void CodePatcher::Emit(Address addr) {
  // masm()->emit(reinterpret_cast<Instr>(addr));
}


void CodePatcher::ChangeBranchCondition(Condition cond) {
  Instr instr = Assembler::instr_at(masm_.pc_);
  DCHECK(Assembler::IsBranch(instr));
  uint32_t opcode = Assembler::GetOpcodeField(instr);
  // Currently only the 'eq' and 'ne' cond values are supported and the simple
  // branch instructions (with opcode being the branch type).
  // There are some special cases (see Assembler::IsBranch()) so extending this
  // would be tricky.
  DCHECK(opcode == BEQ ||
         opcode == BNE ||
        opcode == BLEZ ||
        opcode == BGTZ ||
        opcode == BEQL ||
        opcode == BNEL ||
       opcode == BLEZL ||
       opcode == BGTZL);
  opcode = (cond == eq) ? BEQ : BNE;
  instr = (instr & ~kOpcodeMask) | opcode;
  masm_.emit(instr);
}


void MacroAssembler::TruncatingDiv(Register result,
                                   Register dividend,
                                   int32_t divisor) {
  DCHECK(!dividend.is(result));
  DCHECK(!dividend.is(at));
  DCHECK(!result.is(at));
  base::MagicNumbersForDivision<uint32_t> mag =
  base::SignedDivisionByConstant(static_cast<uint32_t>(divisor));
  li(at, Operand(static_cast<int32_t>(mag.multiplier)));
  Mulh(result, dividend, Operand(at));
  bool neg = (mag.multiplier & (static_cast<uint32_t>(1) << 31)) != 0;
  if (divisor > 0 && neg) {
    Addu(result, result, Operand(dividend));
  }
  if (divisor < 0 && !neg && mag.multiplier > 0) {
    Subu(result, result, Operand(dividend));
  }
  if (mag.shift > 0) sra(result, result, mag.shift);
  srl(at, dividend, 31);
  Addu(result, result, Operand(at));
}


} }  // namespace v8::internal

#endif  // V8_TARGET_ARCH_MIPS64