xref: /openjdk10/hotspot/src/share/vm/opto/library_call.cpp

/*
 * Copyright 1999-2007 Sun Microsystems, Inc.  All Rights Reserved.
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 *
 * This code is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 only, as
 * published by the Free Software Foundation.
 *
 * This code is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
 * version 2 for more details (a copy is included in the LICENSE file that
 * accompanied this code).
 *
 * You should have received a copy of the GNU General Public License version
 * 2 along with this work; if not, write to the Free Software Foundation,
 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
 *
 * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
 * CA 95054 USA or visit www.sun.com if you need additional information or
 * have any questions.
 *
 */

#include "incls/_precompiled.incl"
#include "incls/_library_call.cpp.incl"

class LibraryIntrinsic : public InlineCallGenerator {
  // Extend the set of intrinsics known to the runtime:
 public:
 private:
  bool             _is_virtual;
  vmIntrinsics::ID _intrinsic_id;

 public:
  LibraryIntrinsic(ciMethod* m, bool is_virtual, vmIntrinsics::ID id)
    : InlineCallGenerator(m),
      _is_virtual(is_virtual),
      _intrinsic_id(id)
  {
  }
  virtual bool is_intrinsic() const { return true; }
  virtual bool is_virtual()   const { return _is_virtual; }
  virtual JVMState* generate(JVMState* jvms);
  vmIntrinsics::ID intrinsic_id() const { return _intrinsic_id; }
};


// Local helper class for LibraryIntrinsic:
class LibraryCallKit : public GraphKit {
 private:
  LibraryIntrinsic* _intrinsic;   // the library intrinsic being called

 public:
  LibraryCallKit(JVMState* caller, LibraryIntrinsic* intrinsic)
    : GraphKit(caller),
      _intrinsic(intrinsic)
  {
  }

  ciMethod*         caller()    const    { return jvms()->method(); }
  int               bci()       const    { return jvms()->bci(); }
  LibraryIntrinsic* intrinsic() const    { return _intrinsic; }
  vmIntrinsics::ID  intrinsic_id() const { return _intrinsic->intrinsic_id(); }
  ciMethod*         callee()    const    { return _intrinsic->method(); }
  ciSignature*      signature() const    { return callee()->signature(); }
  int               arg_size()  const    { return callee()->arg_size(); }

  bool try_to_inline();

  // Helper functions to inline natives
  void push_result(RegionNode* region, PhiNode* value);
  Node* generate_guard(Node* test, RegionNode* region, float true_prob);
  Node* generate_slow_guard(Node* test, RegionNode* region);
  Node* generate_fair_guard(Node* test, RegionNode* region);
  Node* generate_negative_guard(Node* index, RegionNode* region,
                                // resulting CastII of index:
                                Node* *pos_index = NULL);
  Node* generate_nonpositive_guard(Node* index, bool never_negative,
                                   // resulting CastII of index:
                                   Node* *pos_index = NULL);
  Node* generate_limit_guard(Node* offset, Node* subseq_length,
                             Node* array_length,
                             RegionNode* region);
  Node* generate_current_thread(Node* &tls_output);
  address basictype2arraycopy(BasicType t, Node *src_offset, Node *dest_offset,
                              bool disjoint_bases, const char* &name);
  Node* load_mirror_from_klass(Node* klass);
  Node* load_klass_from_mirror_common(Node* mirror, bool never_see_null,
                                      int nargs,
                                      RegionNode* region, int null_path,
                                      int offset);
  Node* load_klass_from_mirror(Node* mirror, bool never_see_null, int nargs,
                               RegionNode* region, int null_path) {
    int offset = java_lang_Class::klass_offset_in_bytes();
    return load_klass_from_mirror_common(mirror, never_see_null, nargs,
                                         region, null_path,
                                         offset);
  }
  Node* load_array_klass_from_mirror(Node* mirror, bool never_see_null,
                                     int nargs,
                                     RegionNode* region, int null_path) {
    int offset = java_lang_Class::array_klass_offset_in_bytes();
    return load_klass_from_mirror_common(mirror, never_see_null, nargs,
                                         region, null_path,
                                         offset);
  }
  Node* generate_access_flags_guard(Node* kls,
                                    int modifier_mask, int modifier_bits,
                                    RegionNode* region);
  Node* generate_interface_guard(Node* kls, RegionNode* region);
  Node* generate_array_guard(Node* kls, RegionNode* region) {
    return generate_array_guard_common(kls, region, false, false);
  }
  Node* generate_non_array_guard(Node* kls, RegionNode* region) {
    return generate_array_guard_common(kls, region, false, true);
  }
  Node* generate_objArray_guard(Node* kls, RegionNode* region) {
    return generate_array_guard_common(kls, region, true, false);
  }
  Node* generate_non_objArray_guard(Node* kls, RegionNode* region) {
    return generate_array_guard_common(kls, region, true, true);
  }
  Node* generate_array_guard_common(Node* kls, RegionNode* region,
                                    bool obj_array, bool not_array);
  Node* generate_virtual_guard(Node* obj_klass, RegionNode* slow_region);
  CallJavaNode* generate_method_call(vmIntrinsics::ID method_id,
                                     bool is_virtual = false, bool is_static = false);
  CallJavaNode* generate_method_call_static(vmIntrinsics::ID method_id) {
    return generate_method_call(method_id, false, true);
  }
  CallJavaNode* generate_method_call_virtual(vmIntrinsics::ID method_id) {
    return generate_method_call(method_id, true, false);
  }

  bool inline_string_compareTo();
  bool inline_string_indexOf();
  Node* string_indexOf(Node* string_object, ciTypeArray* target_array, jint offset, jint cache_i, jint md2_i);
  Node* pop_math_arg();
  bool runtime_math(const TypeFunc* call_type, address funcAddr, const char* funcName);
  bool inline_math_native(vmIntrinsics::ID id);
  bool inline_trig(vmIntrinsics::ID id);
  bool inline_trans(vmIntrinsics::ID id);
  bool inline_abs(vmIntrinsics::ID id);
  bool inline_sqrt(vmIntrinsics::ID id);
  bool inline_pow(vmIntrinsics::ID id);
  bool inline_exp(vmIntrinsics::ID id);
  bool inline_min_max(vmIntrinsics::ID id);
  Node* generate_min_max(vmIntrinsics::ID id, Node* x, Node* y);
  // This returns Type::AnyPtr, RawPtr, or OopPtr.
  int classify_unsafe_addr(Node* &base, Node* &offset);
  Node* make_unsafe_address(Node* base, Node* offset);
  bool inline_unsafe_access(bool is_native_ptr, bool is_store, BasicType type, bool is_volatile);
  bool inline_unsafe_prefetch(bool is_native_ptr, bool is_store, bool is_static);
  bool inline_unsafe_allocate();
  bool inline_unsafe_copyMemory();
  bool inline_native_currentThread();
  bool inline_native_time_funcs(bool isNano);
  bool inline_native_isInterrupted();
  bool inline_native_Class_query(vmIntrinsics::ID id);
  bool inline_native_subtype_check();

  bool inline_native_newArray();
  bool inline_native_getLength();
  bool inline_array_copyOf(bool is_copyOfRange);
  bool inline_native_clone(bool is_virtual);
  bool inline_native_Reflection_getCallerClass();
  bool inline_native_AtomicLong_get();
  bool inline_native_AtomicLong_attemptUpdate();
  bool is_method_invoke_or_aux_frame(JVMState* jvms);
  // Helper function for inlining native object hash method
  bool inline_native_hashcode(bool is_virtual, bool is_static);
  bool inline_native_getClass();

  // Helper functions for inlining arraycopy
  bool inline_arraycopy();
  void generate_arraycopy(const TypePtr* adr_type,
                          BasicType basic_elem_type,
                          Node* src,  Node* src_offset,
                          Node* dest, Node* dest_offset,
                          Node* copy_length,
                          int nargs,  // arguments on stack for debug info
                          bool disjoint_bases = false,
                          bool length_never_negative = false,
                          RegionNode* slow_region = NULL);
  AllocateArrayNode* tightly_coupled_allocation(Node* ptr,
                                                RegionNode* slow_region);
  void generate_clear_array(const TypePtr* adr_type,
                            Node* dest,
                            BasicType basic_elem_type,
                            Node* slice_off,
                            Node* slice_len,
                            Node* slice_end);
  bool generate_block_arraycopy(const TypePtr* adr_type,
                                BasicType basic_elem_type,
                                AllocateNode* alloc,
                                Node* src,  Node* src_offset,
                                Node* dest, Node* dest_offset,
                                Node* dest_size);
  void generate_slow_arraycopy(const TypePtr* adr_type,
                               Node* src,  Node* src_offset,
                               Node* dest, Node* dest_offset,
                               Node* copy_length,
                               int nargs);
  Node* generate_checkcast_arraycopy(const TypePtr* adr_type,
                                     Node* dest_elem_klass,
                                     Node* src,  Node* src_offset,
                                     Node* dest, Node* dest_offset,
                                     Node* copy_length, int nargs);
  Node* generate_generic_arraycopy(const TypePtr* adr_type,
                                   Node* src,  Node* src_offset,
                                   Node* dest, Node* dest_offset,
                                   Node* copy_length, int nargs);
  void generate_unchecked_arraycopy(const TypePtr* adr_type,
                                    BasicType basic_elem_type,
                                    bool disjoint_bases,
                                    Node* src,  Node* src_offset,
                                    Node* dest, Node* dest_offset,
                                    Node* copy_length);
  bool inline_unsafe_CAS(BasicType type);
  bool inline_unsafe_ordered_store(BasicType type);
  bool inline_fp_conversions(vmIntrinsics::ID id);
  bool inline_reverseBytes(vmIntrinsics::ID id);
};


//---------------------------make_vm_intrinsic----------------------------
CallGenerator* Compile::make_vm_intrinsic(ciMethod* m, bool is_virtual) {
  vmIntrinsics::ID id = m->intrinsic_id();
  assert(id != vmIntrinsics::_none, "must be a VM intrinsic");

  if (DisableIntrinsic[0] != '\0'
      && strstr(DisableIntrinsic, vmIntrinsics::name_at(id)) != NULL) {
    // disabled by a user request on the command line:
    // example: -XX:DisableIntrinsic=_hashCode,_getClass
    return NULL;
  }

  if (!m->is_loaded()) {
    // do not attempt to inline unloaded methods
    return NULL;
  }

  // Only a few intrinsics implement a virtual dispatch.
  // They are expensive calls which are also frequently overridden.
  if (is_virtual) {
    switch (id) {
    case vmIntrinsics::_hashCode:
    case vmIntrinsics::_clone:
      // OK, Object.hashCode and Object.clone intrinsics come in both flavors
      break;
    default:
      return NULL;
    }
  }

  // -XX:-InlineNatives disables nearly all intrinsics:
  if (!InlineNatives) {
    switch (id) {
    case vmIntrinsics::_indexOf:
    case vmIntrinsics::_compareTo:
      break;  // InlineNatives does not control String.compareTo
    default:
      return NULL;
    }
  }

  switch (id) {
  case vmIntrinsics::_compareTo:
    if (!SpecialStringCompareTo)  return NULL;
    break;
  case vmIntrinsics::_indexOf:
    if (!SpecialStringIndexOf)  return NULL;
    break;
  case vmIntrinsics::_arraycopy:
    if (!InlineArrayCopy)  return NULL;
    break;
  case vmIntrinsics::_copyMemory:
    if (StubRoutines::unsafe_arraycopy() == NULL)  return NULL;
    if (!InlineArrayCopy)  return NULL;
    break;
  case vmIntrinsics::_hashCode:
    if (!InlineObjectHash)  return NULL;
    break;
  case vmIntrinsics::_clone:
  case vmIntrinsics::_copyOf:
  case vmIntrinsics::_copyOfRange:
    if (!InlineObjectCopy)  return NULL;
    // These also use the arraycopy intrinsic mechanism:
    if (!InlineArrayCopy)  return NULL;
    break;
  case vmIntrinsics::_checkIndex:
    // We do not intrinsify this.  The optimizer does fine with it.
    return NULL;

  case vmIntrinsics::_get_AtomicLong:
  case vmIntrinsics::_attemptUpdate:
    if (!InlineAtomicLong)  return NULL;
    break;

  case vmIntrinsics::_Object_init:
  case vmIntrinsics::_invoke:
    // We do not intrinsify these; they are marked for other purposes.
    return NULL;

  case vmIntrinsics::_getCallerClass:
    if (!UseNewReflection)  return NULL;
    if (!InlineReflectionGetCallerClass)  return NULL;
    if (!JDK_Version::is_gte_jdk14x_version())  return NULL;
    break;

 default:
    break;
  }

  // -XX:-InlineClassNatives disables natives from the Class class.
  // The flag applies to all reflective calls, notably Array.newArray
  // (visible to Java programmers as Array.newInstance).
  if (m->holder()->name() == ciSymbol::java_lang_Class() ||
      m->holder()->name() == ciSymbol::java_lang_reflect_Array()) {
    if (!InlineClassNatives)  return NULL;
  }

  // -XX:-InlineThreadNatives disables natives from the Thread class.
  if (m->holder()->name() == ciSymbol::java_lang_Thread()) {
    if (!InlineThreadNatives)  return NULL;
  }

  // -XX:-InlineMathNatives disables natives from the Math,Float and Double classes.
  if (m->holder()->name() == ciSymbol::java_lang_Math() ||
      m->holder()->name() == ciSymbol::java_lang_Float() ||
      m->holder()->name() == ciSymbol::java_lang_Double()) {
    if (!InlineMathNatives)  return NULL;
  }

  // -XX:-InlineUnsafeOps disables natives from the Unsafe class.
  if (m->holder()->name() == ciSymbol::sun_misc_Unsafe()) {
    if (!InlineUnsafeOps)  return NULL;
  }

  return new LibraryIntrinsic(m, is_virtual, (vmIntrinsics::ID) id);
}

//----------------------register_library_intrinsics-----------------------
// Initialize this file's data structures, for each Compile instance.
void Compile::register_library_intrinsics() {
  // Nothing to do here.
}

JVMState* LibraryIntrinsic::generate(JVMState* jvms) {
  LibraryCallKit kit(jvms, this);
  Compile* C = kit.C;
  int nodes = C->unique();
#ifndef PRODUCT
  if ((PrintIntrinsics || PrintInlining NOT_PRODUCT( || PrintOptoInlining) ) && Verbose) {
    char buf[1000];
    const char* str = vmIntrinsics::short_name_as_C_string(intrinsic_id(), buf, sizeof(buf));
    tty->print_cr("Intrinsic %s", str);
  }
#endif
  if (kit.try_to_inline()) {
    if (PrintIntrinsics || PrintInlining NOT_PRODUCT( || PrintOptoInlining) ) {
      tty->print("Inlining intrinsic %s%s at bci:%d in",
                 vmIntrinsics::name_at(intrinsic_id()),
                 (is_virtual() ? " (virtual)" : ""), kit.bci());
      kit.caller()->print_short_name(tty);
      tty->print_cr(" (%d bytes)", kit.caller()->code_size());
    }
    C->gather_intrinsic_statistics(intrinsic_id(), is_virtual(), Compile::_intrinsic_worked);
    if (C->log()) {
      C->log()->elem("intrinsic id='%s'%s nodes='%d'",
                     vmIntrinsics::name_at(intrinsic_id()),
                     (is_virtual() ? " virtual='1'" : ""),
                     C->unique() - nodes);
    }
    return kit.transfer_exceptions_into_jvms();
  }

  if (PrintIntrinsics) {
    switch (intrinsic_id()) {
    case vmIntrinsics::_invoke:
    case vmIntrinsics::_Object_init:
      // We do not expect to inline these, so do not produce any noise about them.
      break;
    default:
      tty->print("Did not inline intrinsic %s%s at bci:%d in",
                 vmIntrinsics::name_at(intrinsic_id()),
                 (is_virtual() ? " (virtual)" : ""), kit.bci());
      kit.caller()->print_short_name(tty);
      tty->print_cr(" (%d bytes)", kit.caller()->code_size());
    }
  }
  C->gather_intrinsic_statistics(intrinsic_id(), is_virtual(), Compile::_intrinsic_failed);
  return NULL;
}

bool LibraryCallKit::try_to_inline() {
  // Handle symbolic names for otherwise undistinguished boolean switches:
  const bool is_store       = true;
  const bool is_native_ptr  = true;
  const bool is_static      = true;

  switch (intrinsic_id()) {
  case vmIntrinsics::_hashCode:
    return inline_native_hashcode(intrinsic()->is_virtual(), !is_static);
  case vmIntrinsics::_identityHashCode:
    return inline_native_hashcode(/*!virtual*/ false, is_static);
  case vmIntrinsics::_getClass:
    return inline_native_getClass();

  case vmIntrinsics::_dsin:
  case vmIntrinsics::_dcos:
  case vmIntrinsics::_dtan:
  case vmIntrinsics::_dabs:
  case vmIntrinsics::_datan2:
  case vmIntrinsics::_dsqrt:
  case vmIntrinsics::_dexp:
  case vmIntrinsics::_dlog:
  case vmIntrinsics::_dlog10:
  case vmIntrinsics::_dpow:
    return inline_math_native(intrinsic_id());

  case vmIntrinsics::_min:
  case vmIntrinsics::_max:
    return inline_min_max(intrinsic_id());

  case vmIntrinsics::_arraycopy:
    return inline_arraycopy();

  case vmIntrinsics::_compareTo:
    return inline_string_compareTo();
  case vmIntrinsics::_indexOf:
    return inline_string_indexOf();

  case vmIntrinsics::_getObject:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_OBJECT, false);
  case vmIntrinsics::_getBoolean:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_BOOLEAN, false);
  case vmIntrinsics::_getByte:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_BYTE, false);
  case vmIntrinsics::_getShort:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_SHORT, false);
  case vmIntrinsics::_getChar:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_CHAR, false);
  case vmIntrinsics::_getInt:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_INT, false);
  case vmIntrinsics::_getLong:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_LONG, false);
  case vmIntrinsics::_getFloat:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_FLOAT, false);
  case vmIntrinsics::_getDouble:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_DOUBLE, false);

  case vmIntrinsics::_putObject:
    return inline_unsafe_access(!is_native_ptr, is_store, T_OBJECT, false);
  case vmIntrinsics::_putBoolean:
    return inline_unsafe_access(!is_native_ptr, is_store, T_BOOLEAN, false);
  case vmIntrinsics::_putByte:
    return inline_unsafe_access(!is_native_ptr, is_store, T_BYTE, false);
  case vmIntrinsics::_putShort:
    return inline_unsafe_access(!is_native_ptr, is_store, T_SHORT, false);
  case vmIntrinsics::_putChar:
    return inline_unsafe_access(!is_native_ptr, is_store, T_CHAR, false);
  case vmIntrinsics::_putInt:
    return inline_unsafe_access(!is_native_ptr, is_store, T_INT, false);
  case vmIntrinsics::_putLong:
    return inline_unsafe_access(!is_native_ptr, is_store, T_LONG, false);
  case vmIntrinsics::_putFloat:
    return inline_unsafe_access(!is_native_ptr, is_store, T_FLOAT, false);
  case vmIntrinsics::_putDouble:
    return inline_unsafe_access(!is_native_ptr, is_store, T_DOUBLE, false);

  case vmIntrinsics::_getByte_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_BYTE, false);
  case vmIntrinsics::_getShort_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_SHORT, false);
  case vmIntrinsics::_getChar_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_CHAR, false);
  case vmIntrinsics::_getInt_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_INT, false);
  case vmIntrinsics::_getLong_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_LONG, false);
  case vmIntrinsics::_getFloat_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_FLOAT, false);
  case vmIntrinsics::_getDouble_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_DOUBLE, false);
  case vmIntrinsics::_getAddress_raw:
    return inline_unsafe_access(is_native_ptr, !is_store, T_ADDRESS, false);

  case vmIntrinsics::_putByte_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_BYTE, false);
  case vmIntrinsics::_putShort_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_SHORT, false);
  case vmIntrinsics::_putChar_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_CHAR, false);
  case vmIntrinsics::_putInt_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_INT, false);
  case vmIntrinsics::_putLong_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_LONG, false);
  case vmIntrinsics::_putFloat_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_FLOAT, false);
  case vmIntrinsics::_putDouble_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_DOUBLE, false);
  case vmIntrinsics::_putAddress_raw:
    return inline_unsafe_access(is_native_ptr, is_store, T_ADDRESS, false);

  case vmIntrinsics::_getObjectVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_OBJECT, true);
  case vmIntrinsics::_getBooleanVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_BOOLEAN, true);
  case vmIntrinsics::_getByteVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_BYTE, true);
  case vmIntrinsics::_getShortVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_SHORT, true);
  case vmIntrinsics::_getCharVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_CHAR, true);
  case vmIntrinsics::_getIntVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_INT, true);
  case vmIntrinsics::_getLongVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_LONG, true);
  case vmIntrinsics::_getFloatVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_FLOAT, true);
  case vmIntrinsics::_getDoubleVolatile:
    return inline_unsafe_access(!is_native_ptr, !is_store, T_DOUBLE, true);

  case vmIntrinsics::_putObjectVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_OBJECT, true);
  case vmIntrinsics::_putBooleanVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_BOOLEAN, true);
  case vmIntrinsics::_putByteVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_BYTE, true);
  case vmIntrinsics::_putShortVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_SHORT, true);
  case vmIntrinsics::_putCharVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_CHAR, true);
  case vmIntrinsics::_putIntVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_INT, true);
  case vmIntrinsics::_putLongVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_LONG, true);
  case vmIntrinsics::_putFloatVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_FLOAT, true);
  case vmIntrinsics::_putDoubleVolatile:
    return inline_unsafe_access(!is_native_ptr, is_store, T_DOUBLE, true);

  case vmIntrinsics::_prefetchRead:
    return inline_unsafe_prefetch(!is_native_ptr, !is_store, !is_static);
  case vmIntrinsics::_prefetchWrite:
    return inline_unsafe_prefetch(!is_native_ptr, is_store, !is_static);
  case vmIntrinsics::_prefetchReadStatic:
    return inline_unsafe_prefetch(!is_native_ptr, !is_store, is_static);
  case vmIntrinsics::_prefetchWriteStatic:
    return inline_unsafe_prefetch(!is_native_ptr, is_store, is_static);

  case vmIntrinsics::_compareAndSwapObject:
    return inline_unsafe_CAS(T_OBJECT);
  case vmIntrinsics::_compareAndSwapInt:
    return inline_unsafe_CAS(T_INT);
  case vmIntrinsics::_compareAndSwapLong:
    return inline_unsafe_CAS(T_LONG);

  case vmIntrinsics::_putOrderedObject:
    return inline_unsafe_ordered_store(T_OBJECT);
  case vmIntrinsics::_putOrderedInt:
    return inline_unsafe_ordered_store(T_INT);
  case vmIntrinsics::_putOrderedLong:
    return inline_unsafe_ordered_store(T_LONG);

  case vmIntrinsics::_currentThread:
    return inline_native_currentThread();
  case vmIntrinsics::_isInterrupted:
    return inline_native_isInterrupted();

  case vmIntrinsics::_currentTimeMillis:
    return inline_native_time_funcs(false);
  case vmIntrinsics::_nanoTime:
    return inline_native_time_funcs(true);
  case vmIntrinsics::_allocateInstance:
    return inline_unsafe_allocate();
  case vmIntrinsics::_copyMemory:
    return inline_unsafe_copyMemory();
  case vmIntrinsics::_newArray:
    return inline_native_newArray();
  case vmIntrinsics::_getLength:
    return inline_native_getLength();
  case vmIntrinsics::_copyOf:
    return inline_array_copyOf(false);
  case vmIntrinsics::_copyOfRange:
    return inline_array_copyOf(true);
  case vmIntrinsics::_clone:
    return inline_native_clone(intrinsic()->is_virtual());

  case vmIntrinsics::_isAssignableFrom:
    return inline_native_subtype_check();

  case vmIntrinsics::_isInstance:
  case vmIntrinsics::_getModifiers:
  case vmIntrinsics::_isInterface:
  case vmIntrinsics::_isArray:
  case vmIntrinsics::_isPrimitive:
  case vmIntrinsics::_getSuperclass:
  case vmIntrinsics::_getComponentType:
  case vmIntrinsics::_getClassAccessFlags:
    return inline_native_Class_query(intrinsic_id());

  case vmIntrinsics::_floatToRawIntBits:
  case vmIntrinsics::_floatToIntBits:
  case vmIntrinsics::_intBitsToFloat:
  case vmIntrinsics::_doubleToRawLongBits:
  case vmIntrinsics::_doubleToLongBits:
  case vmIntrinsics::_longBitsToDouble:
    return inline_fp_conversions(intrinsic_id());

  case vmIntrinsics::_reverseBytes_i:
  case vmIntrinsics::_reverseBytes_l:
    return inline_reverseBytes((vmIntrinsics::ID) intrinsic_id());

  case vmIntrinsics::_get_AtomicLong:
    return inline_native_AtomicLong_get();
  case vmIntrinsics::_attemptUpdate:
    return inline_native_AtomicLong_attemptUpdate();

  case vmIntrinsics::_getCallerClass:
    return inline_native_Reflection_getCallerClass();

  default:
    // If you get here, it may be that someone has added a new intrinsic
    // to the list in vmSymbols.hpp without implementing it here.
#ifndef PRODUCT
    if ((PrintMiscellaneous && (Verbose || WizardMode)) || PrintOpto) {
      tty->print_cr("*** Warning: Unimplemented intrinsic %s(%d)",
                    vmIntrinsics::name_at(intrinsic_id()), intrinsic_id());
    }
#endif
    return false;
  }
}

//------------------------------push_result------------------------------
// Helper function for finishing intrinsics.
void LibraryCallKit::push_result(RegionNode* region, PhiNode* value) {
  record_for_igvn(region);
  set_control(_gvn.transform(region));
  BasicType value_type = value->type()->basic_type();
  push_node(value_type, _gvn.transform(value));
}

//------------------------------generate_guard---------------------------
// Helper function for generating guarded fast-slow graph structures.
// The given 'test', if true, guards a slow path.  If the test fails
// then a fast path can be taken.  (We generally hope it fails.)
// In all cases, GraphKit::control() is updated to the fast path.
// The returned value represents the control for the slow path.
// The return value is never 'top'; it is either a valid control
// or NULL if it is obvious that the slow path can never be taken.
// Also, if region and the slow control are not NULL, the slow edge
// is appended to the region.
Node* LibraryCallKit::generate_guard(Node* test, RegionNode* region, float true_prob) {
  if (stopped()) {
    // Already short circuited.
    return NULL;
  }

  // Build an if node and its projections.
  // If test is true we take the slow path, which we assume is uncommon.
  if (_gvn.type(test) == TypeInt::ZERO) {
    // The slow branch is never taken.  No need to build this guard.
    return NULL;
  }

  IfNode* iff = create_and_map_if(control(), test, true_prob, COUNT_UNKNOWN);

  Node* if_slow = _gvn.transform( new (C, 1) IfTrueNode(iff) );
  if (if_slow == top()) {
    // The slow branch is never taken.  No need to build this guard.
    return NULL;
  }

  if (region != NULL)
    region->add_req(if_slow);

  Node* if_fast = _gvn.transform( new (C, 1) IfFalseNode(iff) );
  set_control(if_fast);

  return if_slow;
}

inline Node* LibraryCallKit::generate_slow_guard(Node* test, RegionNode* region) {
  return generate_guard(test, region, PROB_UNLIKELY_MAG(3));
}
inline Node* LibraryCallKit::generate_fair_guard(Node* test, RegionNode* region) {
  return generate_guard(test, region, PROB_FAIR);
}

inline Node* LibraryCallKit::generate_negative_guard(Node* index, RegionNode* region,
                                                     Node* *pos_index) {
  if (stopped())
    return NULL;                // already stopped
  if (_gvn.type(index)->higher_equal(TypeInt::POS)) // [0,maxint]
    return NULL;                // index is already adequately typed
  Node* cmp_lt = _gvn.transform( new (C, 3) CmpINode(index, intcon(0)) );
  Node* bol_lt = _gvn.transform( new (C, 2) BoolNode(cmp_lt, BoolTest::lt) );
  Node* is_neg = generate_guard(bol_lt, region, PROB_MIN);
  if (is_neg != NULL && pos_index != NULL) {
    // Emulate effect of Parse::adjust_map_after_if.
    Node* ccast = new (C, 2) CastIINode(index, TypeInt::POS);
    ccast->set_req(0, control());
    (*pos_index) = _gvn.transform(ccast);
  }
  return is_neg;
}

inline Node* LibraryCallKit::generate_nonpositive_guard(Node* index, bool never_negative,
                                                        Node* *pos_index) {
  if (stopped())
    return NULL;                // already stopped
  if (_gvn.type(index)->higher_equal(TypeInt::POS1)) // [1,maxint]
    return NULL;                // index is already adequately typed
  Node* cmp_le = _gvn.transform( new (C, 3) CmpINode(index, intcon(0)) );
  BoolTest::mask le_or_eq = (never_negative ? BoolTest::eq : BoolTest::le);
  Node* bol_le = _gvn.transform( new (C, 2) BoolNode(cmp_le, le_or_eq) );
  Node* is_notp = generate_guard(bol_le, NULL, PROB_MIN);
  if (is_notp != NULL && pos_index != NULL) {
    // Emulate effect of Parse::adjust_map_after_if.
    Node* ccast = new (C, 2) CastIINode(index, TypeInt::POS1);
    ccast->set_req(0, control());
    (*pos_index) = _gvn.transform(ccast);
  }
  return is_notp;
}

// Make sure that 'position' is a valid limit index, in [0..length].
// There are two equivalent plans for checking this:
//   A. (offset + copyLength)  unsigned<=  arrayLength
//   B. offset  <=  (arrayLength - copyLength)
// We require that all of the values above, except for the sum and
// difference, are already known to be non-negative.
// Plan A is robust in the face of overflow, if offset and copyLength
// are both hugely positive.
//
// Plan B is less direct and intuitive, but it does not overflow at
// all, since the difference of two non-negatives is always
// representable.  Whenever Java methods must perform the equivalent
// check they generally use Plan B instead of Plan A.
// For the moment we use Plan A.
inline Node* LibraryCallKit::generate_limit_guard(Node* offset,
                                                  Node* subseq_length,
                                                  Node* array_length,
                                                  RegionNode* region) {
  if (stopped())
    return NULL;                // already stopped
  bool zero_offset = _gvn.type(offset) == TypeInt::ZERO;
  if (zero_offset && _gvn.eqv_uncast(subseq_length, array_length))
    return NULL;                // common case of whole-array copy
  Node* last = subseq_length;
  if (!zero_offset)             // last += offset
    last = _gvn.transform( new (C, 3) AddINode(last, offset));
  Node* cmp_lt = _gvn.transform( new (C, 3) CmpUNode(array_length, last) );
  Node* bol_lt = _gvn.transform( new (C, 2) BoolNode(cmp_lt, BoolTest::lt) );
  Node* is_over = generate_guard(bol_lt, region, PROB_MIN);
  return is_over;
}


//--------------------------generate_current_thread--------------------
Node* LibraryCallKit::generate_current_thread(Node* &tls_output) {
  ciKlass*    thread_klass = env()->Thread_klass();
  const Type* thread_type  = TypeOopPtr::make_from_klass(thread_klass)->cast_to_ptr_type(TypePtr::NotNull);
  Node* thread = _gvn.transform(new (C, 1) ThreadLocalNode());
  Node* p = basic_plus_adr(top()/*!oop*/, thread, in_bytes(JavaThread::threadObj_offset()));
  Node* threadObj = make_load(NULL, p, thread_type, T_OBJECT);
  tls_output = thread;
  return threadObj;
}


//------------------------------inline_string_compareTo------------------------
bool LibraryCallKit::inline_string_compareTo() {

  const int value_offset = java_lang_String::value_offset_in_bytes();
  const int count_offset = java_lang_String::count_offset_in_bytes();
  const int offset_offset = java_lang_String::offset_offset_in_bytes();

  _sp += 2;
  Node *argument = pop();  // pop non-receiver first:  it was pushed second
  Node *receiver = pop();

  // Null check on self without removing any arguments.  The argument
  // null check technically happens in the wrong place, which can lead to
  // invalid stack traces when string compare is inlined into a method
  // which handles NullPointerExceptions.
  _sp += 2;
  receiver = do_null_check(receiver, T_OBJECT);
  argument = do_null_check(argument, T_OBJECT);
  _sp -= 2;
  if (stopped()) {
    return true;
  }

  ciInstanceKlass* klass = env()->String_klass();
  const TypeInstPtr* string_type =
    TypeInstPtr::make(TypePtr::BotPTR, klass, false, NULL, 0);

  Node* compare =
    _gvn.transform(new (C, 7) StrCompNode(
                        control(),
                        memory(TypeAryPtr::CHARS),
                        memory(string_type->add_offset(value_offset)),
                        memory(string_type->add_offset(count_offset)),
                        memory(string_type->add_offset(offset_offset)),
                        receiver,
                        argument));
  push(compare);
  return true;
}

// Java version of String.indexOf(constant string)
// class StringDecl {
//   StringDecl(char[] ca) {
//     offset = 0;
//     count = ca.length;
//     value = ca;
//   }
//   int offset;
//   int count;
//   char[] value;
// }
//
// static int string_indexOf_J(StringDecl string_object, char[] target_object,
//                             int targetOffset, int cache_i, int md2) {
//   int cache = cache_i;
//   int sourceOffset = string_object.offset;
//   int sourceCount = string_object.count;
//   int targetCount = target_object.length;
//
//   int targetCountLess1 = targetCount - 1;
//   int sourceEnd = sourceOffset + sourceCount - targetCountLess1;
//
//   char[] source = string_object.value;
//   char[] target = target_object;
//   int lastChar = target[targetCountLess1];
//
//  outer_loop:
//   for (int i = sourceOffset; i < sourceEnd; ) {
//     int src = source[i + targetCountLess1];
//     if (src == lastChar) {
//       // With random strings and a 4-character alphabet,
//       // reverse matching at this point sets up 0.8% fewer
//       // frames, but (paradoxically) makes 0.3% more probes.
//       // Since those probes are nearer the lastChar probe,
//       // there is may be a net D$ win with reverse matching.
//       // But, reversing loop inhibits unroll of inner loop
//       // for unknown reason.  So, does running outer loop from
//       // (sourceOffset - targetCountLess1) to (sourceOffset + sourceCount)
//       for (int j = 0; j < targetCountLess1; j++) {
//         if (target[targetOffset + j] != source[i+j]) {
//           if ((cache & (1 << source[i+j])) == 0) {
//             if (md2 < j+1) {
//               i += j+1;
//               continue outer_loop;
//             }
//           }
//           i += md2;
//           continue outer_loop;
//         }
//       }
//       return i - sourceOffset;
//     }
//     if ((cache & (1 << src)) == 0) {
//       i += targetCountLess1;
//     } // using "i += targetCount;" and an "else i++;" causes a jump to jump.
//     i++;
//   }
//   return -1;
// }

//------------------------------string_indexOf------------------------
Node* LibraryCallKit::string_indexOf(Node* string_object, ciTypeArray* target_array, jint targetOffset_i,
                                     jint cache_i, jint md2_i) {

  Node* no_ctrl  = NULL;
  float likely   = PROB_LIKELY(0.9);
  float unlikely = PROB_UNLIKELY(0.9);

  const int value_offset  = java_lang_String::value_offset_in_bytes();
  const int count_offset  = java_lang_String::count_offset_in_bytes();
  const int offset_offset = java_lang_String::offset_offset_in_bytes();

  ciInstanceKlass* klass = env()->String_klass();
  const TypeInstPtr* string_type = TypeInstPtr::make(TypePtr::BotPTR, klass, false, NULL, 0);
  const TypeAryPtr*  source_type = TypeAryPtr::make(TypePtr::NotNull, TypeAry::make(TypeInt::CHAR,TypeInt::POS), ciTypeArrayKlass::make(T_CHAR), true, 0);

  Node* sourceOffseta = basic_plus_adr(string_object, string_object, offset_offset);
  Node* sourceOffset  = make_load(no_ctrl, sourceOffseta, TypeInt::INT, T_INT, string_type->add_offset(offset_offset));
  Node* sourceCounta  = basic_plus_adr(string_object, string_object, count_offset);
  Node* sourceCount   = make_load(no_ctrl, sourceCounta, TypeInt::INT, T_INT, string_type->add_offset(count_offset));
  Node* sourcea       = basic_plus_adr(string_object, string_object, value_offset);
  Node* source        = make_load(no_ctrl, sourcea, source_type, T_OBJECT, string_type->add_offset(value_offset));

  Node* target = _gvn.transform(ConPNode::make(C, target_array));
  jint target_length = target_array->length();
  const TypeAry* target_array_type = TypeAry::make(TypeInt::CHAR, TypeInt::make(0, target_length, Type::WidenMin));
  const TypeAryPtr* target_type = TypeAryPtr::make(TypePtr::BotPTR, target_array_type, target_array->klass(), true, Type::OffsetBot);

  IdealKit kit(gvn(), control(), merged_memory());
#define __ kit.
  Node* zero             = __ ConI(0);
  Node* one              = __ ConI(1);
  Node* cache            = __ ConI(cache_i);
  Node* md2              = __ ConI(md2_i);
  Node* lastChar         = __ ConI(target_array->char_at(target_length - 1));
  Node* targetCount      = __ ConI(target_length);
  Node* targetCountLess1 = __ ConI(target_length - 1);
  Node* targetOffset     = __ ConI(targetOffset_i);
  Node* sourceEnd        = __ SubI(__ AddI(sourceOffset, sourceCount), targetCountLess1);

  IdealVariable rtn(kit), i(kit), j(kit); __ declares_done();
  Node* outer_loop = __ make_label(2 /* goto */);
  Node* return_    = __ make_label(1);

  __ set(rtn,__ ConI(-1));
  __ loop(i, sourceOffset, BoolTest::lt, sourceEnd); {
       Node* i2  = __ AddI(__ value(i), targetCountLess1);
       // pin to prohibit loading of "next iteration" value which may SEGV (rare)
       Node* src = load_array_element(__ ctrl(), source, i2, TypeAryPtr::CHARS);
       __ if_then(src, BoolTest::eq, lastChar, unlikely); {
         __ loop(j, zero, BoolTest::lt, targetCountLess1); {
              Node* tpj = __ AddI(targetOffset, __ value(j));
              Node* targ = load_array_element(no_ctrl, target, tpj, target_type);
              Node* ipj  = __ AddI(__ value(i), __ value(j));
              Node* src2 = load_array_element(no_ctrl, source, ipj, TypeAryPtr::CHARS);
              __ if_then(targ, BoolTest::ne, src2); {
                __ if_then(__ AndI(cache, __ LShiftI(one, src2)), BoolTest::eq, zero); {
                  __ if_then(md2, BoolTest::lt, __ AddI(__ value(j), one)); {
                    __ increment(i, __ AddI(__ value(j), one));
                    __ goto_(outer_loop);
                  } __ end_if(); __ dead(j);
                }__ end_if(); __ dead(j);
                __ increment(i, md2);
                __ goto_(outer_loop);
              }__ end_if();
              __ increment(j, one);
         }__ end_loop(); __ dead(j);
         __ set(rtn, __ SubI(__ value(i), sourceOffset)); __ dead(i);
         __ goto_(return_);
       }__ end_if();
       __ if_then(__ AndI(cache, __ LShiftI(one, src)), BoolTest::eq, zero, likely); {
         __ increment(i, targetCountLess1);
       }__ end_if();
       __ increment(i, one);
       __ bind(outer_loop);
  }__ end_loop(); __ dead(i);
  __ bind(return_);
  __ drain_delay_transform();

  set_control(__ ctrl());
  Node* result = __ value(rtn);
#undef __
  C->set_has_loops(true);
  return result;
}


//------------------------------inline_string_indexOf------------------------
bool LibraryCallKit::inline_string_indexOf() {

  _sp += 2;
  Node *argument = pop();  // pop non-receiver first:  it was pushed second
  Node *receiver = pop();

  // don't intrinsify is argument isn't a constant string.
  if (!argument->is_Con()) {
    return false;
  }
  const TypeOopPtr* str_type = _gvn.type(argument)->isa_oopptr();
  if (str_type == NULL) {
    return false;
  }
  ciInstanceKlass* klass = env()->String_klass();
  ciObject* str_const = str_type->const_oop();
  if (str_const == NULL || str_const->klass() != klass) {
    return false;
  }
  ciInstance* str = str_const->as_instance();
  assert(str != NULL, "must be instance");

  const int value_offset  = java_lang_String::value_offset_in_bytes();
  const int count_offset  = java_lang_String::count_offset_in_bytes();
  const int offset_offset = java_lang_String::offset_offset_in_bytes();

  ciObject* v = str->field_value_by_offset(value_offset).as_object();
  int       o = str->field_value_by_offset(offset_offset).as_int();
  int       c = str->field_value_by_offset(count_offset).as_int();
  ciTypeArray* pat = v->as_type_array(); // pattern (argument) character array

  // constant strings have no offset and count == length which
  // simplifies the resulting code somewhat so lets optimize for that.
  if (o != 0 || c != pat->length()) {
    return false;
  }

  // Null check on self without removing any arguments.  The argument
  // null check technically happens in the wrong place, which can lead to
  // invalid stack traces when string compare is inlined into a method
  // which handles NullPointerExceptions.
  _sp += 2;
  receiver = do_null_check(receiver, T_OBJECT);
  // No null check on the argument is needed since it's a constant String oop.
  _sp -= 2;
  if (stopped()) {
    return true;
  }

  // The null string as a pattern always returns 0 (match at beginning of string)
  if (c == 0) {
    push(intcon(0));
    return true;
  }

  jchar lastChar = pat->char_at(o + (c - 1));
  int cache = 0;
  int i;
  for (i = 0; i < c - 1; i++) {
    assert(i < pat->length(), "out of range");
    cache |= (1 << (pat->char_at(o + i) & (sizeof(cache) * BitsPerByte - 1)));
  }

  int md2 = c;
  for (i = 0; i < c - 1; i++) {
    assert(i < pat->length(), "out of range");
    if (pat->char_at(o + i) == lastChar) {
      md2 = (c - 1) - i;
    }
  }

  Node* result = string_indexOf(receiver, pat, o, cache, md2);
  push(result);
  return true;
}

//--------------------------pop_math_arg--------------------------------
// Pop a double argument to a math function from the stack
// rounding it if necessary.
Node * LibraryCallKit::pop_math_arg() {
  Node *arg = pop_pair();
  if( Matcher::strict_fp_requires_explicit_rounding && UseSSE<=1 )
    arg = _gvn.transform( new (C, 2) RoundDoubleNode(0, arg) );
  return arg;
}

//------------------------------inline_trig----------------------------------
// Inline sin/cos/tan instructions, if possible.  If rounding is required, do
// argument reduction which will turn into a fast/slow diamond.
bool LibraryCallKit::inline_trig(vmIntrinsics::ID id) {
  _sp += arg_size();            // restore stack pointer
  Node* arg = pop_math_arg();
  Node* trig = NULL;

  switch (id) {
  case vmIntrinsics::_dsin:
    trig = _gvn.transform((Node*)new (C, 2) SinDNode(arg));
    break;
  case vmIntrinsics::_dcos:
    trig = _gvn.transform((Node*)new (C, 2) CosDNode(arg));
    break;
  case vmIntrinsics::_dtan:
    trig = _gvn.transform((Node*)new (C, 2) TanDNode(arg));
    break;
  default:
    assert(false, "bad intrinsic was passed in");
    return false;
  }

  // Rounding required?  Check for argument reduction!
  if( Matcher::strict_fp_requires_explicit_rounding ) {

    static const double     pi_4 =  0.7853981633974483;
    static const double neg_pi_4 = -0.7853981633974483;
    // pi/2 in 80-bit extended precision
    // static const unsigned char pi_2_bits_x[] = {0x35,0xc2,0x68,0x21,0xa2,0xda,0x0f,0xc9,0xff,0x3f,0x00,0x00,0x00,0x00,0x00,0x00};
    // -pi/2 in 80-bit extended precision
    // static const unsigned char neg_pi_2_bits_x[] = {0x35,0xc2,0x68,0x21,0xa2,0xda,0x0f,0xc9,0xff,0xbf,0x00,0x00,0x00,0x00,0x00,0x00};
    // Cutoff value for using this argument reduction technique
    //static const double    pi_2_minus_epsilon =  1.564660403643354;
    //static const double neg_pi_2_plus_epsilon = -1.564660403643354;

    // Pseudocode for sin:
    // if (x <= Math.PI / 4.0) {
    //   if (x >= -Math.PI / 4.0) return  fsin(x);
    //   if (x >= -Math.PI / 2.0) return -fcos(x + Math.PI / 2.0);
    // } else {
    //   if (x <=  Math.PI / 2.0) return  fcos(x - Math.PI / 2.0);
    // }
    // return StrictMath.sin(x);

    // Pseudocode for cos:
    // if (x <= Math.PI / 4.0) {
    //   if (x >= -Math.PI / 4.0) return  fcos(x);
    //   if (x >= -Math.PI / 2.0) return  fsin(x + Math.PI / 2.0);
    // } else {
    //   if (x <=  Math.PI / 2.0) return -fsin(x - Math.PI / 2.0);
    // }
    // return StrictMath.cos(x);

    // Actually, sticking in an 80-bit Intel value into C2 will be tough; it
    // requires a special machine instruction to load it.  Instead we'll try
    // the 'easy' case.  If we really need the extra range +/- PI/2 we'll
    // probably do the math inside the SIN encoding.

    // Make the merge point
    RegionNode *r = new (C, 3) RegionNode(3);
    Node *phi = new (C, 3) PhiNode(r,Type::DOUBLE);

    // Flatten arg so we need only 1 test
    Node *abs = _gvn.transform(new (C, 2) AbsDNode(arg));
    // Node for PI/4 constant
    Node *pi4 = makecon(TypeD::make(pi_4));
    // Check PI/4 : abs(arg)
    Node *cmp = _gvn.transform(new (C, 3) CmpDNode(pi4,abs));
    // Check: If PI/4 < abs(arg) then go slow
    Node *bol = _gvn.transform( new (C, 2) BoolNode( cmp, BoolTest::lt ) );
    // Branch either way
    IfNode *iff = create_and_xform_if(control(),bol, PROB_STATIC_FREQUENT, COUNT_UNKNOWN);
    set_control(opt_iff(r,iff));

    // Set fast path result
    phi->init_req(2,trig);

    // Slow path - non-blocking leaf call
    Node* call = NULL;
    switch (id) {
    case vmIntrinsics::_dsin:
      call = make_runtime_call(RC_LEAF, OptoRuntime::Math_D_D_Type(),
                               CAST_FROM_FN_PTR(address, SharedRuntime::dsin),
                               "Sin", NULL, arg, top());
      break;
    case vmIntrinsics::_dcos:
      call = make_runtime_call(RC_LEAF, OptoRuntime::Math_D_D_Type(),
                               CAST_FROM_FN_PTR(address, SharedRuntime::dcos),
                               "Cos", NULL, arg, top());
      break;
    case vmIntrinsics::_dtan:
      call = make_runtime_call(RC_LEAF, OptoRuntime::Math_D_D_Type(),
                               CAST_FROM_FN_PTR(address, SharedRuntime::dtan),
                               "Tan", NULL, arg, top());
      break;
    }
    assert(control()->in(0) == call, "");
    Node* slow_result = _gvn.transform(new (C, 1) ProjNode(call,TypeFunc::Parms));
    r->init_req(1,control());
    phi->init_req(1,slow_result);

    // Post-merge
    set_control(_gvn.transform(r));
    record_for_igvn(r);
    trig = _gvn.transform(phi);

    C->set_has_split_ifs(true); // Has chance for split-if optimization
  }
  // Push result back on JVM stack
  push_pair(trig);
  return true;
}

//------------------------------inline_sqrt-------------------------------------
// Inline square root instruction, if possible.
bool LibraryCallKit::inline_sqrt(vmIntrinsics::ID id) {
  assert(id == vmIntrinsics::_dsqrt, "Not square root");
  _sp += arg_size();        // restore stack pointer
  push_pair(_gvn.transform(new (C, 2) SqrtDNode(0, pop_math_arg())));
  return true;
}

//------------------------------inline_abs-------------------------------------
// Inline absolute value instruction, if possible.
bool LibraryCallKit::inline_abs(vmIntrinsics::ID id) {
  assert(id == vmIntrinsics::_dabs, "Not absolute value");
  _sp += arg_size();        // restore stack pointer
  push_pair(_gvn.transform(new (C, 2) AbsDNode(pop_math_arg())));
  return true;
}

//------------------------------inline_exp-------------------------------------
// Inline exp instructions, if possible.  The Intel hardware only misses
// really odd corner cases (+/- Infinity).  Just uncommon-trap them.
bool LibraryCallKit::inline_exp(vmIntrinsics::ID id) {
  assert(id == vmIntrinsics::_dexp, "Not exp");

  // If this inlining ever returned NaN in the past, we do not intrinsify it
  // every again.  NaN results requires StrictMath.exp handling.
  if (too_many_traps(Deoptimization::Reason_intrinsic))  return false;

  // Do not intrinsify on older platforms which lack cmove.
  if (ConditionalMoveLimit == 0)  return false;

  _sp += arg_size();        // restore stack pointer
  Node *x = pop_math_arg();
  Node *result = _gvn.transform(new (C, 2) ExpDNode(0,x));

  //-------------------
  //result=(result.isNaN())? StrictMath::exp():result;
  // Check: If isNaN() by checking result!=result? then go to Strict Math
  Node* cmpisnan = _gvn.transform(new (C, 3) CmpDNode(result,result));
  // Build the boolean node
  Node* bolisnum = _gvn.transform( new (C, 2) BoolNode(cmpisnan, BoolTest::eq) );

  { BuildCutout unless(this, bolisnum, PROB_STATIC_FREQUENT);
    // End the current control-flow path
    push_pair(x);
    // Math.exp intrinsic returned a NaN, which requires StrictMath.exp
    // to handle.  Recompile without intrinsifying Math.exp
    uncommon_trap(Deoptimization::Reason_intrinsic,
                  Deoptimization::Action_make_not_entrant);
  }

  C->set_has_split_ifs(true); // Has chance for split-if optimization

  push_pair(result);

  return true;
}

//------------------------------inline_pow-------------------------------------
// Inline power instructions, if possible.
bool LibraryCallKit::inline_pow(vmIntrinsics::ID id) {
  assert(id == vmIntrinsics::_dpow, "Not pow");

  // If this inlining ever returned NaN in the past, we do not intrinsify it
  // every again.  NaN results requires StrictMath.pow handling.
  if (too_many_traps(Deoptimization::Reason_intrinsic))  return false;

  // Do not intrinsify on older platforms which lack cmove.
  if (ConditionalMoveLimit == 0)  return false;

  // Pseudocode for pow
  // if (x <= 0.0) {
  //   if ((double)((int)y)==y) { // if y is int
  //     result = ((1&(int)y)==0)?-DPow(abs(x), y):DPow(abs(x), y)
  //   } else {
  //     result = NaN;
  //   }
  // } else {
  //   result = DPow(x,y);
  // }
  // if (result != result)?  {
  //   ucommon_trap();
  // }
  // return result;

  _sp += arg_size();        // restore stack pointer
  Node* y = pop_math_arg();
  Node* x = pop_math_arg();

  Node *fast_result = _gvn.transform( new (C, 3) PowDNode(0, x, y) );

  // Short form: if not top-level (i.e., Math.pow but inlining Math.pow
  // inside of something) then skip the fancy tests and just check for
  // NaN result.
  Node *result = NULL;
  if( jvms()->depth() >= 1 ) {
    result = fast_result;
  } else {

    // Set the merge point for If node with condition of (x <= 0.0)
    // There are four possible paths to region node and phi node
    RegionNode *r = new (C, 4) RegionNode(4);
    Node *phi = new (C, 4) PhiNode(r, Type::DOUBLE);

    // Build the first if node: if (x <= 0.0)
    // Node for 0 constant
    Node *zeronode = makecon(TypeD::ZERO);
    // Check x:0
    Node *cmp = _gvn.transform(new (C, 3) CmpDNode(x, zeronode));
    // Check: If (x<=0) then go complex path
    Node *bol1 = _gvn.transform( new (C, 2) BoolNode( cmp, BoolTest::le ) );
    // Branch either way
    IfNode *if1 = create_and_xform_if(control(),bol1, PROB_STATIC_INFREQUENT, COUNT_UNKNOWN);
    Node *opt_test = _gvn.transform(if1);
    //assert( opt_test->is_If(), "Expect an IfNode");
    IfNode *opt_if1 = (IfNode*)opt_test;
    // Fast path taken; set region slot 3
    Node *fast_taken = _gvn.transform( new (C, 1) IfFalseNode(opt_if1) );
    r->init_req(3,fast_taken); // Capture fast-control

    // Fast path not-taken, i.e. slow path
    Node *complex_path = _gvn.transform( new (C, 1) IfTrueNode(opt_if1) );

    // Set fast path result
    Node *fast_result = _gvn.transform( new (C, 3) PowDNode(0, y, x) );
    phi->init_req(3, fast_result);

    // Complex path
    // Build the second if node (if y is int)
    // Node for (int)y
    Node *inty = _gvn.transform( new (C, 2) ConvD2INode(y));
    // Node for (double)((int) y)
    Node *doubleinty= _gvn.transform( new (C, 2) ConvI2DNode(inty));
    // Check (double)((int) y) : y
    Node *cmpinty= _gvn.transform(new (C, 3) CmpDNode(doubleinty, y));
    // Check if (y isn't int) then go to slow path

    Node *bol2 = _gvn.transform( new (C, 2) BoolNode( cmpinty, BoolTest::ne ) );
    // Branch eith way
    IfNode *if2 = create_and_xform_if(complex_path,bol2, PROB_STATIC_INFREQUENT, COUNT_UNKNOWN);
    Node *slow_path = opt_iff(r,if2); // Set region path 2

    // Calculate DPow(abs(x), y)*(1 & (int)y)
    // Node for constant 1
    Node *conone = intcon(1);
    // 1& (int)y
    Node *signnode= _gvn.transform( new (C, 3) AndINode(conone, inty) );
    // zero node
    Node *conzero = intcon(0);
    // Check (1&(int)y)==0?
    Node *cmpeq1 = _gvn.transform(new (C, 3) CmpINode(signnode, conzero));
    // Check if (1&(int)y)!=0?, if so the result is negative
    Node *bol3 = _gvn.transform( new (C, 2) BoolNode( cmpeq1, BoolTest::ne ) );
    // abs(x)
    Node *absx=_gvn.transform( new (C, 2) AbsDNode(x));
    // abs(x)^y
    Node *absxpowy = _gvn.transform( new (C, 3) PowDNode(0, y, absx) );
    // -abs(x)^y
    Node *negabsxpowy = _gvn.transform(new (C, 2) NegDNode (absxpowy));
    // (1&(int)y)==1?-DPow(abs(x), y):DPow(abs(x), y)
    Node *signresult = _gvn.transform( CMoveNode::make(C, NULL, bol3, absxpowy, negabsxpowy, Type::DOUBLE));
    // Set complex path fast result
    phi->init_req(2, signresult);

    static const jlong nan_bits = CONST64(0x7ff8000000000000);
    Node *slow_result = makecon(TypeD::make(*(double*)&nan_bits)); // return NaN
    r->init_req(1,slow_path);
    phi->init_req(1,slow_result);

    // Post merge
    set_control(_gvn.transform(r));
    record_for_igvn(r);
    result=_gvn.transform(phi);
  }

  //-------------------
  //result=(result.isNaN())? uncommon_trap():result;
  // Check: If isNaN() by checking result!=result? then go to Strict Math
  Node* cmpisnan = _gvn.transform(new (C, 3) CmpDNode(result,result));
  // Build the boolean node
  Node* bolisnum = _gvn.transform( new (C, 2) BoolNode(cmpisnan, BoolTest::eq) );

  { BuildCutout unless(this, bolisnum, PROB_STATIC_FREQUENT);
    // End the current control-flow path
    push_pair(x);
    push_pair(y);
    // Math.pow intrinsic returned a NaN, which requires StrictMath.pow
    // to handle.  Recompile without intrinsifying Math.pow.
    uncommon_trap(Deoptimization::Reason_intrinsic,
                  Deoptimization::Action_make_not_entrant);
  }

  C->set_has_split_ifs(true); // Has chance for split-if optimization

  push_pair(result);

  return true;
}

//------------------------------inline_trans-------------------------------------
// Inline transcendental instructions, if possible.  The Intel hardware gets
// these right, no funny corner cases missed.
bool LibraryCallKit::inline_trans(vmIntrinsics::ID id) {
  _sp += arg_size();        // restore stack pointer
  Node* arg = pop_math_arg();
  Node* trans = NULL;

  switch (id) {
  case vmIntrinsics::_dlog:
    trans = _gvn.transform((Node*)new (C, 2) LogDNode(arg));
    break;
  case vmIntrinsics::_dlog10:
    trans = _gvn.transform((Node*)new (C, 2) Log10DNode(arg));
    break;
  default:
    assert(false, "bad intrinsic was passed in");
    return false;
  }

  // Push result back on JVM stack
  push_pair(trans);
  return true;
}

//------------------------------runtime_math-----------------------------
bool LibraryCallKit::runtime_math(const TypeFunc* call_type, address funcAddr, const char* funcName) {
  Node* a = NULL;
  Node* b = NULL;

  assert(call_type == OptoRuntime::Math_DD_D_Type() || call_type == OptoRuntime::Math_D_D_Type(),
         "must be (DD)D or (D)D type");

  // Inputs
  _sp += arg_size();        // restore stack pointer
  if (call_type == OptoRuntime::Math_DD_D_Type()) {
    b = pop_math_arg();
  }
  a = pop_math_arg();

  const TypePtr* no_memory_effects = NULL;
  Node* trig = make_runtime_call(RC_LEAF, call_type, funcAddr, funcName,
                                 no_memory_effects,
                                 a, top(), b, b ? top() : NULL);
  Node* value = _gvn.transform(new (C, 1) ProjNode(trig, TypeFunc::Parms+0));
#ifdef ASSERT
  Node* value_top = _gvn.transform(new (C, 1) ProjNode(trig, TypeFunc::Parms+1));
  assert(value_top == top(), "second value must be top");
#endif

  push_pair(value);
  return true;
}

//------------------------------inline_math_native-----------------------------
bool LibraryCallKit::inline_math_native(vmIntrinsics::ID id) {
  switch (id) {
    // These intrinsics are not properly supported on all hardware
  case vmIntrinsics::_dcos: return Matcher::has_match_rule(Op_CosD) ? inline_trig(id) :
    runtime_math(OptoRuntime::Math_D_D_Type(), CAST_FROM_FN_PTR(address, SharedRuntime::dcos), "COS");
  case vmIntrinsics::_dsin: return Matcher::has_match_rule(Op_SinD) ? inline_trig(id) :
    runtime_math(OptoRuntime::Math_D_D_Type(), CAST_FROM_FN_PTR(address, SharedRuntime::dsin), "SIN");
  case vmIntrinsics::_dtan: return Matcher::has_match_rule(Op_TanD) ? inline_trig(id) :
    runtime_math(OptoRuntime::Math_D_D_Type(), CAST_FROM_FN_PTR(address, SharedRuntime::dtan), "TAN");

  case vmIntrinsics::_dlog:   return Matcher::has_match_rule(Op_LogD) ? inline_trans(id) :
    runtime_math(OptoRuntime::Math_D_D_Type(), CAST_FROM_FN_PTR(address, SharedRuntime::dlog), "LOG");
  case vmIntrinsics::_dlog10: return Matcher::has_match_rule(Op_Log10D) ? inline_trans(id) :
    runtime_math(OptoRuntime::Math_D_D_Type(), CAST_FROM_FN_PTR(address, SharedRuntime::dlog10), "LOG10");

    // These intrinsics are supported on all hardware
  case vmIntrinsics::_dsqrt: return Matcher::has_match_rule(Op_SqrtD) ? inline_sqrt(id) : false;
  case vmIntrinsics::_dabs:  return Matcher::has_match_rule(Op_AbsD)  ? inline_abs(id)  : false;

    // These intrinsics don't work on X86.  The ad implementation doesn't
    // handle NaN's properly.  Instead of returning infinity, the ad
    // implementation returns a NaN on overflow. See bug: 6304089
    // Once the ad implementations are fixed, change the code below
    // to match the intrinsics above

  case vmIntrinsics::_dexp:  return
    runtime_math(OptoRuntime::Math_D_D_Type(), CAST_FROM_FN_PTR(address, SharedRuntime::dexp), "EXP");
  case vmIntrinsics::_dpow:  return
    runtime_math(OptoRuntime::Math_DD_D_Type(), CAST_FROM_FN_PTR(address, SharedRuntime::dpow), "POW");

   // These intrinsics are not yet correctly implemented
  case vmIntrinsics::_datan2:
    return false;

  default:
    ShouldNotReachHere();
    return false;
  }
}

static bool is_simple_name(Node* n) {
  return (n->req() == 1         // constant
          || (n->is_Type() && n->as_Type()->type()->singleton())
          || n->is_Proj()       // parameter or return value
          || n->is_Phi()        // local of some sort
          );
}

//----------------------------inline_min_max-----------------------------------
bool LibraryCallKit::inline_min_max(vmIntrinsics::ID id) {
  push(generate_min_max(id, argument(0), argument(1)));

  return true;
}

Node*
LibraryCallKit::generate_min_max(vmIntrinsics::ID id, Node* x0, Node* y0) {
  // These are the candidate return value:
  Node* xvalue = x0;
  Node* yvalue = y0;

  if (xvalue == yvalue) {
    return xvalue;
  }

  bool want_max = (id == vmIntrinsics::_max);

  const TypeInt* txvalue = _gvn.type(xvalue)->isa_int();
  const TypeInt* tyvalue = _gvn.type(yvalue)->isa_int();
  if (txvalue == NULL || tyvalue == NULL)  return top();
  // This is not really necessary, but it is consistent with a
  // hypothetical MaxINode::Value method:
  int widen = MAX2(txvalue->_widen, tyvalue->_widen);

  // %%% This folding logic should (ideally) be in a different place.
  // Some should be inside IfNode, and there to be a more reliable
  // transformation of ?: style patterns into cmoves.  We also want
  // more powerful optimizations around cmove and min/max.

  // Try to find a dominating comparison of these guys.
  // It can simplify the index computation for Arrays.copyOf
  // and similar uses of System.arraycopy.
  // First, compute the normalized version of CmpI(x, y).
  int   cmp_op = Op_CmpI;
  Node* xkey = xvalue;
  Node* ykey = yvalue;
  Node* ideal_cmpxy = _gvn.transform( new(C, 3) CmpINode(xkey, ykey) );
  if (ideal_cmpxy->is_Cmp()) {
    // E.g., if we have CmpI(length - offset, count),
    // it might idealize to CmpI(length, count + offset)
    cmp_op = ideal_cmpxy->Opcode();
    xkey = ideal_cmpxy->in(1);
    ykey = ideal_cmpxy->in(2);
  }

  // Start by locating any relevant comparisons.
  Node* start_from = (xkey->outcnt() < ykey->outcnt()) ? xkey : ykey;
  Node* cmpxy = NULL;
  Node* cmpyx = NULL;
  for (DUIterator_Fast kmax, k = start_from->fast_outs(kmax); k < kmax; k++) {
    Node* cmp = start_from->fast_out(k);
    if (cmp->outcnt() > 0 &&            // must have prior uses
        cmp->in(0) == NULL &&           // must be context-independent
        cmp->Opcode() == cmp_op) {      // right kind of compare
      if (cmp->in(1) == xkey && cmp->in(2) == ykey)  cmpxy = cmp;
      if (cmp->in(1) == ykey && cmp->in(2) == xkey)  cmpyx = cmp;
    }
  }

  const int NCMPS = 2;
  Node* cmps[NCMPS] = { cmpxy, cmpyx };
  int cmpn;
  for (cmpn = 0; cmpn < NCMPS; cmpn++) {
    if (cmps[cmpn] != NULL)  break;     // find a result
  }
  if (cmpn < NCMPS) {
    // Look for a dominating test that tells us the min and max.
    int depth = 0;                // Limit search depth for speed
    Node* dom = control();
    for (; dom != NULL; dom = IfNode::up_one_dom(dom, true)) {
      if (++depth >= 100)  break;
      Node* ifproj = dom;
      if (!ifproj->is_Proj())  continue;
      Node* iff = ifproj->in(0);
      if (!iff->is_If())  continue;
      Node* bol = iff->in(1);
      if (!bol->is_Bool())  continue;
      Node* cmp = bol->in(1);
      if (cmp == NULL)  continue;
      for (cmpn = 0; cmpn < NCMPS; cmpn++)
        if (cmps[cmpn] == cmp)  break;
      if (cmpn == NCMPS)  continue;
      BoolTest::mask btest = bol->as_Bool()->_test._test;
      if (ifproj->is_IfFalse())  btest = BoolTest(btest).negate();
      if (cmp->in(1) == ykey)    btest = BoolTest(btest).commute();
      // At this point, we know that 'x btest y' is true.
      switch (btest) {
      case BoolTest::eq:
        // They are proven equal, so we can collapse the min/max.
        // Either value is the answer.  Choose the simpler.
        if (is_simple_name(yvalue) && !is_simple_name(xvalue))
          return yvalue;
        return xvalue;
      case BoolTest::lt:          // x < y
      case BoolTest::le:          // x <= y
        return (want_max ? yvalue : xvalue);
      case BoolTest::gt:          // x > y
      case BoolTest::ge:          // x >= y
        return (want_max ? xvalue : yvalue);
      }
    }
  }

  // We failed to find a dominating test.
  // Let's pick a test that might GVN with prior tests.
  Node*          best_bol   = NULL;
  BoolTest::mask best_btest = BoolTest::illegal;
  for (cmpn = 0; cmpn < NCMPS; cmpn++) {
    Node* cmp = cmps[cmpn];
    if (cmp == NULL)  continue;
    for (DUIterator_Fast jmax, j = cmp->fast_outs(jmax); j < jmax; j++) {
      Node* bol = cmp->fast_out(j);
      if (!bol->is_Bool())  continue;
      BoolTest::mask btest = bol->as_Bool()->_test._test;
      if (btest == BoolTest::eq || btest == BoolTest::ne)  continue;
      if (cmp->in(1) == ykey)   btest = BoolTest(btest).commute();
      if (bol->outcnt() > (best_bol == NULL ? 0 : best_bol->outcnt())) {
        best_bol   = bol->as_Bool();
        best_btest = btest;
      }
    }
  }

  Node* answer_if_true  = NULL;
  Node* answer_if_false = NULL;
  switch (best_btest) {
  default:
    if (cmpxy == NULL)
      cmpxy = ideal_cmpxy;
    best_bol = _gvn.transform( new(C, 2) BoolNode(cmpxy, BoolTest::lt) );
    // and fall through:
  case BoolTest::lt:          // x < y
  case BoolTest::le:          // x <= y
    answer_if_true  = (want_max ? yvalue : xvalue);
    answer_if_false = (want_max ? xvalue : yvalue);
    break;
  case BoolTest::gt:          // x > y
  case BoolTest::ge:          // x >= y
    answer_if_true  = (want_max ? xvalue : yvalue);
    answer_if_false = (want_max ? yvalue : xvalue);
    break;
  }

  jint hi, lo;
  if (want_max) {
    // We can sharpen the minimum.
    hi = MAX2(txvalue->_hi, tyvalue->_hi);
    lo = MAX2(txvalue->_lo, tyvalue->_lo);
  } else {
    // We can sharpen the maximum.
    hi = MIN2(txvalue->_hi, tyvalue->_hi);
    lo = MIN2(txvalue->_lo, tyvalue->_lo);
  }

  // Use a flow-free graph structure, to avoid creating excess control edges
  // which could hinder other optimizations.
  // Since Math.min/max is often used with arraycopy, we want
  // tightly_coupled_allocation to be able to see beyond min/max expressions.
  Node* cmov = CMoveNode::make(C, NULL, best_bol,
                               answer_if_false, answer_if_true,
                               TypeInt::make(lo, hi, widen));

  return _gvn.transform(cmov);

  /*
  // This is not as desirable as it may seem, since Min and Max
  // nodes do not have a full set of optimizations.
  // And they would interfere, anyway, with 'if' optimizations
  // and with CMoveI canonical forms.
  switch (id) {
  case vmIntrinsics::_min:
    result_val = _gvn.transform(new (C, 3) MinINode(x,y)); break;
  case vmIntrinsics::_max:
    result_val = _gvn.transform(new (C, 3) MaxINode(x,y)); break;
  default:
    ShouldNotReachHere();
  }
  */
}

inline int
LibraryCallKit::classify_unsafe_addr(Node* &base, Node* &offset) {
  const TypePtr* base_type = TypePtr::NULL_PTR;
  if (base != NULL)  base_type = _gvn.type(base)->isa_ptr();
  if (base_type == NULL) {
    // Unknown type.
    return Type::AnyPtr;
  } else if (base_type == TypePtr::NULL_PTR) {
    // Since this is a NULL+long form, we have to switch to a rawptr.
    base   = _gvn.transform( new (C, 2) CastX2PNode(offset) );
    offset = MakeConX(0);
    return Type::RawPtr;
  } else if (base_type->base() == Type::RawPtr) {
    return Type::RawPtr;
  } else if (base_type->isa_oopptr()) {
    // Base is never null => always a heap address.
    if (base_type->ptr() == TypePtr::NotNull) {
      return Type::OopPtr;
    }
    // Offset is small => always a heap address.
    const TypeX* offset_type = _gvn.type(offset)->isa_intptr_t();
    if (offset_type != NULL &&
        base_type->offset() == 0 &&     // (should always be?)
        offset_type->_lo >= 0 &&
        !MacroAssembler::needs_explicit_null_check(offset_type->_hi)) {
      return Type::OopPtr;
    }
    // Otherwise, it might either be oop+off or NULL+addr.
    return Type::AnyPtr;
  } else {
    // No information:
    return Type::AnyPtr;
  }
}

inline Node* LibraryCallKit::make_unsafe_address(Node* base, Node* offset) {
  int kind = classify_unsafe_addr(base, offset);
  if (kind == Type::RawPtr) {
    return basic_plus_adr(top(), base, offset);
  } else {
    return basic_plus_adr(base, offset);
  }
}

//----------------------------inline_reverseBytes_int/long-------------------
// inline Int.reverseBytes(int)
// inline Long.reverseByes(long)
bool LibraryCallKit::inline_reverseBytes(vmIntrinsics::ID id) {
  assert(id == vmIntrinsics::_reverseBytes_i || id == vmIntrinsics::_reverseBytes_l, "not reverse Bytes");
  if (id == vmIntrinsics::_reverseBytes_i && !Matcher::has_match_rule(Op_ReverseBytesI)) return false;
  if (id == vmIntrinsics::_reverseBytes_l && !Matcher::has_match_rule(Op_ReverseBytesL)) return false;
  _sp += arg_size();        // restore stack pointer
  switch (id) {
  case vmIntrinsics::_reverseBytes_i:
    push(_gvn.transform(new (C, 2) ReverseBytesINode(0, pop())));
    break;
  case vmIntrinsics::_reverseBytes_l:
    push_pair(_gvn.transform(new (C, 2) ReverseBytesLNode(0, pop_pair())));
    break;
  default:
    ;
  }
  return true;
}

//----------------------------inline_unsafe_access----------------------------

const static BasicType T_ADDRESS_HOLDER = T_LONG;

// Interpret Unsafe.fieldOffset cookies correctly:
extern jlong Unsafe_field_offset_to_byte_offset(jlong field_offset);

bool LibraryCallKit::inline_unsafe_access(bool is_native_ptr, bool is_store, BasicType type, bool is_volatile) {
  if (callee()->is_static())  return false;  // caller must have the capability!

#ifndef PRODUCT
  {
    ResourceMark rm;
    // Check the signatures.
    ciSignature* sig = signature();
#ifdef ASSERT
    if (!is_store) {
      // Object getObject(Object base, int/long offset), etc.
      BasicType rtype = sig->return_type()->basic_type();
      if (rtype == T_ADDRESS_HOLDER && callee()->name() == ciSymbol::getAddress_name())
          rtype = T_ADDRESS;  // it is really a C void*
      assert(rtype == type, "getter must return the expected value");
      if (!is_native_ptr) {
        assert(sig->count() == 2, "oop getter has 2 arguments");
        assert(sig->type_at(0)->basic_type() == T_OBJECT, "getter base is object");
        assert(sig->type_at(1)->basic_type() == T_LONG, "getter offset is correct");
      } else {
        assert(sig->count() == 1, "native getter has 1 argument");
        assert(sig->type_at(0)->basic_type() == T_LONG, "getter base is long");
      }
    } else {
      // void putObject(Object base, int/long offset, Object x), etc.
      assert(sig->return_type()->basic_type() == T_VOID, "putter must not return a value");
      if (!is_native_ptr) {
        assert(sig->count() == 3, "oop putter has 3 arguments");
        assert(sig->type_at(0)->basic_type() == T_OBJECT, "putter base is object");
        assert(sig->type_at(1)->basic_type() == T_LONG, "putter offset is correct");
      } else {
        assert(sig->count() == 2, "native putter has 2 arguments");
        assert(sig->type_at(0)->basic_type() == T_LONG, "putter base is long");
      }
      BasicType vtype = sig->type_at(sig->count()-1)->basic_type();
      if (vtype == T_ADDRESS_HOLDER && callee()->name() == ciSymbol::putAddress_name())
        vtype = T_ADDRESS;  // it is really a C void*
      assert(vtype == type, "putter must accept the expected value");
    }
#endif // ASSERT
 }
#endif //PRODUCT

  C->set_has_unsafe_access(true);  // Mark eventual nmethod as "unsafe".

  int type_words = type2size[ (type == T_ADDRESS) ? T_LONG : type ];

  // Argument words:  "this" plus (oop/offset) or (lo/hi) args plus maybe 1 or 2 value words
  int nargs = 1 + (is_native_ptr ? 2 : 3) + (is_store ? type_words : 0);

  debug_only(int saved_sp = _sp);
  _sp += nargs;

  Node* val;
  debug_only(val = (Node*)(uintptr_t)-1);


  if (is_store) {
    // Get the value being stored.  (Pop it first; it was pushed last.)
    switch (type) {
    case T_DOUBLE:
    case T_LONG:
    case T_ADDRESS:
      val = pop_pair();
      break;
    default:
      val = pop();
    }
  }

  // Build address expression.  See the code in inline_unsafe_prefetch.
  Node *adr;
  Node *heap_base_oop = top();
  if (!is_native_ptr) {
    // The offset is a value produced by Unsafe.staticFieldOffset or Unsafe.objectFieldOffset
    Node* offset = pop_pair();
    // The base is either a Java object or a value produced by Unsafe.staticFieldBase
    Node* base   = pop();
    // We currently rely on the cookies produced by Unsafe.xxxFieldOffset
    // to be plain byte offsets, which are also the same as those accepted
    // by oopDesc::field_base.
    assert(Unsafe_field_offset_to_byte_offset(11) == 11,
           "fieldOffset must be byte-scaled");
    // 32-bit machines ignore the high half!
    offset = ConvL2X(offset);
    adr = make_unsafe_address(base, offset);
    heap_base_oop = base;
  } else {
    Node* ptr = pop_pair();
    // Adjust Java long to machine word:
    ptr = ConvL2X(ptr);
    adr = make_unsafe_address(NULL, ptr);
  }

  // Pop receiver last:  it was pushed first.
  Node *receiver = pop();

  assert(saved_sp == _sp, "must have correct argument count");

  const TypePtr *adr_type = _gvn.type(adr)->isa_ptr();

  // First guess at the value type.
  const Type *value_type = Type::get_const_basic_type(type);

  // Try to categorize the address.  If it comes up as TypeJavaPtr::BOTTOM,
  // there was not enough information to nail it down.
  Compile::AliasType* alias_type = C->alias_type(adr_type);
  assert(alias_type->index() != Compile::AliasIdxBot, "no bare pointers here");

  // We will need memory barriers unless we can determine a unique
  // alias category for this reference.  (Note:  If for some reason
  // the barriers get omitted and the unsafe reference begins to "pollute"
  // the alias analysis of the rest of the graph, either Compile::can_alias
  // or Compile::must_alias will throw a diagnostic assert.)
  bool need_mem_bar = (alias_type->adr_type() == TypeOopPtr::BOTTOM);

  if (!is_store && type == T_OBJECT) {
    // Attempt to infer a sharper value type from the offset and base type.
    ciKlass* sharpened_klass = NULL;

    // See if it is an instance field, with an object type.
    if (alias_type->field() != NULL) {
      assert(!is_native_ptr, "native pointer op cannot use a java address");
      if (alias_type->field()->type()->is_klass()) {
        sharpened_klass = alias_type->field()->type()->as_klass();
      }
    }

    // See if it is a narrow oop array.
    if (adr_type->isa_aryptr()) {
      if (adr_type->offset() >= objArrayOopDesc::header_size() * wordSize) {
        const TypeOopPtr *elem_type = adr_type->is_aryptr()->elem()->isa_oopptr();
        if (elem_type != NULL) {
          sharpened_klass = elem_type->klass();
        }
      }
    }

    if (sharpened_klass != NULL) {
      const TypeOopPtr* tjp = TypeOopPtr::make_from_klass(sharpened_klass);

      // Sharpen the value type.
      value_type = tjp;

#ifndef PRODUCT
      if (PrintIntrinsics || PrintInlining || PrintOptoInlining) {
        tty->print("  from base type:  ");   adr_type->dump();
        tty->print("  sharpened value: "); value_type->dump();
      }
#endif
    }
  }

  // Null check on self without removing any arguments.  The argument
  // null check technically happens in the wrong place, which can lead to
  // invalid stack traces when the primitive is inlined into a method
  // which handles NullPointerExceptions.
  _sp += nargs;
  do_null_check(receiver, T_OBJECT);
  _sp -= nargs;
  if (stopped()) {
    return true;
  }
  // Heap pointers get a null-check from the interpreter,
  // as a courtesy.  However, this is not guaranteed by Unsafe,
  // and it is not possible to fully distinguish unintended nulls
  // from intended ones in this API.

  if (is_volatile) {
    // We need to emit leading and trailing CPU membars (see below) in
    // addition to memory membars when is_volatile. This is a little
    // too strong, but avoids the need to insert per-alias-type
    // volatile membars (for stores; compare Parse::do_put_xxx), which
    // we cannot do effctively here because we probably only have a
    // rough approximation of type.
    need_mem_bar = true;
    // For Stores, place a memory ordering barrier now.
    if (is_store)
      insert_mem_bar(Op_MemBarRelease);
  }

  // Memory barrier to prevent normal and 'unsafe' accesses from
  // bypassing each other.  Happens after null checks, so the
  // exception paths do not take memory state from the memory barrier,
  // so there's no problems making a strong assert about mixing users
  // of safe & unsafe memory.  Otherwise fails in a CTW of rt.jar
  // around 5701, class sun/reflect/UnsafeBooleanFieldAccessorImpl.
  if (need_mem_bar) insert_mem_bar(Op_MemBarCPUOrder);

  if (!is_store) {
    Node* p = make_load(control(), adr, value_type, type, adr_type, is_volatile);
    // load value and push onto stack
    switch (type) {
    case T_BOOLEAN:
    case T_CHAR:
    case T_BYTE:
    case T_SHORT:
    case T_INT:
    case T_FLOAT:
    case T_OBJECT:
      push( p );
      break;
    case T_ADDRESS:
      // Cast to an int type.
      p = _gvn.transform( new (C, 2) CastP2XNode(NULL,p) );
      p = ConvX2L(p);
      push_pair(p);
      break;
    case T_DOUBLE:
    case T_LONG:
      push_pair( p );
      break;
    default: ShouldNotReachHere();
    }
  } else {
    // place effect of store into memory
    switch (type) {
    case T_DOUBLE:
      val = dstore_rounding(val);
      break;
    case T_ADDRESS:
      // Repackage the long as a pointer.
      val = ConvL2X(val);
      val = _gvn.transform( new (C, 2) CastX2PNode(val) );
      break;
    }

    if (type != T_OBJECT ) {
      (void) store_to_memory(control(), adr, val, type, adr_type, is_volatile);
    } else {
      // Possibly an oop being stored to Java heap or native memory
      if (!TypePtr::NULL_PTR->higher_equal(_gvn.type(heap_base_oop))) {
        // oop to Java heap.
        (void) store_oop_to_unknown(control(), heap_base_oop, adr, adr_type, val, val->bottom_type(), type);
      } else {

        // We can't tell at compile time if we are storing in the Java heap or outside
        // of it. So we need to emit code to conditionally do the proper type of
        // store.

        IdealKit kit(gvn(), control(),  merged_memory());
        kit.declares_done();
        // QQQ who knows what probability is here??
        kit.if_then(heap_base_oop, BoolTest::ne, null(), PROB_UNLIKELY(0.999)); {
          (void) store_oop_to_unknown(control(), heap_base_oop, adr, adr_type, val, val->bottom_type(), type);
        } kit.else_(); {
          (void) store_to_memory(control(), adr, val, type, adr_type, is_volatile);
        } kit.end_if();
      }
    }
  }

  if (is_volatile) {
    if (!is_store)
      insert_mem_bar(Op_MemBarAcquire);
    else
      insert_mem_bar(Op_MemBarVolatile);
  }

  if (need_mem_bar) insert_mem_bar(Op_MemBarCPUOrder);

  return true;
}

//----------------------------inline_unsafe_prefetch----------------------------

bool LibraryCallKit::inline_unsafe_prefetch(bool is_native_ptr, bool is_store, bool is_static) {
#ifndef PRODUCT
  {
    ResourceMark rm;
    // Check the signatures.
    ciSignature* sig = signature();
#ifdef ASSERT
    // Object getObject(Object base, int/long offset), etc.
    BasicType rtype = sig->return_type()->basic_type();
    if (!is_native_ptr) {
      assert(sig->count() == 2, "oop prefetch has 2 arguments");
      assert(sig->type_at(0)->basic_type() == T_OBJECT, "prefetch base is object");
      assert(sig->type_at(1)->basic_type() == T_LONG, "prefetcha offset is correct");
    } else {
      assert(sig->count() == 1, "native prefetch has 1 argument");
      assert(sig->type_at(0)->basic_type() == T_LONG, "prefetch base is long");
    }
#endif // ASSERT
  }
#endif // !PRODUCT

  C->set_has_unsafe_access(true);  // Mark eventual nmethod as "unsafe".

  // Argument words:  "this" if not static, plus (oop/offset) or (lo/hi) args
  int nargs = (is_static ? 0 : 1) + (is_native_ptr ? 2 : 3);

  debug_only(int saved_sp = _sp);
  _sp += nargs;

  // Build address expression.  See the code in inline_unsafe_access.
  Node *adr;
  if (!is_native_ptr) {
    // The offset is a value produced by Unsafe.staticFieldOffset or Unsafe.objectFieldOffset
    Node* offset = pop_pair();
    // The base is either a Java object or a value produced by Unsafe.staticFieldBase
    Node* base   = pop();
    // We currently rely on the cookies produced by Unsafe.xxxFieldOffset
    // to be plain byte offsets, which are also the same as those accepted
    // by oopDesc::field_base.
    assert(Unsafe_field_offset_to_byte_offset(11) == 11,
           "fieldOffset must be byte-scaled");
    // 32-bit machines ignore the high half!
    offset = ConvL2X(offset);
    adr = make_unsafe_address(base, offset);
  } else {
    Node* ptr = pop_pair();
    // Adjust Java long to machine word:
    ptr = ConvL2X(ptr);
    adr = make_unsafe_address(NULL, ptr);
  }

  if (is_static) {
    assert(saved_sp == _sp, "must have correct argument count");
  } else {
    // Pop receiver last:  it was pushed first.
    Node *receiver = pop();
    assert(saved_sp == _sp, "must have correct argument count");

    // Null check on self without removing any arguments.  The argument
    // null check technically happens in the wrong place, which can lead to
    // invalid stack traces when the primitive is inlined into a method
    // which handles NullPointerExceptions.
    _sp += nargs;
    do_null_check(receiver, T_OBJECT);
    _sp -= nargs;
    if (stopped()) {
      return true;
    }
  }

  // Generate the read or write prefetch
  Node *prefetch;
  if (is_store) {
    prefetch = new (C, 3) PrefetchWriteNode(i_o(), adr);
  } else {
    prefetch = new (C, 3) PrefetchReadNode(i_o(), adr);
  }
  prefetch->init_req(0, control());
  set_i_o(_gvn.transform(prefetch));

  return true;
}

//----------------------------inline_unsafe_CAS----------------------------

bool LibraryCallKit::inline_unsafe_CAS(BasicType type) {
  // This basic scheme here is the same as inline_unsafe_access, but
  // differs in enough details that combining them would make the code
  // overly confusing.  (This is a true fact! I originally combined
  // them, but even I was confused by it!) As much code/comments as
  // possible are retained from inline_unsafe_access though to make
  // the correspondances clearer. - dl

  if (callee()->is_static())  return false;  // caller must have the capability!

#ifndef PRODUCT
  {
    ResourceMark rm;
    // Check the signatures.
    ciSignature* sig = signature();
#ifdef ASSERT
    BasicType rtype = sig->return_type()->basic_type();
    assert(rtype == T_BOOLEAN, "CAS must return boolean");
    assert(sig->count() == 4, "CAS has 4 arguments");
    assert(sig->type_at(0)->basic_type() == T_OBJECT, "CAS base is object");
    assert(sig->type_at(1)->basic_type() == T_LONG, "CAS offset is long");
#endif // ASSERT
  }
#endif //PRODUCT

  // number of stack slots per value argument (1 or 2)
  int type_words = type2size[type];

  // Cannot inline wide CAS on machines that don't support it natively
  if (type2aelembytes[type] > BytesPerInt && !VM_Version::supports_cx8())
    return false;

  C->set_has_unsafe_access(true);  // Mark eventual nmethod as "unsafe".

  // Argument words:  "this" plus oop plus offset plus oldvalue plus newvalue;
  int nargs = 1 + 1 + 2  + type_words + type_words;

  // pop arguments: newval, oldval, offset, base, and receiver
  debug_only(int saved_sp = _sp);
  _sp += nargs;
  Node* newval   = (type_words == 1) ? pop() : pop_pair();
  Node* oldval   = (type_words == 1) ? pop() : pop_pair();
  Node *offset   = pop_pair();
  Node *base     = pop();
  Node *receiver = pop();
  assert(saved_sp == _sp, "must have correct argument count");

  //  Null check receiver.
  _sp += nargs;
  do_null_check(receiver, T_OBJECT);
  _sp -= nargs;
  if (stopped()) {
    return true;
  }

  // Build field offset expression.
  // We currently rely on the cookies produced by Unsafe.xxxFieldOffset
  // to be plain byte offsets, which are also the same as those accepted
  // by oopDesc::field_base.
  assert(Unsafe_field_offset_to_byte_offset(11) == 11, "fieldOffset must be byte-scaled");
  // 32-bit machines ignore the high half of long offsets
  offset = ConvL2X(offset);
  Node* adr = make_unsafe_address(base, offset);
  const TypePtr *adr_type = _gvn.type(adr)->isa_ptr();

  // (Unlike inline_unsafe_access, there seems no point in trying
  // to refine types. Just use the coarse types here.
  const Type *value_type = Type::get_const_basic_type(type);
  Compile::AliasType* alias_type = C->alias_type(adr_type);
  assert(alias_type->index() != Compile::AliasIdxBot, "no bare pointers here");
  int alias_idx = C->get_alias_index(adr_type);

  // Memory-model-wise, a CAS acts like a little synchronized block,
  // so needs barriers on each side.  These don't't translate into
  // actual barriers on most machines, but we still need rest of
  // compiler to respect ordering.

  insert_mem_bar(Op_MemBarRelease);
  insert_mem_bar(Op_MemBarCPUOrder);

  // 4984716: MemBars must be inserted before this
  //          memory node in order to avoid a false
  //          dependency which will confuse the scheduler.
  Node *mem = memory(alias_idx);

  // For now, we handle only those cases that actually exist: ints,
  // longs, and Object. Adding others should be straightforward.
  Node* cas;
  switch(type) {
  case T_INT:
    cas = _gvn.transform(new (C, 5) CompareAndSwapINode(control(), mem, adr, newval, oldval));
    break;
  case T_LONG:
    cas = _gvn.transform(new (C, 5) CompareAndSwapLNode(control(), mem, adr, newval, oldval));
    break;
  case T_OBJECT:
    // reference stores need a store barrier.
    // (They don't if CAS fails, but it isn't worth checking.)
    pre_barrier(control(), base, adr, alias_idx, newval, value_type, T_OBJECT);
    cas = _gvn.transform(new (C, 5) CompareAndSwapPNode(control(), mem, adr, newval, oldval));
    post_barrier(control(), cas, base, adr, alias_idx, newval, T_OBJECT, true);
    break;
  default:
    ShouldNotReachHere();
    break;
  }

  // SCMemProjNodes represent the memory state of CAS. Their main
  // role is to prevent CAS nodes from being optimized away when their
  // results aren't used.
  Node* proj = _gvn.transform( new (C, 1) SCMemProjNode(cas));
  set_memory(proj, alias_idx);

  // Add the trailing membar surrounding the access
  insert_mem_bar(Op_MemBarCPUOrder);
  insert_mem_bar(Op_MemBarAcquire);

  push(cas);
  return true;
}

bool LibraryCallKit::inline_unsafe_ordered_store(BasicType type) {
  // This is another variant of inline_unsafe_access, differing in
  // that it always issues store-store ("release") barrier and ensures
  // store-atomicity (which only matters for "long").

  if (callee()->is_static())  return false;  // caller must have the capability!

#ifndef PRODUCT
  {
    ResourceMark rm;
    // Check the signatures.
    ciSignature* sig = signature();
#ifdef ASSERT
    BasicType rtype = sig->return_type()->basic_type();
    assert(rtype == T_VOID, "must return void");
    assert(sig->count() == 3, "has 3 arguments");
    assert(sig->type_at(0)->basic_type() == T_OBJECT, "base is object");
    assert(sig->type_at(1)->basic_type() == T_LONG, "offset is long");
#endif // ASSERT
  }
#endif //PRODUCT

  // number of stack slots per value argument (1 or 2)
  int type_words = type2size[type];

  C->set_has_unsafe_access(true);  // Mark eventual nmethod as "unsafe".

  // Argument words:  "this" plus oop plus offset plus value;
  int nargs = 1 + 1 + 2 + type_words;

  // pop arguments: val, offset, base, and receiver
  debug_only(int saved_sp = _sp);
  _sp += nargs;
  Node* val      = (type_words == 1) ? pop() : pop_pair();
  Node *offset   = pop_pair();
  Node *base     = pop();
  Node *receiver = pop();
  assert(saved_sp == _sp, "must have correct argument count");

  //  Null check receiver.
  _sp += nargs;
  do_null_check(receiver, T_OBJECT);
  _sp -= nargs;
  if (stopped()) {
    return true;
  }

  // Build field offset expression.
  assert(Unsafe_field_offset_to_byte_offset(11) == 11, "fieldOffset must be byte-scaled");
  // 32-bit machines ignore the high half of long offsets
  offset = ConvL2X(offset);
  Node* adr = make_unsafe_address(base, offset);
  const TypePtr *adr_type = _gvn.type(adr)->isa_ptr();
  const Type *value_type = Type::get_const_basic_type(type);
  Compile::AliasType* alias_type = C->alias_type(adr_type);

  insert_mem_bar(Op_MemBarRelease);
  insert_mem_bar(Op_MemBarCPUOrder);
  // Ensure that the store is atomic for longs:
  bool require_atomic_access = true;
  Node* store;
  if (type == T_OBJECT) // reference stores need a store barrier.
    store = store_oop_to_unknown(control(), base, adr, adr_type, val, value_type, type);
  else {
    store = store_to_memory(control(), adr, val, type, adr_type, require_atomic_access);
  }
  insert_mem_bar(Op_MemBarCPUOrder);
  return true;
}

bool LibraryCallKit::inline_unsafe_allocate() {
  if (callee()->is_static())  return false;  // caller must have the capability!
  int nargs = 1 + 1;
  assert(signature()->size() == nargs-1, "alloc has 1 argument");
  null_check_receiver(callee());  // check then ignore argument(0)
  _sp += nargs;  // set original stack for use by uncommon_trap
  Node* cls = do_null_check(argument(1), T_OBJECT);
  _sp -= nargs;
  if (stopped())  return true;

  Node* kls = load_klass_from_mirror(cls, false, nargs, NULL, 0);
  _sp += nargs;  // set original stack for use by uncommon_trap
  kls = do_null_check(kls, T_OBJECT);
  _sp -= nargs;
  if (stopped())  return true;  // argument was like int.class

  // Note:  The argument might still be an illegal value like
  // Serializable.class or Object[].class.   The runtime will handle it.
  // But we must make an explicit check for initialization.
  Node* insp = basic_plus_adr(kls, instanceKlass::init_state_offset_in_bytes() + sizeof(oopDesc));
  Node* inst = make_load(NULL, insp, TypeInt::INT, T_INT);
  Node* bits = intcon(instanceKlass::fully_initialized);
  Node* test = _gvn.transform( new (C, 3) SubINode(inst, bits) );
  // The 'test' is non-zero if we need to take a slow path.

  Node* obj = new_instance(kls, test);
  push(obj);

  return true;
}

//------------------------inline_native_time_funcs--------------
// inline code for System.currentTimeMillis() and System.nanoTime()
// these have the same type and signature
bool LibraryCallKit::inline_native_time_funcs(bool isNano) {
  address funcAddr = isNano ? CAST_FROM_FN_PTR(address, os::javaTimeNanos) :
                              CAST_FROM_FN_PTR(address, os::javaTimeMillis);
  const char * funcName = isNano ? "nanoTime" : "currentTimeMillis";
  const TypeFunc *tf = OptoRuntime::current_time_millis_Type();
  const TypePtr* no_memory_effects = NULL;
  Node* time = make_runtime_call(RC_LEAF, tf, funcAddr, funcName, no_memory_effects);
  Node* value = _gvn.transform(new (C, 1) ProjNode(time, TypeFunc::Parms+0));
#ifdef ASSERT
  Node* value_top = _gvn.transform(new (C, 1) ProjNode(time, TypeFunc::Parms + 1));
  assert(value_top == top(), "second value must be top");
#endif
  push_pair(value);
  return true;
}

//------------------------inline_native_currentThread------------------
bool LibraryCallKit::inline_native_currentThread() {
  Node* junk = NULL;
  push(generate_current_thread(junk));
  return true;
}

//------------------------inline_native_isInterrupted------------------
bool LibraryCallKit::inline_native_isInterrupted() {
  const int nargs = 1+1;  // receiver + boolean
  assert(nargs == arg_size(), "sanity");
  // Add a fast path to t.isInterrupted(clear_int):
  //   (t == Thread.current() && (!TLS._osthread._interrupted || !clear_int))
  //   ? TLS._osthread._interrupted : /*slow path:*/ t.isInterrupted(clear_int)
  // So, in the common case that the interrupt bit is false,
  // we avoid making a call into the VM.  Even if the interrupt bit
  // is true, if the clear_int argument is false, we avoid the VM call.
  // However, if the receiver is not currentThread, we must call the VM,
  // because there must be some locking done around the operation.

  // We only go to the fast case code if we pass two guards.
  // Paths which do not pass are accumulated in the slow_region.
  RegionNode* slow_region = new (C, 1) RegionNode(1);
  record_for_igvn(slow_region);
  RegionNode* result_rgn = new (C, 4) RegionNode(1+3); // fast1, fast2, slow
  PhiNode*    result_val = new (C, 4) PhiNode(result_rgn, TypeInt::BOOL);
  enum { no_int_result_path   = 1,
         no_clear_result_path = 2,
         slow_result_path     = 3
  };

  // (a) Receiving thread must be the current thread.
  Node* rec_thr = argument(0);
  Node* tls_ptr = NULL;
  Node* cur_thr = generate_current_thread(tls_ptr);
  Node* cmp_thr = _gvn.transform( new (C, 3) CmpPNode(cur_thr, rec_thr) );
  Node* bol_thr = _gvn.transform( new (C, 2) BoolNode(cmp_thr, BoolTest::ne) );

  bool known_current_thread = (_gvn.type(bol_thr) == TypeInt::ZERO);
  if (!known_current_thread)
    generate_slow_guard(bol_thr, slow_region);

  // (b) Interrupt bit on TLS must be false.
  Node* p = basic_plus_adr(top()/*!oop*/, tls_ptr, in_bytes(JavaThread::osthread_offset()));
  Node* osthread = make_load(NULL, p, TypeRawPtr::NOTNULL, T_ADDRESS);
  p = basic_plus_adr(top()/*!oop*/, osthread, in_bytes(OSThread::interrupted_offset()));
  Node* int_bit = make_load(NULL, p, TypeInt::BOOL, T_INT);
  Node* cmp_bit = _gvn.transform( new (C, 3) CmpINode(int_bit, intcon(0)) );
  Node* bol_bit = _gvn.transform( new (C, 2) BoolNode(cmp_bit, BoolTest::ne) );

  IfNode* iff_bit = create_and_map_if(control(), bol_bit, PROB_UNLIKELY_MAG(3), COUNT_UNKNOWN);

  // First fast path:  if (!TLS._interrupted) return false;
  Node* false_bit = _gvn.transform( new (C, 1) IfFalseNode(iff_bit) );
  result_rgn->init_req(no_int_result_path, false_bit);
  result_val->init_req(no_int_result_path, intcon(0));

  // drop through to next case
  set_control( _gvn.transform(new (C, 1) IfTrueNode(iff_bit)) );

  // (c) Or, if interrupt bit is set and clear_int is false, use 2nd fast path.
  Node* clr_arg = argument(1);
  Node* cmp_arg = _gvn.transform( new (C, 3) CmpINode(clr_arg, intcon(0)) );
  Node* bol_arg = _gvn.transform( new (C, 2) BoolNode(cmp_arg, BoolTest::ne) );
  IfNode* iff_arg = create_and_map_if(control(), bol_arg, PROB_FAIR, COUNT_UNKNOWN);

  // Second fast path:  ... else if (!clear_int) return true;
  Node* false_arg = _gvn.transform( new (C, 1) IfFalseNode(iff_arg) );
  result_rgn->init_req(no_clear_result_path, false_arg);
  result_val->init_req(no_clear_result_path, intcon(1));

  // drop through to next case
  set_control( _gvn.transform(new (C, 1) IfTrueNode(iff_arg)) );

  // (d) Otherwise, go to the slow path.
  slow_region->add_req(control());
  set_control( _gvn.transform(slow_region) );

  if (stopped()) {
    // There is no slow path.
    result_rgn->init_req(slow_result_path, top());
    result_val->init_req(slow_result_path, top());
  } else {
    // non-virtual because it is a private non-static
    CallJavaNode* slow_call = generate_method_call(vmIntrinsics::_isInterrupted);

    Node* slow_val = set_results_for_java_call(slow_call);
    // this->control() comes from set_results_for_java_call

    // If we know that the result of the slow call will be true, tell the optimizer!
    if (known_current_thread)  slow_val = intcon(1);

    Node* fast_io  = slow_call->in(TypeFunc::I_O);
    Node* fast_mem = slow_call->in(TypeFunc::Memory);
    // These two phis are pre-filled with copies of of the fast IO and Memory
    Node* io_phi   = PhiNode::make(result_rgn, fast_io,  Type::ABIO);
    Node* mem_phi  = PhiNode::make(result_rgn, fast_mem, Type::MEMORY, TypePtr::BOTTOM);

    result_rgn->init_req(slow_result_path, control());
    io_phi    ->init_req(slow_result_path, i_o());
    mem_phi   ->init_req(slow_result_path, reset_memory());
    result_val->init_req(slow_result_path, slow_val);

    set_all_memory( _gvn.transform(mem_phi) );
    set_i_o(        _gvn.transform(io_phi) );
  }

  push_result(result_rgn, result_val);
  C->set_has_split_ifs(true); // Has chance for split-if optimization

  return true;
}

//---------------------------load_mirror_from_klass----------------------------
// Given a klass oop, load its java mirror (a java.lang.Class oop).
Node* LibraryCallKit::load_mirror_from_klass(Node* klass) {
  Node* p = basic_plus_adr(klass, Klass::java_mirror_offset_in_bytes() + sizeof(oopDesc));
  return make_load(NULL, p, TypeInstPtr::MIRROR, T_OBJECT);
}

//-----------------------load_klass_from_mirror_common-------------------------
// Given a java mirror (a java.lang.Class oop), load its corresponding klass oop.
// Test the klass oop for null (signifying a primitive Class like Integer.TYPE),
// and branch to the given path on the region.
// If never_see_null, take an uncommon trap on null, so we can optimistically
// compile for the non-null case.
// If the region is NULL, force never_see_null = true.
Node* LibraryCallKit::load_klass_from_mirror_common(Node* mirror,
                                                    bool never_see_null,
                                                    int nargs,
                                                    RegionNode* region,
                                                    int null_path,
                                                    int offset) {
  if (region == NULL)  never_see_null = true;
  Node* p = basic_plus_adr(mirror, offset);
  const TypeKlassPtr*  kls_type = TypeKlassPtr::OBJECT_OR_NULL;
  Node* kls = _gvn.transform(new (C, 3) LoadKlassNode(0, immutable_memory(), p, TypeRawPtr::BOTTOM, kls_type));
  _sp += nargs; // any deopt will start just before call to enclosing method
  Node* null_ctl = top();
  kls = null_check_oop(kls, &null_ctl, never_see_null);
  if (region != NULL) {
    // Set region->in(null_path) if the mirror is a primitive (e.g, int.class).
    region->init_req(null_path, null_ctl);
  } else {
    assert(null_ctl == top(), "no loose ends");
  }
  _sp -= nargs;
  return kls;
}

//--------------------(inline_native_Class_query helpers)---------------------
// Use this for JVM_ACC_INTERFACE, JVM_ACC_IS_CLONEABLE, JVM_ACC_HAS_FINALIZER.
// Fall through if (mods & mask) == bits, take the guard otherwise.
Node* LibraryCallKit::generate_access_flags_guard(Node* kls, int modifier_mask, int modifier_bits, RegionNode* region) {
  // Branch around if the given klass has the given modifier bit set.
  // Like generate_guard, adds a new path onto the region.
  Node* modp = basic_plus_adr(kls, Klass::access_flags_offset_in_bytes() + sizeof(oopDesc));
  Node* mods = make_load(NULL, modp, TypeInt::INT, T_INT);
  Node* mask = intcon(modifier_mask);
  Node* bits = intcon(modifier_bits);
  Node* mbit = _gvn.transform( new (C, 3) AndINode(mods, mask) );
  Node* cmp  = _gvn.transform( new (C, 3) CmpINode(mbit, bits) );
  Node* bol  = _gvn.transform( new (C, 2) BoolNode(cmp, BoolTest::ne) );
  return generate_fair_guard(bol, region);
}
Node* LibraryCallKit::generate_interface_guard(Node* kls, RegionNode* region) {
  return generate_access_flags_guard(kls, JVM_ACC_INTERFACE, 0, region);
}

//-------------------------inline_native_Class_query-------------------
bool LibraryCallKit::inline_native_Class_query(vmIntrinsics::ID id) {
  int nargs = 1+0;  // just the Class mirror, in most cases
  const Type* return_type = TypeInt::BOOL;
  Node* prim_return_value = top();  // what happens if it's a primitive class?
  bool never_see_null = !too_many_traps(Deoptimization::Reason_null_check);
  bool expect_prim = false;     // most of these guys expect to work on refs

  enum { _normal_path = 1, _prim_path = 2, PATH_LIMIT };

  switch (id) {
  case vmIntrinsics::_isInstance:
    nargs = 1+1;  // the Class mirror, plus the object getting queried about
    // nothing is an instance of a primitive type
    prim_return_value = intcon(0);
    break;
  case vmIntrinsics::_getModifiers:
    prim_return_value = intcon(JVM_ACC_ABSTRACT | JVM_ACC_FINAL | JVM_ACC_PUBLIC);
    assert(is_power_of_2((int)JVM_ACC_WRITTEN_FLAGS+1), "change next line");
    return_type = TypeInt::make(0, JVM_ACC_WRITTEN_FLAGS, Type::WidenMin);
    break;
  case vmIntrinsics::_isInterface:
    prim_return_value = intcon(0);
    break;
  case vmIntrinsics::_isArray:
    prim_return_value = intcon(0);
    expect_prim = true;  // cf. ObjectStreamClass.getClassSignature
    break;
  case vmIntrinsics::_isPrimitive:
    prim_return_value = intcon(1);
    expect_prim = true;  // obviously
    break;
  case vmIntrinsics::_getSuperclass:
    prim_return_value = null();
    return_type = TypeInstPtr::MIRROR->cast_to_ptr_type(TypePtr::BotPTR);
    break;
  case vmIntrinsics::_getComponentType:
    prim_return_value = null();
    return_type = TypeInstPtr::MIRROR->cast_to_ptr_type(TypePtr::BotPTR);
    break;
  case vmIntrinsics::_getClassAccessFlags:
    prim_return_value = intcon(JVM_ACC_ABSTRACT | JVM_ACC_FINAL | JVM_ACC_PUBLIC);
    return_type = TypeInt::INT;  // not bool!  6297094
    break;
  default:
    ShouldNotReachHere();
  }

  Node* mirror =                      argument(0);
  Node* obj    = (nargs <= 1)? top(): argument(1);

  const TypeInstPtr* mirror_con = _gvn.type(mirror)->isa_instptr();
  if (mirror_con == NULL)  return false;  // cannot happen?

#ifndef PRODUCT
  if (PrintIntrinsics || PrintInlining || PrintOptoInlining) {
    ciType* k = mirror_con->java_mirror_type();
    if (k) {
      tty->print("Inlining %s on constant Class ", vmIntrinsics::name_at(intrinsic_id()));
      k->print_name();
      tty->cr();
    }
  }
#endif

  // Null-check the mirror, and the mirror's klass ptr (in case it is a primitive).
  RegionNode* region = new (C, PATH_LIMIT) RegionNode(PATH_LIMIT);
  record_for_igvn(region);
  PhiNode* phi = new (C, PATH_LIMIT) PhiNode(region, return_type);

  // The mirror will never be null of Reflection.getClassAccessFlags, however
  // it may be null for Class.isInstance or Class.getModifiers. Throw a NPE
  // if it is. See bug 4774291.

  // For Reflection.getClassAccessFlags(), the null check occurs in
  // the wrong place; see inline_unsafe_access(), above, for a similar
  // situation.
  _sp += nargs;  // set original stack for use by uncommon_trap
  mirror = do_null_check(mirror, T_OBJECT);
  _sp -= nargs;
  // If mirror or obj is dead, only null-path is taken.
  if (stopped())  return true;

  if (expect_prim)  never_see_null = false;  // expect nulls (meaning prims)

  // Now load the mirror's klass metaobject, and null-check it.
  // Side-effects region with the control path if the klass is null.
  Node* kls = load_klass_from_mirror(mirror, never_see_null, nargs,
                                     region, _prim_path);
  // If kls is null, we have a primitive mirror.
  phi->init_req(_prim_path, prim_return_value);
  if (stopped()) { push_result(region, phi); return true; }

  Node* p;  // handy temp
  Node* null_ctl;

  // Now that we have the non-null klass, we can perform the real query.
  // For constant classes, the query will constant-fold in LoadNode::Value.
  Node* query_value = top();
  switch (id) {
  case vmIntrinsics::_isInstance:
    // nothing is an instance of a primitive type
    query_value = gen_instanceof(obj, kls);
    break;

  case vmIntrinsics::_getModifiers:
    p = basic_plus_adr(kls, Klass::modifier_flags_offset_in_bytes() + sizeof(oopDesc));
    query_value = make_load(NULL, p, TypeInt::INT, T_INT);
    break;

  case vmIntrinsics::_isInterface:
    // (To verify this code sequence, check the asserts in JVM_IsInterface.)
    if (generate_interface_guard(kls, region) != NULL)
      // A guard was added.  If the guard is taken, it was an interface.
      phi->add_req(intcon(1));
    // If we fall through, it's a plain class.
    query_value = intcon(0);
    break;

  case vmIntrinsics::_isArray:
    // (To verify this code sequence, check the asserts in JVM_IsArrayClass.)
    if (generate_array_guard(kls, region) != NULL)
      // A guard was added.  If the guard is taken, it was an array.
      phi->add_req(intcon(1));
    // If we fall through, it's a plain class.
    query_value = intcon(0);
    break;

  case vmIntrinsics::_isPrimitive:
    query_value = intcon(0); // "normal" path produces false
    break;

  case vmIntrinsics::_getSuperclass:
    // The rules here are somewhat unfortunate, but we can still do better
    // with random logic than with a JNI call.
    // Interfaces store null or Object as _super, but must report null.
    // Arrays store an intermediate super as _super, but must report Object.
    // Other types can report the actual _super.
    // (To verify this code sequence, check the asserts in JVM_IsInterface.)
    if (generate_interface_guard(kls, region) != NULL)
      // A guard was added.  If the guard is taken, it was an interface.
      phi->add_req(null());
    if (generate_array_guard(kls, region) != NULL)
      // A guard was added.  If the guard is taken, it was an array.
      phi->add_req(makecon(TypeInstPtr::make(env()->Object_klass()->java_mirror())));
    // If we fall through, it's a plain class.  Get its _super.
    p = basic_plus_adr(kls, Klass::super_offset_in_bytes() + sizeof(oopDesc));
    kls = _gvn.transform(new (C, 3) LoadKlassNode(0, immutable_memory(), p, TypeRawPtr::BOTTOM, TypeKlassPtr::OBJECT_OR_NULL));
    null_ctl = top();
    kls = null_check_oop(kls, &null_ctl);
    if (null_ctl != top()) {
      // If the guard is taken, Object.superClass is null (both klass and mirror).
      region->add_req(null_ctl);
      phi   ->add_req(null());
    }
    if (!stopped()) {
      query_value = load_mirror_from_klass(kls);
    }
    break;

  case vmIntrinsics::_getComponentType:
    if (generate_array_guard(kls, region) != NULL) {
      // Be sure to pin the oop load to the guard edge just created:
      Node* is_array_ctrl = region->in(region->req()-1);
      Node* cma = basic_plus_adr(kls, in_bytes(arrayKlass::component_mirror_offset()) + sizeof(oopDesc));
      Node* cmo = make_load(is_array_ctrl, cma, TypeInstPtr::MIRROR, T_OBJECT);
      phi->add_req(cmo);
    }
    query_value = null();  // non-array case is null
    break;

  case vmIntrinsics::_getClassAccessFlags:
    p = basic_plus_adr(kls, Klass::access_flags_offset_in_bytes() + sizeof(oopDesc));
    query_value = make_load(NULL, p, TypeInt::INT, T_INT);
    break;

  default:
    ShouldNotReachHere();
  }

  // Fall-through is the normal case of a query to a real class.
  phi->init_req(1, query_value);
  region->init_req(1, control());

  push_result(region, phi);
  C->set_has_split_ifs(true); // Has chance for split-if optimization

  return true;
}

//--------------------------inline_native_subtype_check------------------------
// This intrinsic takes the JNI calls out of the heart of
// UnsafeFieldAccessorImpl.set, which improves Field.set, readObject, etc.
bool LibraryCallKit::inline_native_subtype_check() {
  int nargs = 1+1;  // the Class mirror, plus the other class getting examined

  // Pull both arguments off the stack.
  Node* args[2];                // two java.lang.Class mirrors: superc, subc
  args[0] = argument(0);
  args[1] = argument(1);
  Node* klasses[2];             // corresponding Klasses: superk, subk
  klasses[0] = klasses[1] = top();

  enum {
    // A full decision tree on {superc is prim, subc is prim}:
    _prim_0_path = 1,           // {P,N} => false
                                // {P,P} & superc!=subc => false
    _prim_same_path,            // {P,P} & superc==subc => true
    _prim_1_path,               // {N,P} => false
    _ref_subtype_path,          // {N,N} & subtype check wins => true
    _both_ref_path,             // {N,N} & subtype check loses => false
    PATH_LIMIT
  };

  RegionNode* region = new (C, PATH_LIMIT) RegionNode(PATH_LIMIT);
  Node*       phi    = new (C, PATH_LIMIT) PhiNode(region, TypeInt::BOOL);
  record_for_igvn(region);

  const TypePtr* adr_type = TypeRawPtr::BOTTOM;   // memory type of loads
  const TypeKlassPtr* kls_type = TypeKlassPtr::OBJECT_OR_NULL;
  int class_klass_offset = java_lang_Class::klass_offset_in_bytes();

  // First null-check both mirrors and load each mirror's klass metaobject.
  int which_arg;
  for (which_arg = 0; which_arg <= 1; which_arg++) {
    Node* arg = args[which_arg];
    _sp += nargs;  // set original stack for use by uncommon_trap
    arg = do_null_check(arg, T_OBJECT);
    _sp -= nargs;
    if (stopped())  break;
    args[which_arg] = _gvn.transform(arg);

    Node* p = basic_plus_adr(arg, class_klass_offset);
    Node* kls = new (C, 3) LoadKlassNode(0, immutable_memory(), p, adr_type, kls_type);
    klasses[which_arg] = _gvn.transform(kls);
  }

  // Having loaded both klasses, test each for null.
  bool never_see_null = !too_many_traps(Deoptimization::Reason_null_check);
  for (which_arg = 0; which_arg <= 1; which_arg++) {
    Node* kls = klasses[which_arg];
    Node* null_ctl = top();
    _sp += nargs;  // set original stack for use by uncommon_trap
    kls = null_check_oop(kls, &null_ctl, never_see_null);
    _sp -= nargs;
    int prim_path = (which_arg == 0 ? _prim_0_path : _prim_1_path);
    region->init_req(prim_path, null_ctl);
    if (stopped())  break;
    klasses[which_arg] = kls;
  }

  if (!stopped()) {
    // now we have two reference types, in klasses[0..1]
    Node* subk   = klasses[1];  // the argument to isAssignableFrom
    Node* superk = klasses[0];  // the receiver
    region->set_req(_both_ref_path, gen_subtype_check(subk, superk));
    // now we have a successful reference subtype check
    region->set_req(_ref_subtype_path, control());
  }

  // If both operands are primitive (both klasses null), then
  // we must return true when they are identical primitives.
  // It is convenient to test this after the first null klass check.
  set_control(region->in(_prim_0_path)); // go back to first null check
  if (!stopped()) {
    // Since superc is primitive, make a guard for the superc==subc case.
    Node* cmp_eq = _gvn.transform( new (C, 3) CmpPNode(args[0], args[1]) );
    Node* bol_eq = _gvn.transform( new (C, 2) BoolNode(cmp_eq, BoolTest::eq) );
    generate_guard(bol_eq, region, PROB_FAIR);
    if (region->req() == PATH_LIMIT+1) {
      // A guard was added.  If the added guard is taken, superc==subc.
      region->swap_edges(PATH_LIMIT, _prim_same_path);
      region->del_req(PATH_LIMIT);
    }
    region->set_req(_prim_0_path, control()); // Not equal after all.
  }

  // these are the only paths that produce 'true':
  phi->set_req(_prim_same_path,   intcon(1));
  phi->set_req(_ref_subtype_path, intcon(1));

  // pull together the cases:
  assert(region->req() == PATH_LIMIT, "sane region");
  for (uint i = 1; i < region->req(); i++) {
    Node* ctl = region->in(i);
    if (ctl == NULL || ctl == top()) {
      region->set_req(i, top());
      phi   ->set_req(i, top());
    } else if (phi->in(i) == NULL) {
      phi->set_req(i, intcon(0)); // all other paths produce 'false'
    }
  }

  set_control(_gvn.transform(region));
  push(_gvn.transform(phi));

  return true;
}

//---------------------generate_array_guard_common------------------------
Node* LibraryCallKit::generate_array_guard_common(Node* kls, RegionNode* region,
                                                  bool obj_array, bool not_array) {
  // If obj_array/non_array==false/false:
  // Branch around if the given klass is in fact an array (either obj or prim).
  // If obj_array/non_array==false/true:
  // Branch around if the given klass is not an array klass of any kind.
  // If obj_array/non_array==true/true:
  // Branch around if the kls is not an oop array (kls is int[], String, etc.)
  // If obj_array/non_array==true/false:
  // Branch around if the kls is an oop array (Object[] or subtype)
  //
  // Like generate_guard, adds a new path onto the region.
  jint  layout_con = 0;
  Node* layout_val = get_layout_helper(kls, layout_con);
  if (layout_val == NULL) {
    bool query = (obj_array
                  ? Klass::layout_helper_is_objArray(layout_con)
                  : Klass::layout_helper_is_javaArray(layout_con));
    if (query == not_array) {
      return NULL;                       // never a branch
    } else {                             // always a branch
      Node* always_branch = control();
      if (region != NULL)
        region->add_req(always_branch);
      set_control(top());
      return always_branch;
    }
  }
  // Now test the correct condition.
  jint  nval = (obj_array
                ? ((jint)Klass::_lh_array_tag_type_value
                   <<    Klass::_lh_array_tag_shift)
                : Klass::_lh_neutral_value);
  Node* cmp = _gvn.transform( new(C, 3) CmpINode(layout_val, intcon(nval)) );
  BoolTest::mask btest = BoolTest::lt;  // correct for testing is_[obj]array
  // invert the test if we are looking for a non-array
  if (not_array)  btest = BoolTest(btest).negate();
  Node* bol = _gvn.transform( new(C, 2) BoolNode(cmp, btest) );
  return generate_fair_guard(bol, region);
}


//-----------------------inline_native_newArray--------------------------
bool LibraryCallKit::inline_native_newArray() {
  int nargs = 2;
  Node* mirror    = argument(0);
  Node* count_val = argument(1);

  _sp += nargs;  // set original stack for use by uncommon_trap
  mirror = do_null_check(mirror, T_OBJECT);
  _sp -= nargs;

  enum { _normal_path = 1, _slow_path = 2, PATH_LIMIT };
  RegionNode* result_reg = new(C, PATH_LIMIT) RegionNode(PATH_LIMIT);
  PhiNode*    result_val = new(C, PATH_LIMIT) PhiNode(result_reg,
                                                      TypeInstPtr::NOTNULL);
  PhiNode*    result_io  = new(C, PATH_LIMIT) PhiNode(result_reg, Type::ABIO);
  PhiNode*    result_mem = new(C, PATH_LIMIT) PhiNode(result_reg, Type::MEMORY,
                                                      TypePtr::BOTTOM);

  bool never_see_null = !too_many_traps(Deoptimization::Reason_null_check);
  Node* klass_node = load_array_klass_from_mirror(mirror, never_see_null,
                                                  nargs,
                                                  result_reg, _slow_path);
  Node* normal_ctl   = control();
  Node* no_array_ctl = result_reg->in(_slow_path);

  // Generate code for the slow case.  We make a call to newArray().
  set_control(no_array_ctl);
  if (!stopped()) {
    // Either the input type is void.class, or else the
    // array klass has not yet been cached.  Either the
    // ensuing call will throw an exception, or else it
    // will cache the array klass for next time.
    PreserveJVMState pjvms(this);
    CallJavaNode* slow_call = generate_method_call_static(vmIntrinsics::_newArray);
    Node* slow_result = set_results_for_java_call(slow_call);
    // this->control() comes from set_results_for_java_call
    result_reg->set_req(_slow_path, control());
    result_val->set_req(_slow_path, slow_result);
    result_io ->set_req(_slow_path, i_o());
    result_mem->set_req(_slow_path, reset_memory());
  }

  set_control(normal_ctl);
  if (!stopped()) {
    // Normal case:  The array type has been cached in the java.lang.Class.
    // The following call works fine even if the array type is polymorphic.
    // It could be a dynamic mix of int[], boolean[], Object[], etc.
    _sp += nargs;  // set original stack for use by uncommon_trap
    Node* obj = new_array(klass_node, count_val);
    _sp -= nargs;
    result_reg->init_req(_normal_path, control());
    result_val->init_req(_normal_path, obj);
    result_io ->init_req(_normal_path, i_o());
    result_mem->init_req(_normal_path, reset_memory());
  }

  // Return the combined state.
  set_i_o(        _gvn.transform(result_io)  );
  set_all_memory( _gvn.transform(result_mem) );
  push_result(result_reg, result_val);
  C->set_has_split_ifs(true); // Has chance for split-if optimization

  return true;
}

//----------------------inline_native_getLength--------------------------
bool LibraryCallKit::inline_native_getLength() {
  if (too_many_traps(Deoptimization::Reason_intrinsic))  return false;

  int nargs = 1;
  Node* array = argument(0);

  _sp += nargs;  // set original stack for use by uncommon_trap
  array = do_null_check(array, T_OBJECT);
  _sp -= nargs;

  // If array is dead, only null-path is taken.
  if (stopped())  return true;

  // Deoptimize if it is a non-array.
  Node* non_array = generate_non_array_guard(load_object_klass(array), NULL);

  if (non_array != NULL) {
    PreserveJVMState pjvms(this);
    set_control(non_array);
    _sp += nargs;  // push the arguments back on the stack
    uncommon_trap(Deoptimization::Reason_intrinsic,
                  Deoptimization::Action_maybe_recompile);
  }

  // If control is dead, only non-array-path is taken.
  if (stopped())  return true;

  // The works fine even if the array type is polymorphic.
  // It could be a dynamic mix of int[], boolean[], Object[], etc.
  push( load_array_length(array) );

  C->set_has_split_ifs(true); // Has chance for split-if optimization

  return true;
}

//------------------------inline_array_copyOf----------------------------
bool LibraryCallKit::inline_array_copyOf(bool is_copyOfRange) {
  if (too_many_traps(Deoptimization::Reason_intrinsic))  return false;

  // Restore the stack and pop off the arguments.
  int nargs = 3 + (is_copyOfRange? 1: 0);
  Node* original          = argument(0);
  Node* start             = is_copyOfRange? argument(1): intcon(0);
  Node* end               = is_copyOfRange? argument(2): argument(1);
  Node* array_type_mirror = is_copyOfRange? argument(3): argument(2);

  _sp += nargs;  // set original stack for use by uncommon_trap
  array_type_mirror = do_null_check(array_type_mirror, T_OBJECT);
  original          = do_null_check(original, T_OBJECT);
  _sp -= nargs;

  // Check if a null path was taken unconditionally.
  if (stopped())  return true;

  Node* orig_length = load_array_length(original);

  Node* klass_node = load_klass_from_mirror(array_type_mirror, false, nargs,
                                            NULL, 0);
  _sp += nargs;  // set original stack for use by uncommon_trap
  klass_node = do_null_check(klass_node, T_OBJECT);
  _sp -= nargs;

  RegionNode* bailout = new (C, 1) RegionNode(1);
  record_for_igvn(bailout);

  // Despite the generic type of Arrays.copyOf, the mirror might be int, int[], etc.
  // Bail out if that is so.
  Node* not_objArray = generate_non_objArray_guard(klass_node, bailout);
  if (not_objArray != NULL) {
    // Improve the klass node's type from the new optimistic assumption:
    ciKlass* ak = ciArrayKlass::make(env()->Object_klass());
    const Type* akls = TypeKlassPtr::make(TypePtr::NotNull, ak, 0/*offset*/);
    Node* cast = new (C, 2) CastPPNode(klass_node, akls);
    cast->init_req(0, control());
    klass_node = _gvn.transform(cast);
  }

  // Bail out if either start or end is negative.
  generate_negative_guard(start, bailout, &start);
  generate_negative_guard(end,   bailout, &end);

  Node* length = end;
  if (_gvn.type(start) != TypeInt::ZERO) {
    length = _gvn.transform( new (C, 3) SubINode(end, start) );
  }

  // Bail out if length is negative.
  // ...Not needed, since the new_array will throw the right exception.
  //generate_negative_guard(length, bailout, &length);

  if (bailout->req() > 1) {
    PreserveJVMState pjvms(this);
    set_control( _gvn.transform(bailout) );
    _sp += nargs;  // push the arguments back on the stack
    uncommon_trap(Deoptimization::Reason_intrinsic,
                  Deoptimization::Action_maybe_recompile);
  }

  if (!stopped()) {
    // How many elements will we copy from the original?
    // The answer is MinI(orig_length - start, length).
    Node* orig_tail = _gvn.transform( new(C, 3) SubINode(orig_length, start) );
    Node* moved = generate_min_max(vmIntrinsics::_min, orig_tail, length);

    _sp += nargs;  // set original stack for use by uncommon_trap
    Node* newcopy = new_array(klass_node, length);
    _sp -= nargs;

    // Generate a direct call to the right arraycopy function(s).
    // We know the copy is disjoint but we might not know if the
    // oop stores need checking.
    // Extreme case:  Arrays.copyOf((Integer[])x, 10, String[].class).
    // This will fail a store-check if x contains any non-nulls.
    bool disjoint_bases = true;
    bool length_never_negative = true;
    generate_arraycopy(TypeAryPtr::OOPS, T_OBJECT,
                       original, start, newcopy, intcon(0), moved,
                       nargs, disjoint_bases, length_never_negative);

    push(newcopy);
  }

  C->set_has_split_ifs(true); // Has chance for split-if optimization

  return true;
}


//----------------------generate_virtual_guard---------------------------
// Helper for hashCode and clone.  Peeks inside the vtable to avoid a call.
Node* LibraryCallKit::generate_virtual_guard(Node* obj_klass,
                                             RegionNode* slow_region) {
  ciMethod* method = callee();
  int vtable_index = method->vtable_index();
  // Get the methodOop out of the appropriate vtable entry.
  int entry_offset  = (instanceKlass::vtable_start_offset() +
                     vtable_index*vtableEntry::size()) * wordSize +
                     vtableEntry::method_offset_in_bytes();
  Node* entry_addr  = basic_plus_adr(obj_klass, entry_offset);
  Node* target_call = make_load(NULL, entry_addr, TypeInstPtr::NOTNULL, T_OBJECT);

  // Compare the target method with the expected method (e.g., Object.hashCode).
  const TypeInstPtr* native_call_addr = TypeInstPtr::make(method);

  Node* native_call = makecon(native_call_addr);
  Node* chk_native  = _gvn.transform( new(C, 3) CmpPNode(target_call, native_call) );
  Node* test_native = _gvn.transform( new(C, 2) BoolNode(chk_native, BoolTest::ne) );

  return generate_slow_guard(test_native, slow_region);
}

//-----------------------generate_method_call----------------------------
// Use generate_method_call to make a slow-call to the real
// method if the fast path fails.  An alternative would be to
// use a stub like OptoRuntime::slow_arraycopy_Java.
// This only works for expanding the current library call,
// not another intrinsic.  (E.g., don't use this for making an
// arraycopy call inside of the copyOf intrinsic.)
CallJavaNode*
LibraryCallKit::generate_method_call(vmIntrinsics::ID method_id, bool is_virtual, bool is_static) {
  // When compiling the intrinsic method itself, do not use this technique.
  guarantee(callee() != C->method(), "cannot make slow-call to self");

  ciMethod* method = callee();
  // ensure the JVMS we have will be correct for this call
  guarantee(method_id == method->intrinsic_id(), "must match");

  const TypeFunc* tf = TypeFunc::make(method);
  int tfdc = tf->domain()->cnt();
  CallJavaNode* slow_call;
  if (is_static) {
    assert(!is_virtual, "");
    slow_call = new(C, tfdc) CallStaticJavaNode(tf,
                                SharedRuntime::get_resolve_static_call_stub(),
                                method, bci());
  } else if (is_virtual) {
    null_check_receiver(method);
    int vtable_index = methodOopDesc::invalid_vtable_index;
    if (UseInlineCaches) {
      // Suppress the vtable call
    } else {
      // hashCode and clone are not a miranda methods,
      // so the vtable index is fixed.
      // No need to use the linkResolver to get it.
       vtable_index = method->vtable_index();
    }
    slow_call = new(C, tfdc) CallDynamicJavaNode(tf,
                                SharedRuntime::get_resolve_virtual_call_stub(),
                                method, vtable_index, bci());
  } else {  // neither virtual nor static:  opt_virtual
    null_check_receiver(method);
    slow_call = new(C, tfdc) CallStaticJavaNode(tf,
                                SharedRuntime::get_resolve_opt_virtual_call_stub(),
                                method, bci());
    slow_call->set_optimized_virtual(true);
  }
  set_arguments_for_java_call(slow_call);
  set_edges_for_java_call(slow_call);
  return slow_call;
}


//------------------------------inline_native_hashcode--------------------
// Build special case code for calls to hashCode on an object.
bool LibraryCallKit::inline_native_hashcode(bool is_virtual, bool is_static) {
  assert(is_static == callee()->is_static(), "correct intrinsic selection");
  assert(!(is_virtual && is_static), "either virtual, special, or static");

  enum { _slow_path = 1, _fast_path, _null_path, PATH_LIMIT };

  RegionNode* result_reg = new(C, PATH_LIMIT) RegionNode(PATH_LIMIT);
  PhiNode*    result_val = new(C, PATH_LIMIT) PhiNode(result_reg,
                                                      TypeInt::INT);
  PhiNode*    result_io  = new(C, PATH_LIMIT) PhiNode(result_reg, Type::ABIO);
  PhiNode*    result_mem = new(C, PATH_LIMIT) PhiNode(result_reg, Type::MEMORY,
                                                      TypePtr::BOTTOM);
  Node* obj = NULL;
  if (!is_static) {
    // Check for hashing null object
    obj = null_check_receiver(callee());
    if (stopped())  return true;        // unconditionally null
    result_reg->init_req(_null_path, top());
    result_val->init_req(_null_path, top());
  } else {
    // Do a null check, and return zero if null.
    // System.identityHashCode(null) == 0
    obj = argument(0);
    Node* null_ctl = top();
    obj = null_check_oop(obj, &null_ctl);
    result_reg->init_req(_null_path, null_ctl);
    result_val->init_req(_null_path, _gvn.intcon(0));
  }

  // Unconditionally null?  Then return right away.
  if (stopped()) {
    set_control( result_reg->in(_null_path) );
    if (!stopped())
      push(      result_val ->in(_null_path) );
    return true;
  }

  // After null check, get the object's klass.
  Node* obj_klass = load_object_klass(obj);

  // This call may be virtual (invokevirtual) or bound (invokespecial).
  // For each case we generate slightly different code.

  // We only go to the fast case code if we pass a number of guards.  The
  // paths which do not pass are accumulated in the slow_region.
  RegionNode* slow_region = new (C, 1) RegionNode(1);
  record_for_igvn(slow_region);

  // If this is a virtual call, we generate a funny guard.  We pull out
  // the vtable entry corresponding to hashCode() from the target object.
  // If the target method which we are calling happens to be the native
  // Object hashCode() method, we pass the guard.  We do not need this
  // guard for non-virtual calls -- the caller is known to be the native
  // Object hashCode().
  if (is_virtual) {
    generate_virtual_guard(obj_klass, slow_region);
  }

  // Get the header out of the object, use LoadMarkNode when available
  Node* header_addr = basic_plus_adr(obj, oopDesc::mark_offset_in_bytes());
  Node* header = make_load(NULL, header_addr, TypeRawPtr::BOTTOM, T_ADDRESS);
  header = _gvn.transform( new (C, 2) CastP2XNode(NULL, header) );

  // Test the header to see if it is unlocked.
  Node *lock_mask      = _gvn.MakeConX(markOopDesc::biased_lock_mask_in_place);
  Node *lmasked_header = _gvn.transform( new (C, 3) AndXNode(header, lock_mask) );
  Node *unlocked_val   = _gvn.MakeConX(markOopDesc::unlocked_value);
  Node *chk_unlocked   = _gvn.transform( new (C, 3) CmpXNode( lmasked_header, unlocked_val));
  Node *test_unlocked  = _gvn.transform( new (C, 2) BoolNode( chk_unlocked, BoolTest::ne) );

  generate_slow_guard(test_unlocked, slow_region);

  // Get the hash value and check to see that it has been properly assigned.
  // We depend on hash_mask being at most 32 bits and avoid the use of
  // hash_mask_in_place because it could be larger than 32 bits in a 64-bit
  // vm: see markOop.hpp.
  Node *hash_mask      = _gvn.intcon(markOopDesc::hash_mask);
  Node *hash_shift     = _gvn.intcon(markOopDesc::hash_shift);
  Node *hshifted_header= _gvn.transform( new (C, 3) URShiftXNode(header, hash_shift) );
  // This hack lets the hash bits live anywhere in the mark object now, as long
  // as the shift drops the relevent bits into the low 32 bits.  Note that
  // Java spec says that HashCode is an int so there's no point in capturing
  // an 'X'-sized hashcode (32 in 32-bit build or 64 in 64-bit build).
  hshifted_header      = ConvX2I(hshifted_header);
  Node *hash_val       = _gvn.transform( new (C, 3) AndINode(hshifted_header, hash_mask) );

  Node *no_hash_val    = _gvn.intcon(markOopDesc::no_hash);
  Node *chk_assigned   = _gvn.transform( new (C, 3) CmpINode( hash_val, no_hash_val));
  Node *test_assigned  = _gvn.transform( new (C, 2) BoolNode( chk_assigned, BoolTest::eq) );

  generate_slow_guard(test_assigned, slow_region);

  Node* init_mem = reset_memory();
  // fill in the rest of the null path:
  result_io ->init_req(_null_path, i_o());
  result_mem->init_req(_null_path, init_mem);

  result_val->init_req(_fast_path, hash_val);
  result_reg->init_req(_fast_path, control());
  result_io ->init_req(_fast_path, i_o());
  result_mem->init_req(_fast_path, init_mem);

  // Generate code for the slow case.  We make a call to hashCode().
  set_control(_gvn.transform(slow_region));
  if (!stopped()) {
    // No need for PreserveJVMState, because we're using up the present state.
    set_all_memory(init_mem);
    vmIntrinsics::ID hashCode_id = vmIntrinsics::_hashCode;
    if (is_static)   hashCode_id = vmIntrinsics::_identityHashCode;
    CallJavaNode* slow_call = generate_method_call(hashCode_id, is_virtual, is_static);
    Node* slow_result = set_results_for_java_call(slow_call);
    // this->control() comes from set_results_for_java_call
    result_reg->init_req(_slow_path, control());
    result_val->init_req(_slow_path, slow_result);
    result_io  ->set_req(_slow_path, i_o());
    result_mem ->set_req(_slow_path, reset_memory());
  }

  // Return the combined state.
  set_i_o(        _gvn.transform(result_io)  );
  set_all_memory( _gvn.transform(result_mem) );
  push_result(result_reg, result_val);

  return true;
}

//---------------------------inline_native_getClass----------------------------
// Build special case code for calls to hashCode on an object.
bool LibraryCallKit::inline_native_getClass() {
  Node* obj = null_check_receiver(callee());
  if (stopped())  return true;
  push( load_mirror_from_klass(load_object_klass(obj)) );
  return true;
}

//-----------------inline_native_Reflection_getCallerClass---------------------
// In the presence of deep enough inlining, getCallerClass() becomes a no-op.
//
// NOTE that this code must perform the same logic as
// vframeStream::security_get_caller_frame in that it must skip
// Method.invoke() and auxiliary frames.




bool LibraryCallKit::inline_native_Reflection_getCallerClass() {
  ciMethod*       method = callee();

#ifndef PRODUCT
  if ((PrintIntrinsics || PrintInlining || PrintOptoInlining) && Verbose) {
    tty->print_cr("Attempting to inline sun.reflect.Reflection.getCallerClass");
  }
#endif

  debug_only(int saved_sp = _sp);

  // Argument words:  (int depth)
  int nargs = 1;

  _sp += nargs;
  Node* caller_depth_node = pop();

  assert(saved_sp == _sp, "must have correct argument count");

  // The depth value must be a constant in order for the runtime call
  // to be eliminated.
  const TypeInt* caller_depth_type = _gvn.type(caller_depth_node)->isa_int();
  if (caller_depth_type == NULL || !caller_depth_type->is_con()) {
#ifndef PRODUCT
    if ((PrintIntrinsics || PrintInlining || PrintOptoInlining) && Verbose) {
      tty->print_cr("  Bailing out because caller depth was not a constant");
    }
#endif
    return false;
  }
  // Note that the JVM state at this point does not include the
  // getCallerClass() frame which we are trying to inline. The
  // semantics of getCallerClass(), however, are that the "first"
  // frame is the getCallerClass() frame, so we subtract one from the
  // requested depth before continuing. We don't inline requests of
  // getCallerClass(0).
  int caller_depth = caller_depth_type->get_con() - 1;
  if (caller_depth < 0) {
#ifndef PRODUCT
    if ((PrintIntrinsics || PrintInlining || PrintOptoInlining) && Verbose) {
      tty->print_cr("  Bailing out because caller depth was %d", caller_depth);
    }
#endif
    return false;
  }

  if (!jvms()->has_method()) {
#ifndef PRODUCT
    if ((PrintIntrinsics || PrintInlining || PrintOptoInlining) && Verbose) {
      tty->print_cr("  Bailing out because intrinsic was inlined at top level");
    }
#endif
    return false;
  }
  int _depth = jvms()->depth();  // cache call chain depth

  // Walk back up the JVM state to find the caller at the required
  // depth. NOTE that this code must perform the same logic as
  // vframeStream::security_get_caller_frame in that it must skip
  // Method.invoke() and auxiliary frames. Note also that depth is
  // 1-based (1 is the bottom of the inlining).
  int inlining_depth = _depth;
  JVMState* caller_jvms = NULL;

  if (inlining_depth > 0) {
    caller_jvms = jvms();
    assert(caller_jvms = jvms()->of_depth(inlining_depth), "inlining_depth == our depth");
    do {
      // The following if-tests should be performed in this order
      if (is_method_invoke_or_aux_frame(caller_jvms)) {
        // Skip a Method.invoke() or auxiliary frame
      } else if (caller_depth > 0) {
        // Skip real frame
        --caller_depth;
      } else {
        // We're done: reached desired caller after skipping.
        break;
      }
      caller_jvms = caller_jvms->caller();
      --inlining_depth;
    } while (inlining_depth > 0);
  }

  if (inlining_depth == 0) {
#ifndef PRODUCT
    if ((PrintIntrinsics || PrintInlining || PrintOptoInlining) && Verbose) {
      tty->print_cr("  Bailing out because caller depth (%d) exceeded inlining depth (%d)", caller_depth_type->get_con(), _depth);
      tty->print_cr("  JVM state at this point:");
      for (int i = _depth; i >= 1; i--) {
        tty->print_cr("   %d) %s", i, jvms()->of_depth(i)->method()->name()->as_utf8());
      }
    }
#endif
    return false; // Reached end of inlining
  }

  // Acquire method holder as java.lang.Class
  ciInstanceKlass* caller_klass  = caller_jvms->method()->holder();
  ciInstance*      caller_mirror = caller_klass->java_mirror();
  // Push this as a constant
  push(makecon(TypeInstPtr::make(caller_mirror)));
#ifndef PRODUCT
  if ((PrintIntrinsics || PrintInlining || PrintOptoInlining) && Verbose) {
    tty->print_cr("  Succeeded: caller = %s.%s, caller depth = %d, depth = %d", caller_klass->name()->as_utf8(), caller_jvms->method()->name()->as_utf8(), caller_depth_type->get_con(), _depth);
    tty->print_cr("  JVM state at this point:");
    for (int i = _depth; i >= 1; i--) {
      tty->print_cr("   %d) %s", i, jvms()->of_depth(i)->method()->name()->as_utf8());
    }
  }
#endif
  return true;
}

// Helper routine for above
bool LibraryCallKit::is_method_invoke_or_aux_frame(JVMState* jvms) {
  // Is this the Method.invoke method itself?
  if (jvms->method()->intrinsic_id() == vmIntrinsics::_invoke)
    return true;

  // Is this a helper, defined somewhere underneath MethodAccessorImpl.
  ciKlass* k = jvms->method()->holder();
  if (k->is_instance_klass()) {
    ciInstanceKlass* ik = k->as_instance_klass();
    for (; ik != NULL; ik = ik->super()) {
      if (ik->name() == ciSymbol::sun_reflect_MethodAccessorImpl() &&
          ik == env()->find_system_klass(ik->name())) {
        return true;
      }
    }
  }

  return false;
}

static int value_field_offset = -1;  // offset of the "value" field of AtomicLongCSImpl.  This is needed by
                                     // inline_native_AtomicLong_attemptUpdate() but it has no way of
                                     // computing it since there is no lookup field by name function in the
                                     // CI interface.  This is computed and set by inline_native_AtomicLong_get().
                                     // Using a static variable here is safe even if we have multiple compilation
                                     // threads because the offset is constant.  At worst the same offset will be
                                     // computed and  stored multiple

bool LibraryCallKit::inline_native_AtomicLong_get() {
  // Restore the stack and pop off the argument
  _sp+=1;
  Node *obj = pop();

  // get the offset of the "value" field. Since the CI interfaces
  // does not provide a way to look up a field by name, we scan the bytecodes
  // to get the field index.  We expect the first 2 instructions of the method
  // to be:
  //    0 aload_0
  //    1 getfield "value"
  ciMethod* method = callee();
  if (value_field_offset == -1)
  {
    ciField* value_field;
    ciBytecodeStream iter(method);
    Bytecodes::Code bc = iter.next();

    if ((bc != Bytecodes::_aload_0) &&
              ((bc != Bytecodes::_aload) || (iter.get_index() != 0)))
      return false;
    bc = iter.next();
    if (bc != Bytecodes::_getfield)
      return false;
    bool ignore;
    value_field = iter.get_field(ignore);
    value_field_offset = value_field->offset_in_bytes();
  }

  // Null check without removing any arguments.
  _sp++;
  obj = do_null_check(obj, T_OBJECT);
  _sp--;
  // Check for locking null object
  if (stopped()) return true;

  Node *adr = basic_plus_adr(obj, obj, value_field_offset);
  const TypePtr *adr_type = _gvn.type(adr)->is_ptr();
  int alias_idx = C->get_alias_index(adr_type);

  Node *result = _gvn.transform(new (C, 3) LoadLLockedNode(control(), memory(alias_idx), adr));

  push_pair(result);

  return true;
}

bool LibraryCallKit::inline_native_AtomicLong_attemptUpdate() {
  // Restore the stack and pop off the arguments
  _sp+=5;
  Node *newVal = pop_pair();
  Node *oldVal = pop_pair();
  Node *obj = pop();

  // we need the offset of the "value" field which was computed when
  // inlining the get() method.  Give up if we don't have it.
  if (value_field_offset == -1)
    return false;

  // Null check without removing any arguments.
  _sp+=5;
  obj = do_null_check(obj, T_OBJECT);
  _sp-=5;
  // Check for locking null object
  if (stopped()) return true;

  Node *adr = basic_plus_adr(obj, obj, value_field_offset);
  const TypePtr *adr_type = _gvn.type(adr)->is_ptr();
  int alias_idx = C->get_alias_index(adr_type);

  Node *result = _gvn.transform(new (C, 5) StoreLConditionalNode(control(), memory(alias_idx), adr, newVal, oldVal));
  Node *store_proj = _gvn.transform( new (C, 1) SCMemProjNode(result));
  set_memory(store_proj, alias_idx);

  push(result);
  return true;
}

bool LibraryCallKit::inline_fp_conversions(vmIntrinsics::ID id) {
  // restore the arguments
  _sp += arg_size();

  switch (id) {
  case vmIntrinsics::_floatToRawIntBits:
    push(_gvn.transform( new (C, 2) MoveF2INode(pop())));
    break;

  case vmIntrinsics::_intBitsToFloat:
    push(_gvn.transform( new (C, 2) MoveI2FNode(pop())));
    break;

  case vmIntrinsics::_doubleToRawLongBits:
    push_pair(_gvn.transform( new (C, 2) MoveD2LNode(pop_pair())));
    break;

  case vmIntrinsics::_longBitsToDouble:
    push_pair(_gvn.transform( new (C, 2) MoveL2DNode(pop_pair())));
    break;

  case vmIntrinsics::_doubleToLongBits: {
    Node* value = pop_pair();

    // two paths (plus control) merge in a wood
    RegionNode *r = new (C, 3) RegionNode(3);
    Node *phi = new (C, 3) PhiNode(r, TypeLong::LONG);

    Node *cmpisnan = _gvn.transform( new (C, 3) CmpDNode(value, value));
    // Build the boolean node
    Node *bolisnan = _gvn.transform( new (C, 2) BoolNode( cmpisnan, BoolTest::ne ) );

    // Branch either way.
    // NaN case is less traveled, which makes all the difference.
    IfNode *ifisnan = create_and_xform_if(control(), bolisnan, PROB_STATIC_FREQUENT, COUNT_UNKNOWN);
    Node *opt_isnan = _gvn.transform(ifisnan);
    assert( opt_isnan->is_If(), "Expect an IfNode");
    IfNode *opt_ifisnan = (IfNode*)opt_isnan;
    Node *iftrue = _gvn.transform( new (C, 1) IfTrueNode(opt_ifisnan) );

    set_control(iftrue);

    static const jlong nan_bits = CONST64(0x7ff8000000000000);
    Node *slow_result = longcon(nan_bits); // return NaN
    phi->init_req(1, _gvn.transform( slow_result ));
    r->init_req(1, iftrue);

    // Else fall through
    Node *iffalse = _gvn.transform( new (C, 1) IfFalseNode(opt_ifisnan) );
    set_control(iffalse);

    phi->init_req(2, _gvn.transform( new (C, 2) MoveD2LNode(value)));
    r->init_req(2, iffalse);

    // Post merge
    set_control(_gvn.transform(r));
    record_for_igvn(r);

    Node* result = _gvn.transform(phi);
    assert(result->bottom_type()->isa_long(), "must be");
    push_pair(result);

    C->set_has_split_ifs(true); // Has chance for split-if optimization

    break;
  }

  case vmIntrinsics::_floatToIntBits: {
    Node* value = pop();

    // two paths (plus control) merge in a wood
    RegionNode *r = new (C, 3) RegionNode(3);
    Node *phi = new (C, 3) PhiNode(r, TypeInt::INT);

    Node *cmpisnan = _gvn.transform( new (C, 3) CmpFNode(value, value));
    // Build the boolean node
    Node *bolisnan = _gvn.transform( new (C, 2) BoolNode( cmpisnan, BoolTest::ne ) );

    // Branch either way.
    // NaN case is less traveled, which makes all the difference.
    IfNode *ifisnan = create_and_xform_if(control(), bolisnan, PROB_STATIC_FREQUENT, COUNT_UNKNOWN);
    Node *opt_isnan = _gvn.transform(ifisnan);
    assert( opt_isnan->is_If(), "Expect an IfNode");
    IfNode *opt_ifisnan = (IfNode*)opt_isnan;
    Node *iftrue = _gvn.transform( new (C, 1) IfTrueNode(opt_ifisnan) );

    set_control(iftrue);

    static const jint nan_bits = 0x7fc00000;
    Node *slow_result = makecon(TypeInt::make(nan_bits)); // return NaN
    phi->init_req(1, _gvn.transform( slow_result ));
    r->init_req(1, iftrue);

    // Else fall through
    Node *iffalse = _gvn.transform( new (C, 1) IfFalseNode(opt_ifisnan) );
    set_control(iffalse);

    phi->init_req(2, _gvn.transform( new (C, 2) MoveF2INode(value)));
    r->init_req(2, iffalse);

    // Post merge
    set_control(_gvn.transform(r));
    record_for_igvn(r);

    Node* result = _gvn.transform(phi);
    assert(result->bottom_type()->isa_int(), "must be");
    push(result);

    C->set_has_split_ifs(true); // Has chance for split-if optimization

    break;
  }

  default:
    ShouldNotReachHere();
  }

  return true;
}

#ifdef _LP64
#define XTOP ,top() /*additional argument*/
#else  //_LP64
#define XTOP        /*no additional argument*/
#endif //_LP64

//----------------------inline_unsafe_copyMemory-------------------------
bool LibraryCallKit::inline_unsafe_copyMemory() {
  if (callee()->is_static())  return false;  // caller must have the capability!
  int nargs = 1 + 5 + 3;  // 5 args:  (src: ptr,off, dst: ptr,off, size)
  assert(signature()->size() == nargs-1, "copy has 5 arguments");
  null_check_receiver(callee());  // check then ignore argument(0)
  if (stopped())  return true;

  C->set_has_unsafe_access(true);  // Mark eventual nmethod as "unsafe".

  Node* src_ptr = argument(1);
  Node* src_off = ConvL2X(argument(2));
  assert(argument(3)->is_top(), "2nd half of long");
  Node* dst_ptr = argument(4);
  Node* dst_off = ConvL2X(argument(5));
  assert(argument(6)->is_top(), "2nd half of long");
  Node* size    = ConvL2X(argument(7));
  assert(argument(8)->is_top(), "2nd half of long");

  assert(Unsafe_field_offset_to_byte_offset(11) == 11,
         "fieldOffset must be byte-scaled");

  Node* src = make_unsafe_address(src_ptr, src_off);
  Node* dst = make_unsafe_address(dst_ptr, dst_off);

  // Conservatively insert a memory barrier on all memory slices.
  // Do not let writes of the copy source or destination float below the copy.
  insert_mem_bar(Op_MemBarCPUOrder);

  // Call it.  Note that the length argument is not scaled.
  make_runtime_call(RC_LEAF|RC_NO_FP,
                    OptoRuntime::fast_arraycopy_Type(),
                    StubRoutines::unsafe_arraycopy(),
                    "unsafe_arraycopy",
                    TypeRawPtr::BOTTOM,
                    src, dst, size XTOP);

  // Do not let reads of the copy destination float above the copy.
  insert_mem_bar(Op_MemBarCPUOrder);

  return true;
}


//------------------------inline_native_clone----------------------------
// Here are the simple edge cases:
//  null receiver => normal trap
//  virtual and clone was overridden => slow path to out-of-line clone
//  not cloneable or finalizer => slow path to out-of-line Object.clone
//
// The general case has two steps, allocation and copying.
// Allocation has two cases, and uses GraphKit::new_instance or new_array.
//
// Copying also has two cases, oop arrays and everything else.
// Oop arrays use arrayof_oop_arraycopy (same as System.arraycopy).
// Everything else uses the tight inline loop supplied by CopyArrayNode.
//
// These steps fold up nicely if and when the cloned object's klass
// can be sharply typed as an object array, a type array, or an instance.
//
bool LibraryCallKit::inline_native_clone(bool is_virtual) {
  int nargs = 1;
  Node* obj = null_check_receiver(callee());
  if (stopped())  return true;
  Node* obj_klass = load_object_klass(obj);
  const TypeKlassPtr* tklass = _gvn.type(obj_klass)->isa_klassptr();
  const TypeOopPtr*   toop   = ((tklass != NULL)
                                ? tklass->as_instance_type()
                                : TypeInstPtr::NOTNULL);

  // Conservatively insert a memory barrier on all memory slices.
  // Do not let writes into the original float below the clone.
  insert_mem_bar(Op_MemBarCPUOrder);

  // paths into result_reg:
  enum {
    _slow_path = 1,     // out-of-line call to clone method (virtual or not)
    _objArray_path,     // plain allocation, plus arrayof_oop_arraycopy
    _fast_path,         // plain allocation, plus a CopyArray operation
    PATH_LIMIT
  };
  RegionNode* result_reg = new(C, PATH_LIMIT) RegionNode(PATH_LIMIT);
  PhiNode*    result_val = new(C, PATH_LIMIT) PhiNode(result_reg,
                                                      TypeInstPtr::NOTNULL);
  PhiNode*    result_i_o = new(C, PATH_LIMIT) PhiNode(result_reg, Type::ABIO);
  PhiNode*    result_mem = new(C, PATH_LIMIT) PhiNode(result_reg, Type::MEMORY,
                                                      TypePtr::BOTTOM);
  record_for_igvn(result_reg);

  const TypePtr* raw_adr_type = TypeRawPtr::BOTTOM;
  int raw_adr_idx = Compile::AliasIdxRaw;
  const bool raw_mem_only = true;

  // paths into alloc_reg (on the fast path, just before the CopyArray):
  enum { _typeArray_alloc = 1, _instance_alloc, ALLOC_LIMIT };
  RegionNode* alloc_reg = new(C, ALLOC_LIMIT) RegionNode(ALLOC_LIMIT);
  PhiNode*    alloc_val = new(C, ALLOC_LIMIT) PhiNode(alloc_reg, raw_adr_type);
  PhiNode*    alloc_siz = new(C, ALLOC_LIMIT) PhiNode(alloc_reg, TypeX_X);
  PhiNode*    alloc_i_o = new(C, ALLOC_LIMIT) PhiNode(alloc_reg, Type::ABIO);
  PhiNode*    alloc_mem = new(C, ALLOC_LIMIT) PhiNode(alloc_reg, Type::MEMORY,
                                                      raw_adr_type);
  record_for_igvn(alloc_reg);

  bool card_mark = false;  // (see below)

  Node* array_ctl = generate_array_guard(obj_klass, (RegionNode*)NULL);
  if (array_ctl != NULL) {
    // It's an array.
    PreserveJVMState pjvms(this);
    set_control(array_ctl);
    Node* obj_length = load_array_length(obj);
    Node* obj_size = NULL;
    _sp += nargs;  // set original stack for use by uncommon_trap
    Node* alloc_obj = new_array(obj_klass, obj_length,
                                raw_mem_only, &obj_size);
    _sp -= nargs;
    assert(obj_size != NULL, "");
    Node* raw_obj = alloc_obj->in(1);
    assert(raw_obj->is_Proj() && raw_obj->in(0)->is_Allocate(), "");
    if (ReduceBulkZeroing) {
      AllocateNode* alloc = AllocateNode::Ideal_allocation(alloc_obj, &_gvn);
      if (alloc != NULL) {
        // We will be completely responsible for initializing this object.
        alloc->maybe_set_complete(&_gvn);
      }
    }

    if (!use_ReduceInitialCardMarks()) {
      // If it is an oop array, it requires very special treatment,
      // because card marking is required on each card of the array.
      Node* is_obja = generate_objArray_guard(obj_klass, (RegionNode*)NULL);
      if (is_obja != NULL) {
        PreserveJVMState pjvms2(this);
        set_control(is_obja);
        // Generate a direct call to the right arraycopy function(s).
        bool disjoint_bases = true;
        bool length_never_negative = true;
        generate_arraycopy(TypeAryPtr::OOPS, T_OBJECT,
                           obj, intcon(0), alloc_obj, intcon(0),
                           obj_length, nargs,
                           disjoint_bases, length_never_negative);
        result_reg->init_req(_objArray_path, control());
        result_val->init_req(_objArray_path, alloc_obj);
        result_i_o ->set_req(_objArray_path, i_o());
        result_mem ->set_req(_objArray_path, reset_memory());
      }
    }
    // We can dispense with card marks if we know the allocation
    // comes out of eden (TLAB)...  In fact, ReduceInitialCardMarks
    // causes the non-eden paths to simulate a fresh allocation,
    // insofar that no further card marks are required to initialize
    // the object.

    // Otherwise, there are no card marks to worry about.
    alloc_val->init_req(_typeArray_alloc, raw_obj);
    alloc_siz->init_req(_typeArray_alloc, obj_size);
    alloc_reg->init_req(_typeArray_alloc, control());
    alloc_i_o->init_req(_typeArray_alloc, i_o());
    alloc_mem->init_req(_typeArray_alloc, memory(raw_adr_type));
  }

  // We only go to the fast case code if we pass a number of guards.
  // The paths which do not pass are accumulated in the slow_region.
  RegionNode* slow_region = new (C, 1) RegionNode(1);
  record_for_igvn(slow_region);
  if (!stopped()) {
    // It's an instance.  Make the slow-path tests.
    // If this is a virtual call, we generate a funny guard.  We grab
    // the vtable entry corresponding to clone() from the target object.
    // If the target method which we are calling happens to be the
    // Object clone() method, we pass the guard.  We do not need this
    // guard for non-virtual calls; the caller is known to be the native
    // Object clone().
    if (is_virtual) {
      generate_virtual_guard(obj_klass, slow_region);
    }

    // The object must be cloneable and must not have a finalizer.
    // Both of these conditions may be checked in a single test.
    // We could optimize the cloneable test further, but we don't care.
    generate_access_flags_guard(obj_klass,
                                // Test both conditions:
                                JVM_ACC_IS_CLONEABLE | JVM_ACC_HAS_FINALIZER,
                                // Must be cloneable but not finalizer:
                                JVM_ACC_IS_CLONEABLE,
                                slow_region);
  }

  if (!stopped()) {
    // It's an instance, and it passed the slow-path tests.
    PreserveJVMState pjvms(this);
    Node* obj_size = NULL;
    Node* alloc_obj = new_instance(obj_klass, NULL, raw_mem_only, &obj_size);
    assert(obj_size != NULL, "");
    Node* raw_obj = alloc_obj->in(1);
    assert(raw_obj->is_Proj() && raw_obj->in(0)->is_Allocate(), "");
    if (ReduceBulkZeroing) {
      AllocateNode* alloc = AllocateNode::Ideal_allocation(alloc_obj, &_gvn);
      if (alloc != NULL && !alloc->maybe_set_complete(&_gvn))
        alloc = NULL;
    }
    if (!use_ReduceInitialCardMarks()) {
      // Put in store barrier for any and all oops we are sticking
      // into this object.  (We could avoid this if we could prove
      // that the object type contains no oop fields at all.)
      card_mark = true;
    }
    alloc_val->init_req(_instance_alloc, raw_obj);
    alloc_siz->init_req(_instance_alloc, obj_size);
    alloc_reg->init_req(_instance_alloc, control());
    alloc_i_o->init_req(_instance_alloc, i_o());
    alloc_mem->init_req(_instance_alloc, memory(raw_adr_type));
  }

  // Generate code for the slow case.  We make a call to clone().
  set_control(_gvn.transform(slow_region));
  if (!stopped()) {
    PreserveJVMState pjvms(this);
    CallJavaNode* slow_call = generate_method_call(vmIntrinsics::_clone, is_virtual);
    Node* slow_result = set_results_for_java_call(slow_call);
    // this->control() comes from set_results_for_java_call
    result_reg->init_req(_slow_path, control());
    result_val->init_req(_slow_path, slow_result);
    result_i_o ->set_req(_slow_path, i_o());
    result_mem ->set_req(_slow_path, reset_memory());
  }

  // The object is allocated, as an array and/or an instance.  Now copy it.
  set_control( _gvn.transform(alloc_reg) );
  set_i_o(     _gvn.transform(alloc_i_o) );
  set_memory(  _gvn.transform(alloc_mem), raw_adr_type );
  Node* raw_obj  = _gvn.transform(alloc_val);

  if (!stopped()) {
    // Copy the fastest available way.
    // (No need for PreserveJVMState, since we're using it all up now.)
    Node* src  = obj;
    Node* dest = raw_obj;
    Node* end  = dest;
    Node* size = _gvn.transform(alloc_siz);

    // Exclude the header.
    int base_off = sizeof(oopDesc);
    src  = basic_plus_adr(src,  base_off);
    dest = basic_plus_adr(dest, base_off);
    end  = basic_plus_adr(end,  size);

    // Compute the length also, if needed:
    Node* countx = size;
    countx = _gvn.transform( new (C, 3) SubXNode(countx, MakeConX(base_off)) );
    countx = _gvn.transform( new (C, 3) URShiftXNode(countx, intcon(LogBytesPerLong) ));

    // Select an appropriate instruction to initialize the range.
    // The CopyArray instruction (if supported) can be optimized
    // into a discrete set of scalar loads and stores.
    bool disjoint_bases = true;
    generate_unchecked_arraycopy(raw_adr_type, T_LONG, disjoint_bases,
                                 src, NULL, dest, NULL, countx);

    // Now that the object is properly initialized, type it as an oop.
    // Use a secondary InitializeNode memory barrier.
    InitializeNode* init = insert_mem_bar_volatile(Op_Initialize, raw_adr_idx,
                                                   raw_obj)->as_Initialize();
    init->set_complete(&_gvn);  // (there is no corresponding AllocateNode)
    Node* new_obj = new(C, 2) CheckCastPPNode(control(), raw_obj,
                                              TypeInstPtr::NOTNULL);
    new_obj = _gvn.transform(new_obj);

    // If necessary, emit some card marks afterwards.  (Non-arrays only.)
    if (card_mark) {
      Node* no_particular_value = NULL;
      Node* no_particular_field = NULL;
      post_barrier(control(),
                   memory(raw_adr_type),
                   new_obj,
                   no_particular_field,
                   raw_adr_idx,
                   no_particular_value,
                   T_OBJECT,
                   false);
    }
    // Present the results of the slow call.
    result_reg->init_req(_fast_path, control());
    result_val->init_req(_fast_path, new_obj);
    result_i_o ->set_req(_fast_path, i_o());
    result_mem ->set_req(_fast_path, reset_memory());
  }

  // Return the combined state.
  set_control(    _gvn.transform(result_reg) );
  set_i_o(        _gvn.transform(result_i_o) );
  set_all_memory( _gvn.transform(result_mem) );

  // Cast the result to a sharper type, since we know what clone does.
  Node* new_obj = _gvn.transform(result_val);
  Node* cast    = new (C, 2) CheckCastPPNode(control(), new_obj, toop);
  push(_gvn.transform(cast));

  return true;
}


// constants for computing the copy function
enum {
  COPYFUNC_UNALIGNED = 0,
  COPYFUNC_ALIGNED = 1,                 // src, dest aligned to HeapWordSize
  COPYFUNC_CONJOINT = 0,
  COPYFUNC_DISJOINT = 2                 // src != dest, or transfer can descend
};

// Note:  The condition "disjoint" applies also for overlapping copies
// where an descending copy is permitted (i.e., dest_offset <= src_offset).
static address
select_arraycopy_function(BasicType t, bool aligned, bool disjoint, const char* &name) {
  int selector =
    (aligned  ? COPYFUNC_ALIGNED  : COPYFUNC_UNALIGNED) +
    (disjoint ? COPYFUNC_DISJOINT : COPYFUNC_CONJOINT);

#define RETURN_STUB(xxx_arraycopy) { \
  name = #xxx_arraycopy; \
  return StubRoutines::xxx_arraycopy(); }

  switch (t) {
  case T_BYTE:
  case T_BOOLEAN:
    switch (selector) {
    case COPYFUNC_CONJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jbyte_arraycopy);
    case COPYFUNC_CONJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jbyte_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jbyte_disjoint_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jbyte_disjoint_arraycopy);
    }
  case T_CHAR:
  case T_SHORT:
    switch (selector) {
    case COPYFUNC_CONJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jshort_arraycopy);
    case COPYFUNC_CONJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jshort_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jshort_disjoint_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jshort_disjoint_arraycopy);
    }
  case T_INT:
  case T_FLOAT:
    switch (selector) {
    case COPYFUNC_CONJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jint_arraycopy);
    case COPYFUNC_CONJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jint_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jint_disjoint_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jint_disjoint_arraycopy);
    }
  case T_DOUBLE:
  case T_LONG:
    switch (selector) {
    case COPYFUNC_CONJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jlong_arraycopy);
    case COPYFUNC_CONJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jlong_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(jlong_disjoint_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_jlong_disjoint_arraycopy);
    }
  case T_ARRAY:
  case T_OBJECT:
    switch (selector) {
    case COPYFUNC_CONJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(oop_arraycopy);
    case COPYFUNC_CONJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_oop_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_UNALIGNED:  RETURN_STUB(oop_disjoint_arraycopy);
    case COPYFUNC_DISJOINT | COPYFUNC_ALIGNED:    RETURN_STUB(arrayof_oop_disjoint_arraycopy);
    }
  default:
    ShouldNotReachHere();
    return NULL;
  }

#undef RETURN_STUB
}

//------------------------------basictype2arraycopy----------------------------
address LibraryCallKit::basictype2arraycopy(BasicType t,
                                            Node* src_offset,
                                            Node* dest_offset,
                                            bool disjoint_bases,
                                            const char* &name) {
  const TypeInt* src_offset_inttype  = gvn().find_int_type(src_offset);;
  const TypeInt* dest_offset_inttype = gvn().find_int_type(dest_offset);;

  bool aligned = false;
  bool disjoint = disjoint_bases;

  // if the offsets are the same, we can treat the memory regions as
  // disjoint, because either the memory regions are in different arrays,
  // or they are identical (which we can treat as disjoint.)  We can also
  // treat a copy with a destination index  less that the source index
  // as disjoint since a low->high copy will work correctly in this case.
  if (src_offset_inttype != NULL && src_offset_inttype->is_con() &&
      dest_offset_inttype != NULL && dest_offset_inttype->is_con()) {
    // both indices are constants
    int s_offs = src_offset_inttype->get_con();
    int d_offs = dest_offset_inttype->get_con();
    int element_size = type2aelembytes[t];
    aligned = ((arrayOopDesc::base_offset_in_bytes(t) + s_offs * element_size) % HeapWordSize == 0) &&
              ((arrayOopDesc::base_offset_in_bytes(t) + d_offs * element_size) % HeapWordSize == 0);
    if (s_offs >= d_offs)  disjoint = true;
  } else if (src_offset == dest_offset && src_offset != NULL) {
    // This can occur if the offsets are identical non-constants.
    disjoint = true;
  }

  return select_arraycopy_function(t, aligned, disjoint, name);
}


//------------------------------inline_arraycopy-----------------------
bool LibraryCallKit::inline_arraycopy() {
  // Restore the stack and pop off the arguments.
  int nargs = 5;  // 2 oops, 3 ints, no size_t or long
  assert(callee()->signature()->size() == nargs, "copy has 5 arguments");

  Node *src         = argument(0);
  Node *src_offset  = argument(1);
  Node *dest        = argument(2);
  Node *dest_offset = argument(3);
  Node *length      = argument(4);

  // Compile time checks.  If any of these checks cannot be verified at compile time,
  // we do not make a fast path for this call.  Instead, we let the call remain as it
  // is.  The checks we choose to mandate at compile time are:
  //
  // (1) src and dest are arrays.
  const Type* src_type = src->Value(&_gvn);
  const Type* dest_type = dest->Value(&_gvn);
  const TypeAryPtr* top_src = src_type->isa_aryptr();
  const TypeAryPtr* top_dest = dest_type->isa_aryptr();
  if (top_src  == NULL || top_src->klass()  == NULL ||
      top_dest == NULL || top_dest->klass() == NULL) {
    // Conservatively insert a memory barrier on all memory slices.
    // Do not let writes into the source float below the arraycopy.
    insert_mem_bar(Op_MemBarCPUOrder);

    // Call StubRoutines::generic_arraycopy stub.
    generate_arraycopy(TypeRawPtr::BOTTOM, T_CONFLICT,
                       src, src_offset, dest, dest_offset, length,
                       nargs);

    // Do not let reads from the destination float above the arraycopy.
    // Since we cannot type the arrays, we don't know which slices
    // might be affected.  We could restrict this barrier only to those
    // memory slices which pertain to array elements--but don't bother.
    if (!InsertMemBarAfterArraycopy)
      // (If InsertMemBarAfterArraycopy, there is already one in place.)
      insert_mem_bar(Op_MemBarCPUOrder);
    return true;
  }

  // (2) src and dest arrays must have elements of the same BasicType
  // Figure out the size and type of the elements we will be copying.
  BasicType src_elem  =  top_src->klass()->as_array_klass()->element_type()->basic_type();
  BasicType dest_elem = top_dest->klass()->as_array_klass()->element_type()->basic_type();
  if (src_elem  == T_ARRAY)  src_elem  = T_OBJECT;
  if (dest_elem == T_ARRAY)  dest_elem = T_OBJECT;

  if (src_elem != dest_elem || dest_elem == T_VOID) {
    // The component types are not the same or are not recognized.  Punt.
    // (But, avoid the native method wrapper to JVM_ArrayCopy.)
    generate_slow_arraycopy(TypePtr::BOTTOM,
                            src, src_offset, dest, dest_offset, length,
                            nargs);
    return true;
  }

  //---------------------------------------------------------------------------
  // We will make a fast path for this call to arraycopy.

  // We have the following tests left to perform:
  //
  // (3) src and dest must not be null.
  // (4) src_offset must not be negative.
  // (5) dest_offset must not be negative.
  // (6) length must not be negative.
  // (7) src_offset + length must not exceed length of src.
  // (8) dest_offset + length must not exceed length of dest.
  // (9) each element of an oop array must be assignable

  RegionNode* slow_region = new (C, 1) RegionNode(1);
  record_for_igvn(slow_region);

  // (3) operands must not be null
  // We currently perform our null checks with the do_null_check routine.
  // This means that the null exceptions will be reported in the caller
  // rather than (correctly) reported inside of the native arraycopy call.
  // This should be corrected, given time.  We do our null check with the
  // stack pointer restored.
  _sp += nargs;
  src  = do_null_check(src,  T_ARRAY);
  dest = do_null_check(dest, T_ARRAY);
  _sp -= nargs;

  // (4) src_offset must not be negative.
  generate_negative_guard(src_offset, slow_region);

  // (5) dest_offset must not be negative.
  generate_negative_guard(dest_offset, slow_region);

  // (6) length must not be negative (moved to generate_arraycopy()).
  // generate_negative_guard(length, slow_region);

  // (7) src_offset + length must not exceed length of src.
  generate_limit_guard(src_offset, length,
                       load_array_length(src),
                       slow_region);

  // (8) dest_offset + length must not exceed length of dest.
  generate_limit_guard(dest_offset, length,
                       load_array_length(dest),
                       slow_region);

  // (9) each element of an oop array must be assignable
  // The generate_arraycopy subroutine checks this.

  // This is where the memory effects are placed:
  const TypePtr* adr_type = TypeAryPtr::get_array_body_type(dest_elem);
  generate_arraycopy(adr_type, dest_elem,
                     src, src_offset, dest, dest_offset, length,
                     nargs, false, false, slow_region);

  return true;
}

//-----------------------------generate_arraycopy----------------------
// Generate an optimized call to arraycopy.
// Caller must guard against non-arrays.
// Caller must determine a common array basic-type for both arrays.
// Caller must validate offsets against array bounds.
// The slow_region has already collected guard failure paths
// (such as out of bounds length or non-conformable array types).
// The generated code has this shape, in general:
//
//     if (length == 0)  return   // via zero_path
//     slowval = -1
//     if (types unknown) {
//       slowval = call generic copy loop
//       if (slowval == 0)  return  // via checked_path
//     } else if (indexes in bounds) {
//       if ((is object array) && !(array type check)) {
//         slowval = call checked copy loop
//         if (slowval == 0)  return  // via checked_path
//       } else {
//         call bulk copy loop
//         return  // via fast_path
//       }
//     }
//     // adjust params for remaining work:
//     if (slowval != -1) {
//       n = -1^slowval; src_offset += n; dest_offset += n; length -= n
//     }
//   slow_region:
//     call slow arraycopy(src, src_offset, dest, dest_offset, length)
//     return  // via slow_call_path
//
// This routine is used from several intrinsics:  System.arraycopy,
// Object.clone (the array subcase), and Arrays.copyOf[Range].
//
void
LibraryCallKit::generate_arraycopy(const TypePtr* adr_type,
                                   BasicType basic_elem_type,
                                   Node* src,  Node* src_offset,
                                   Node* dest, Node* dest_offset,
                                   Node* copy_length,
                                   int nargs,
                                   bool disjoint_bases,
                                   bool length_never_negative,
                                   RegionNode* slow_region) {

  if (slow_region == NULL) {
    slow_region = new(C,1) RegionNode(1);
    record_for_igvn(slow_region);
  }

  Node* original_dest      = dest;
  AllocateArrayNode* alloc = NULL;  // used for zeroing, if needed
  Node* raw_dest           = NULL;  // used before zeroing, if needed
  bool  must_clear_dest    = false;

  // See if this is the initialization of a newly-allocated array.
  // If so, we will take responsibility here for initializing it to zero.
  // (Note:  Because tightly_coupled_allocation performs checks on the
  // out-edges of the dest, we need to avoid making derived pointers
  // from it until we have checked its uses.)
  if (ReduceBulkZeroing
      && !ZeroTLAB              // pointless if already zeroed
      && basic_elem_type != T_CONFLICT // avoid corner case
      && !_gvn.eqv_uncast(src, dest)
      && ((alloc = tightly_coupled_allocation(dest, slow_region))
          != NULL)
      && alloc->maybe_set_complete(&_gvn)) {
    // "You break it, you buy it."
    InitializeNode* init = alloc->initialization();
    assert(init->is_complete(), "we just did this");
    assert(dest->Opcode() == Op_CheckCastPP, "sanity");
    assert(dest->in(0)->in(0) == init, "dest pinned");
    raw_dest = dest->in(1);  // grab the raw pointer!
    original_dest = dest;
    dest = raw_dest;
    adr_type = TypeRawPtr::BOTTOM;  // all initializations are into raw memory
    // Decouple the original InitializeNode, turning it into a simple membar.
    // We will build a new one at the end of this routine.
    init->set_req(InitializeNode::RawAddress, top());
    // From this point on, every exit path is responsible for
    // initializing any non-copied parts of the object to zero.
    must_clear_dest = true;
  } else {
    // No zeroing elimination here.
    alloc             = NULL;
    //original_dest   = dest;
    //must_clear_dest = false;
  }

  // Results are placed here:
  enum { fast_path        = 1,  // normal void-returning assembly stub
         checked_path     = 2,  // special assembly stub with cleanup
         slow_call_path   = 3,  // something went wrong; call the VM
         zero_path        = 4,  // bypass when length of copy is zero
         bcopy_path       = 5,  // copy primitive array by 64-bit blocks
         PATH_LIMIT       = 6
  };
  RegionNode* result_region = new(C, PATH_LIMIT) RegionNode(PATH_LIMIT);
  PhiNode*    result_i_o    = new(C, PATH_LIMIT) PhiNode(result_region, Type::ABIO);
  PhiNode*    result_memory = new(C, PATH_LIMIT) PhiNode(result_region, Type::MEMORY, adr_type);
  record_for_igvn(result_region);
  _gvn.set_type_bottom(result_i_o);
  _gvn.set_type_bottom(result_memory);
  assert(adr_type != TypePtr::BOTTOM, "must be RawMem or a T[] slice");

  // The slow_control path:
  Node* slow_control;
  Node* slow_i_o = i_o();
  Node* slow_mem = memory(adr_type);
  debug_only(slow_control = (Node*) badAddress);

  // Checked control path:
  Node* checked_control = top();
  Node* checked_mem     = NULL;
  Node* checked_i_o     = NULL;
  Node* checked_value   = NULL;

  if (basic_elem_type == T_CONFLICT) {
    assert(!must_clear_dest, "");
    Node* cv = generate_generic_arraycopy(adr_type,
                                          src, src_offset, dest, dest_offset,
                                          copy_length, nargs);
    if (cv == NULL)  cv = intcon(-1);  // failure (no stub available)
    checked_control = control();
    checked_i_o     = i_o();
    checked_mem     = memory(adr_type);
    checked_value   = cv;
    set_control(top());         // no fast path
  }

  Node* not_pos = generate_nonpositive_guard(copy_length, length_never_negative);
  if (not_pos != NULL) {
    PreserveJVMState pjvms(this);
    set_control(not_pos);

    // (6) length must not be negative.
    if (!length_never_negative) {
      generate_negative_guard(copy_length, slow_region);
    }

    if (!stopped() && must_clear_dest) {
      Node* dest_length = alloc->in(AllocateNode::ALength);
      if (_gvn.eqv_uncast(copy_length, dest_length)
          || _gvn.find_int_con(dest_length, 1) <= 0) {
        // There is no zeroing to do.
      } else {
        // Clear the whole thing since there are no source elements to copy.
        generate_clear_array(adr_type, dest, basic_elem_type,
                             intcon(0), NULL,
                             alloc->in(AllocateNode::AllocSize));
      }
    }

    // Present the results of the fast call.
    result_region->init_req(zero_path, control());
    result_i_o   ->init_req(zero_path, i_o());
    result_memory->init_req(zero_path, memory(adr_type));
  }

  if (!stopped() && must_clear_dest) {
    // We have to initialize the *uncopied* part of the array to zero.
    // The copy destination is the slice dest[off..off+len].  The other slices
    // are dest_head = dest[0..off] and dest_tail = dest[off+len..dest.length].
    Node* dest_size   = alloc->in(AllocateNode::AllocSize);
    Node* dest_length = alloc->in(AllocateNode::ALength);
    Node* dest_tail   = _gvn.transform( new(C,3) AddINode(dest_offset,
                                                          copy_length) );

    // If there is a head section that needs zeroing, do it now.
    if (find_int_con(dest_offset, -1) != 0) {
      generate_clear_array(adr_type, dest, basic_elem_type,
                           intcon(0), dest_offset,
                           NULL);
    }

    // Next, perform a dynamic check on the tail length.
    // It is often zero, and we can win big if we prove this.
    // There are two wins:  Avoid generating the ClearArray
    // with its attendant messy index arithmetic, and upgrade
    // the copy to a more hardware-friendly word size of 64 bits.
    Node* tail_ctl = NULL;
    if (!stopped() && !_gvn.eqv_uncast(dest_tail, dest_length)) {
      Node* cmp_lt   = _gvn.transform( new(C,3) CmpINode(dest_tail, dest_length) );
      Node* bol_lt   = _gvn.transform( new(C,2) BoolNode(cmp_lt, BoolTest::lt) );
      tail_ctl = generate_slow_guard(bol_lt, NULL);
      assert(tail_ctl != NULL || !stopped(), "must be an outcome");
    }

    // At this point, let's assume there is no tail.
    if (!stopped() && alloc != NULL && basic_elem_type != T_OBJECT) {
      // There is no tail.  Try an upgrade to a 64-bit copy.
      bool didit = false;
      { PreserveJVMState pjvms(this);
        didit = generate_block_arraycopy(adr_type, basic_elem_type, alloc,
                                         src, src_offset, dest, dest_offset,
                                         dest_size);
        if (didit) {
          // Present the results of the block-copying fast call.
          result_region->init_req(bcopy_path, control());
          result_i_o   ->init_req(bcopy_path, i_o());
          result_memory->init_req(bcopy_path, memory(adr_type));
        }
      }
      if (didit)
        set_control(top());     // no regular fast path
    }

    // Clear the tail, if any.
    if (tail_ctl != NULL) {
      Node* notail_ctl = stopped() ? NULL : control();
      set_control(tail_ctl);
      if (notail_ctl == NULL) {
        generate_clear_array(adr_type, dest, basic_elem_type,
                             dest_tail, NULL,
                             dest_size);
      } else {
        // Make a local merge.
        Node* done_ctl = new(C,3) RegionNode(3);
        Node* done_mem = new(C,3) PhiNode(done_ctl, Type::MEMORY, adr_type);
        done_ctl->init_req(1, notail_ctl);
        done_mem->init_req(1, memory(adr_type));
        generate_clear_array(adr_type, dest, basic_elem_type,
                             dest_tail, NULL,
                             dest_size);
        done_ctl->init_req(2, control());
        done_mem->init_req(2, memory(adr_type));
        set_control( _gvn.transform(done_ctl) );
        set_memory(  _gvn.transform(done_mem), adr_type );
      }
    }
  }

  BasicType copy_type = basic_elem_type;
  assert(basic_elem_type != T_ARRAY, "caller must fix this");
  if (!stopped() && copy_type == T_OBJECT) {
    // If src and dest have compatible element types, we can copy bits.
    // Types S[] and D[] are compatible if D is a supertype of S.
    //
    // If they are not, we will use checked_oop_disjoint_arraycopy,
    // which performs a fast optimistic per-oop check, and backs off
    // further to JVM_ArrayCopy on the first per-oop check that fails.
    // (Actually, we don't move raw bits only; the GC requires card marks.)

    // Get the klassOop for both src and dest
    Node* src_klass  = load_object_klass(src);
    Node* dest_klass = load_object_klass(dest);

    // Generate the subtype check.
    // This might fold up statically, or then again it might not.
    //
    // Non-static example:  Copying List<String>.elements to a new String[].
    // The backing store for a List<String> is always an Object[],
    // but its elements are always type String, if the generic types
    // are correct at the source level.
    //
    // Test S[] against D[], not S against D, because (probably)
    // the secondary supertype cache is less busy for S[] than S.
    // This usually only matters when D is an interface.
    Node* not_subtype_ctrl = gen_subtype_check(src_klass, dest_klass);
    // Plug failing path into checked_oop_disjoint_arraycopy
    if (not_subtype_ctrl != top()) {
      PreserveJVMState pjvms(this);
      set_control(not_subtype_ctrl);
      // (At this point we can assume disjoint_bases, since types differ.)
      int ek_offset = objArrayKlass::element_klass_offset_in_bytes() + sizeof(oopDesc);
      Node* p1 = basic_plus_adr(dest_klass, ek_offset);
      Node* n1 = new (C, 3) LoadKlassNode(0, immutable_memory(), p1, TypeRawPtr::BOTTOM);
      Node* dest_elem_klass = _gvn.transform(n1);
      Node* cv = generate_checkcast_arraycopy(adr_type,
                                              dest_elem_klass,
                                              src, src_offset, dest, dest_offset,
                                              copy_length,
                                              nargs);
      if (cv == NULL)  cv = intcon(-1);  // failure (no stub available)
      checked_control = control();
      checked_i_o     = i_o();
      checked_mem     = memory(adr_type);
      checked_value   = cv;
    }
    // At this point we know we do not need type checks on oop stores.

    // Let's see if we need card marks:
    if (alloc != NULL && use_ReduceInitialCardMarks()) {
      // If we do not need card marks, copy using the jint or jlong stub.
      copy_type = LP64_ONLY(T_LONG) NOT_LP64(T_INT);
      assert(type2aelembytes[basic_elem_type] == type2aelembytes[copy_type],
             "sizes agree");
    }
  }

  if (!stopped()) {
    // Generate the fast path, if possible.
    PreserveJVMState pjvms(this);
    generate_unchecked_arraycopy(adr_type, copy_type, disjoint_bases,
                                 src, src_offset, dest, dest_offset,
                                 ConvI2X(copy_length));

    // Present the results of the fast call.
    result_region->init_req(fast_path, control());
    result_i_o   ->init_req(fast_path, i_o());
    result_memory->init_req(fast_path, memory(adr_type));
  }

  // Here are all the slow paths up to this point, in one bundle:
  slow_control = top();
  if (slow_region != NULL)
    slow_control = _gvn.transform(slow_region);
  debug_only(slow_region = (RegionNode*)badAddress);

  set_control(checked_control);
  if (!stopped()) {
    // Clean up after the checked call.
    // The returned value is either 0 or -1^K,
    // where K = number of partially transferred array elements.
    Node* cmp = _gvn.transform( new(C, 3) CmpINode(checked_value, intcon(0)) );
    Node* bol = _gvn.transform( new(C, 2) BoolNode(cmp, BoolTest::eq) );
    IfNode* iff = create_and_map_if(control(), bol, PROB_MAX, COUNT_UNKNOWN);

    // If it is 0, we are done, so transfer to the end.
    Node* checks_done = _gvn.transform( new(C, 1) IfTrueNode(iff) );
    result_region->init_req(checked_path, checks_done);
    result_i_o   ->init_req(checked_path, checked_i_o);
    result_memory->init_req(checked_path, checked_mem);

    // If it is not zero, merge into the slow call.
    set_control( _gvn.transform( new(C, 1) IfFalseNode(iff) ));
    RegionNode* slow_reg2 = new(C, 3) RegionNode(3);
    PhiNode*    slow_i_o2 = new(C, 3) PhiNode(slow_reg2, Type::ABIO);
    PhiNode*    slow_mem2 = new(C, 3) PhiNode(slow_reg2, Type::MEMORY, adr_type);
    record_for_igvn(slow_reg2);
    slow_reg2  ->init_req(1, slow_control);
    slow_i_o2  ->init_req(1, slow_i_o);
    slow_mem2  ->init_req(1, slow_mem);
    slow_reg2  ->init_req(2, control());
    slow_i_o2  ->init_req(2, i_o());
    slow_mem2  ->init_req(2, memory(adr_type));

    slow_control = _gvn.transform(slow_reg2);
    slow_i_o     = _gvn.transform(slow_i_o2);
    slow_mem     = _gvn.transform(slow_mem2);

    if (alloc != NULL) {
      // We'll restart from the very beginning, after zeroing the whole thing.
      // This can cause double writes, but that's OK since dest is brand new.
      // So we ignore the low 31 bits of the value returned from the stub.
    } else {
      // We must continue the copy exactly where it failed, or else
      // another thread might see the wrong number of writes to dest.
      Node* checked_offset = _gvn.transform( new(C, 3) XorINode(checked_value, intcon(-1)) );
      Node* slow_offset    = new(C, 3) PhiNode(slow_reg2, TypeInt::INT);
      slow_offset->init_req(1, intcon(0));
      slow_offset->init_req(2, checked_offset);
      slow_offset  = _gvn.transform(slow_offset);

      // Adjust the arguments by the conditionally incoming offset.
      Node* src_off_plus  = _gvn.transform( new(C, 3) AddINode(src_offset,  slow_offset) );
      Node* dest_off_plus = _gvn.transform( new(C, 3) AddINode(dest_offset, slow_offset) );
      Node* length_minus  = _gvn.transform( new(C, 3) SubINode(copy_length, slow_offset) );

      // Tweak the node variables to adjust the code produced below:
      src_offset  = src_off_plus;
      dest_offset = dest_off_plus;
      copy_length = length_minus;
    }
  }

  set_control(slow_control);
  if (!stopped()) {
    // Generate the slow path, if needed.
    PreserveJVMState pjvms(this);   // replace_in_map may trash the map

    set_memory(slow_mem, adr_type);
    set_i_o(slow_i_o);

    if (must_clear_dest) {
      generate_clear_array(adr_type, dest, basic_elem_type,
                           intcon(0), NULL,
                           alloc->in(AllocateNode::AllocSize));
    }

    if (dest != original_dest) {
      // Promote from rawptr to oop, so it looks right in the call's GC map.
      dest = _gvn.transform( new(C,2) CheckCastPPNode(control(), dest,
                                                      TypeInstPtr::NOTNULL) );

      // Edit the call's debug-info to avoid referring to original_dest.
      // (The problem with original_dest is that it isn't ready until
      // after the InitializeNode completes, but this stuff is before.)
      // Substitute in the locally valid dest_oop.
      replace_in_map(original_dest, dest);
    }

    generate_slow_arraycopy(adr_type,
                            src, src_offset, dest, dest_offset,
                            copy_length, nargs);

    result_region->init_req(slow_call_path, control());
    result_i_o   ->init_req(slow_call_path, i_o());
    result_memory->init_req(slow_call_path, memory(adr_type));
  }

  // Remove unused edges.
  for (uint i = 1; i < result_region->req(); i++) {
    if (result_region->in(i) == NULL)
      result_region->init_req(i, top());
  }

  // Finished; return the combined state.
  set_control( _gvn.transform(result_region) );
  set_i_o(     _gvn.transform(result_i_o)    );
  set_memory(  _gvn.transform(result_memory), adr_type );

  if (dest != original_dest) {
    // Pin the "finished" array node after the arraycopy/zeroing operations.
    // Use a secondary InitializeNode memory barrier.
    InitializeNode* init = insert_mem_bar_volatile(Op_Initialize,
                                                   Compile::AliasIdxRaw,
                                                   raw_dest)->as_Initialize();
    init->set_complete(&_gvn);  // (there is no corresponding AllocateNode)
    _gvn.hash_delete(original_dest);
    original_dest->set_req(0, control());
    _gvn.hash_find_insert(original_dest);  // put back into GVN table
  }

  // The memory edges above are precise in order to model effects around
  // array copyies accurately to allow value numbering of field loads around
  // arraycopy.  Such field loads, both before and after, are common in Java
  // collections and similar classes involving header/array data structures.
  //
  // But with low number of register or when some registers are used or killed
  // by arraycopy calls it causes registers spilling on stack. See 6544710.
  // The next memory barrier is added to avoid it. If the arraycopy can be
  // optimized away (which it can, sometimes) then we can manually remove
  // the membar also.
  if (InsertMemBarAfterArraycopy)
    insert_mem_bar(Op_MemBarCPUOrder);
}


// Helper function which determines if an arraycopy immediately follows
// an allocation, with no intervening tests or other escapes for the object.
AllocateArrayNode*
LibraryCallKit::tightly_coupled_allocation(Node* ptr,
                                           RegionNode* slow_region) {
  if (stopped())             return NULL;  // no fast path
  if (C->AliasLevel() == 0)  return NULL;  // no MergeMems around

  AllocateArrayNode* alloc = AllocateArrayNode::Ideal_array_allocation(ptr, &_gvn);
  if (alloc == NULL)  return NULL;

  Node* rawmem = memory(Compile::AliasIdxRaw);
  // Is the allocation's memory state untouched?
  if (!(rawmem->is_Proj() && rawmem->in(0)->is_Initialize())) {
    // Bail out if there have been raw-memory effects since the allocation.
    // (Example:  There might have been a call or safepoint.)
    return NULL;
  }
  rawmem = rawmem->in(0)->as_Initialize()->memory(Compile::AliasIdxRaw);
  if (!(rawmem->is_Proj() && rawmem->in(0) == alloc)) {
    return NULL;
  }

  // There must be no unexpected observers of this allocation.
  for (DUIterator_Fast imax, i = ptr->fast_outs(imax); i < imax; i++) {
    Node* obs = ptr->fast_out(i);
    if (obs != this->map()) {
      return NULL;
    }
  }

  // This arraycopy must unconditionally follow the allocation of the ptr.
  Node* alloc_ctl = ptr->in(0);
  assert(just_allocated_object(alloc_ctl) == ptr, "most recent allo");

  Node* ctl = control();
  while (ctl != alloc_ctl) {
    // There may be guards which feed into the slow_region.
    // Any other control flow means that we might not get a chance
    // to finish initializing the allocated object.
    if ((ctl->is_IfFalse() || ctl->is_IfTrue()) && ctl->in(0)->is_If()) {
      IfNode* iff = ctl->in(0)->as_If();
      Node* not_ctl = iff->proj_out(1 - ctl->as_Proj()->_con);
      assert(not_ctl != NULL && not_ctl != ctl, "found alternate");
      if (slow_region != NULL && slow_region->find_edge(not_ctl) >= 1) {
        ctl = iff->in(0);       // This test feeds the known slow_region.
        continue;
      }
      // One more try:  Various low-level checks bottom out in
      // uncommon traps.  If the debug-info of the trap omits
      // any reference to the allocation, as we've already
      // observed, then there can be no objection to the trap.
      bool found_trap = false;
      for (DUIterator_Fast jmax, j = not_ctl->fast_outs(jmax); j < jmax; j++) {
        Node* obs = not_ctl->fast_out(j);
        if (obs->in(0) == not_ctl && obs->is_Call() &&
            (obs->as_Call()->entry_point() ==
             SharedRuntime::uncommon_trap_blob()->instructions_begin())) {
          found_trap = true; break;
        }
      }
      if (found_trap) {
        ctl = iff->in(0);       // This test feeds a harmless uncommon trap.
        continue;
      }
    }
    return NULL;
  }

  // If we get this far, we have an allocation which immediately
  // precedes the arraycopy, and we can take over zeroing the new object.
  // The arraycopy will finish the initialization, and provide
  // a new control state to which we will anchor the destination pointer.

  return alloc;
}

// Helper for initialization of arrays, creating a ClearArray.
// It writes zero bits in [start..end), within the body of an array object.
// The memory effects are all chained onto the 'adr_type' alias category.
//
// Since the object is otherwise uninitialized, we are free
// to put a little "slop" around the edges of the cleared area,
// as long as it does not go back into the array's header,
// or beyond the array end within the heap.
//
// The lower edge can be rounded down to the nearest jint and the
// upper edge can be rounded up to the nearest MinObjAlignmentInBytes.
//
// Arguments:
//   adr_type           memory slice where writes are generated
//   dest               oop of the destination array
//   basic_elem_type    element type of the destination
//   slice_idx          array index of first element to store
//   slice_len          number of elements to store (or NULL)
//   dest_size          total size in bytes of the array object
//
// Exactly one of slice_len or dest_size must be non-NULL.
// If dest_size is non-NULL, zeroing extends to the end of the object.
// If slice_len is non-NULL, the slice_idx value must be a constant.
void
LibraryCallKit::generate_clear_array(const TypePtr* adr_type,
                                     Node* dest,
                                     BasicType basic_elem_type,
                                     Node* slice_idx,
                                     Node* slice_len,
                                     Node* dest_size) {
  // one or the other but not both of slice_len and dest_size:
  assert((slice_len != NULL? 1: 0) + (dest_size != NULL? 1: 0) == 1, "");
  if (slice_len == NULL)  slice_len = top();
  if (dest_size == NULL)  dest_size = top();

  // operate on this memory slice:
  Node* mem = memory(adr_type); // memory slice to operate on

  // scaling and rounding of indexes:
  int scale = exact_log2(type2aelembytes[basic_elem_type]);
  int abase = arrayOopDesc::base_offset_in_bytes(basic_elem_type);
  int clear_low = (-1 << scale) & (BytesPerInt  - 1);
  int bump_bit  = (-1 << scale) & BytesPerInt;

  // determine constant starts and ends
  const intptr_t BIG_NEG = -128;
  assert(BIG_NEG + 2*abase < 0, "neg enough");
  intptr_t slice_idx_con = (intptr_t) find_int_con(slice_idx, BIG_NEG);
  intptr_t slice_len_con = (intptr_t) find_int_con(slice_len, BIG_NEG);
  if (slice_len_con == 0) {
    return;                     // nothing to do here
  }
  intptr_t start_con = (abase + (slice_idx_con << scale)) & ~clear_low;
  intptr_t end_con   = find_intptr_t_con(dest_size, -1);
  if (slice_idx_con >= 0 && slice_len_con >= 0) {
    assert(end_con < 0, "not two cons");
    end_con = round_to(abase + ((slice_idx_con + slice_len_con) << scale),
                       BytesPerLong);
  }

  if (start_con >= 0 && end_con >= 0) {
    // Constant start and end.  Simple.
    mem = ClearArrayNode::clear_memory(control(), mem, dest,
                                       start_con, end_con, &_gvn);
  } else if (start_con >= 0 && dest_size != top()) {
    // Constant start, pre-rounded end after the tail of the array.
    Node* end = dest_size;
    mem = ClearArrayNode::clear_memory(control(), mem, dest,
                                       start_con, end, &_gvn);
  } else if (start_con >= 0 && slice_len != top()) {
    // Constant start, non-constant end.  End needs rounding up.
    // End offset = round_up(abase + ((slice_idx_con + slice_len) << scale), 8)
    intptr_t end_base  = abase + (slice_idx_con << scale);
    int      end_round = (-1 << scale) & (BytesPerLong  - 1);
    Node*    end       = ConvI2X(slice_len);
    if (scale != 0)
      end = _gvn.transform( new(C,3) LShiftXNode(end, intcon(scale) ));
    end_base += end_round;
    end = _gvn.transform( new(C,3) AddXNode(end, MakeConX(end_base)) );
    end = _gvn.transform( new(C,3) AndXNode(end, MakeConX(~end_round)) );
    mem = ClearArrayNode::clear_memory(control(), mem, dest,
                                       start_con, end, &_gvn);
  } else if (start_con < 0 && dest_size != top()) {
    // Non-constant start, pre-rounded end after the tail of the array.
    // This is almost certainly a "round-to-end" operation.
    Node* start = slice_idx;
    start = ConvI2X(start);
    if (scale != 0)
      start = _gvn.transform( new(C,3) LShiftXNode( start, intcon(scale) ));
    start = _gvn.transform( new(C,3) AddXNode(start, MakeConX(abase)) );
    if ((bump_bit | clear_low) != 0) {
      int to_clear = (bump_bit | clear_low);
      // Align up mod 8, then store a jint zero unconditionally
      // just before the mod-8 boundary.
      // This would only fail if the first array element were immediately
      // after the length field, and were also at an even offset mod 8.
      assert(((abase + bump_bit) & ~to_clear) - BytesPerInt
             >= arrayOopDesc::length_offset_in_bytes() + BytesPerInt,
             "store must not trash length field");

      // Bump 'start' up to (or past) the next jint boundary:
      start = _gvn.transform( new(C,3) AddXNode(start, MakeConX(bump_bit)) );
      // Round bumped 'start' down to jlong boundary in body of array.
      start = _gvn.transform( new(C,3) AndXNode(start, MakeConX(~to_clear)) );
      // Store a zero to the immediately preceding jint:
      Node* x1 = _gvn.transform( new(C,3) AddXNode(start, MakeConX(-BytesPerInt)) );
      Node* p1 = basic_plus_adr(dest, x1);
      mem = StoreNode::make(C, control(), mem, p1, adr_type, intcon(0), T_INT);
      mem = _gvn.transform(mem);
    }

    Node* end = dest_size; // pre-rounded
    mem = ClearArrayNode::clear_memory(control(), mem, dest,
                                       start, end, &_gvn);
  } else {
    // Non-constant start, unrounded non-constant end.
    // (Nobody zeroes a random midsection of an array using this routine.)
    ShouldNotReachHere();       // fix caller
  }

  // Done.
  set_memory(mem, adr_type);
}


bool
LibraryCallKit::generate_block_arraycopy(const TypePtr* adr_type,
                                         BasicType basic_elem_type,
                                         AllocateNode* alloc,
                                         Node* src,  Node* src_offset,
                                         Node* dest, Node* dest_offset,
                                         Node* dest_size) {
  // See if there is an advantage from block transfer.
  int scale = exact_log2(type2aelembytes[basic_elem_type]);
  if (scale >= LogBytesPerLong)
    return false;               // it is already a block transfer

  // Look at the alignment of the starting offsets.
  int abase = arrayOopDesc::base_offset_in_bytes(basic_elem_type);
  const intptr_t BIG_NEG = -128;
  assert(BIG_NEG + 2*abase < 0, "neg enough");

  intptr_t src_off  = abase + ((intptr_t) find_int_con(src_offset, -1)  << scale);
  intptr_t dest_off = abase + ((intptr_t) find_int_con(dest_offset, -1) << scale);
  if (src_off < 0 || dest_off < 0)
    // At present, we can only understand constants.
    return false;

  if (((src_off | dest_off) & (BytesPerLong-1)) != 0) {
    // Non-aligned; too bad.
    // One more chance:  Pick off an initial 32-bit word.
    // This is a common case, since abase can be odd mod 8.
    if (((src_off | dest_off) & (BytesPerLong-1)) == BytesPerInt &&
        ((src_off ^ dest_off) & (BytesPerLong-1)) == 0) {
      Node* sptr = basic_plus_adr(src,  src_off);
      Node* dptr = basic_plus_adr(dest, dest_off);
      Node* sval = make_load(control(), sptr, TypeInt::INT, T_INT, adr_type);
      store_to_memory(control(), dptr, sval, T_INT, adr_type);
      src_off += BytesPerInt;
      dest_off += BytesPerInt;
    } else {
      return false;
    }
  }
  assert(src_off % BytesPerLong == 0, "");
  assert(dest_off % BytesPerLong == 0, "");

  // Do this copy by giant steps.
  Node* sptr  = basic_plus_adr(src,  src_off);
  Node* dptr  = basic_plus_adr(dest, dest_off);
  Node* countx = dest_size;
  countx = _gvn.transform( new (C, 3) SubXNode(countx, MakeConX(dest_off)) );
  countx = _gvn.transform( new (C, 3) URShiftXNode(countx, intcon(LogBytesPerLong)) );

  bool disjoint_bases = true;   // since alloc != NULL
  generate_unchecked_arraycopy(adr_type, T_LONG, disjoint_bases,
                               sptr, NULL, dptr, NULL, countx);

  return true;
}


// Helper function; generates code for the slow case.
// We make a call to a runtime method which emulates the native method,
// but without the native wrapper overhead.
void
LibraryCallKit::generate_slow_arraycopy(const TypePtr* adr_type,
                                        Node* src,  Node* src_offset,
                                        Node* dest, Node* dest_offset,
                                        Node* copy_length,
                                        int nargs) {
  _sp += nargs; // any deopt will start just before call to enclosing method
  Node* call = make_runtime_call(RC_NO_LEAF | RC_UNCOMMON,
                                 OptoRuntime::slow_arraycopy_Type(),
                                 OptoRuntime::slow_arraycopy_Java(),
                                 "slow_arraycopy", adr_type,
                                 src, src_offset, dest, dest_offset,
                                 copy_length);
  _sp -= nargs;

  // Handle exceptions thrown by this fellow:
  make_slow_call_ex(call, env()->Throwable_klass(), false);
}

// Helper function; generates code for cases requiring runtime checks.
Node*
LibraryCallKit::generate_checkcast_arraycopy(const TypePtr* adr_type,
                                             Node* dest_elem_klass,
                                             Node* src,  Node* src_offset,
                                             Node* dest, Node* dest_offset,
                                             Node* copy_length,
                                             int nargs) {
  if (stopped())  return NULL;

  address copyfunc_addr = StubRoutines::checkcast_arraycopy();
  if (copyfunc_addr == NULL) { // Stub was not generated, go slow path.
    return NULL;
  }

  // Pick out the parameters required to perform a store-check
  // for the target array.  This is an optimistic check.  It will
  // look in each non-null element's class, at the desired klass's
  // super_check_offset, for the desired klass.
  int sco_offset = Klass::super_check_offset_offset_in_bytes() + sizeof(oopDesc);
  Node* p3 = basic_plus_adr(dest_elem_klass, sco_offset);
  Node* n3 = new(C, 3) LoadINode(NULL, immutable_memory(), p3, TypeRawPtr::BOTTOM);
  Node* check_offset = _gvn.transform(n3);
  Node* check_value  = dest_elem_klass;

  Node* src_start  = array_element_address(src,  src_offset,  T_OBJECT);
  Node* dest_start = array_element_address(dest, dest_offset, T_OBJECT);

  // (We know the arrays are never conjoint, because their types differ.)
  Node* call = make_runtime_call(RC_LEAF|RC_NO_FP,
                                 OptoRuntime::checkcast_arraycopy_Type(),
                                 copyfunc_addr, "checkcast_arraycopy", adr_type,
                                 // five arguments, of which two are
                                 // intptr_t (jlong in LP64)
                                 src_start, dest_start,
                                 copy_length XTOP,
                                 check_offset XTOP,
                                 check_value);

  return _gvn.transform(new (C, 1) ProjNode(call, TypeFunc::Parms));
}


// Helper function; generates code for cases requiring runtime checks.
Node*
LibraryCallKit::generate_generic_arraycopy(const TypePtr* adr_type,
                                           Node* src,  Node* src_offset,
                                           Node* dest, Node* dest_offset,
                                           Node* copy_length,
                                           int nargs) {
  if (stopped())  return NULL;

  address copyfunc_addr = StubRoutines::generic_arraycopy();
  if (copyfunc_addr == NULL) { // Stub was not generated, go slow path.
    return NULL;
  }

  Node* call = make_runtime_call(RC_LEAF|RC_NO_FP,
                    OptoRuntime::generic_arraycopy_Type(),
                    copyfunc_addr, "generic_arraycopy", adr_type,
                    src, src_offset, dest, dest_offset, copy_length);

  return _gvn.transform(new (C, 1) ProjNode(call, TypeFunc::Parms));
}

// Helper function; generates the fast out-of-line call to an arraycopy stub.
void
LibraryCallKit::generate_unchecked_arraycopy(const TypePtr* adr_type,
                                             BasicType basic_elem_type,
                                             bool disjoint_bases,
                                             Node* src,  Node* src_offset,
                                             Node* dest, Node* dest_offset,
                                             Node* copy_length) {
  if (stopped())  return;               // nothing to do

  Node* src_start  = src;
  Node* dest_start = dest;
  if (src_offset != NULL || dest_offset != NULL) {
    assert(src_offset != NULL && dest_offset != NULL, "");
    src_start  = array_element_address(src,  src_offset,  basic_elem_type);
    dest_start = array_element_address(dest, dest_offset, basic_elem_type);
  }

  // Figure out which arraycopy runtime method to call.
  const char* copyfunc_name = "arraycopy";
  address     copyfunc_addr =
      basictype2arraycopy(basic_elem_type, src_offset, dest_offset,
                          disjoint_bases, copyfunc_name);

  // Call it.  Note that the count_ix value is not scaled to a byte-size.
  make_runtime_call(RC_LEAF|RC_NO_FP,
                    OptoRuntime::fast_arraycopy_Type(),
                    copyfunc_addr, copyfunc_name, adr_type,
                    src_start, dest_start, copy_length XTOP);
}