src/hotspot/share/oops/stackChunkOop.cpp

/*
 * Copyright (c) 2021, 2025, Oracle and/or its affiliates. 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 Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
 * or visit www.oracle.com if you need additional information or have any
 * questions.
 *
 */

#include "code/nmethod.hpp"
#include "code/scopeDesc.hpp"
#include "gc/shared/barrierSet.hpp"
#include "gc/shared/barrierSetStackChunk.hpp"
#include "logging/log.hpp"
#include "logging/logStream.hpp"
#include "memory/memRegion.hpp"
#include "oops/instanceStackChunkKlass.inline.hpp"
#include "oops/oop.inline.hpp"
#include "oops/stackChunkOop.inline.hpp"
#include "runtime/frame.hpp"
#include "runtime/registerMap.hpp"
#include "runtime/smallRegisterMap.inline.hpp"
#include "runtime/stackChunkFrameStream.inline.hpp"

// Note: Some functions in this file work with stale object pointers, e.g.
//       DerivedPointerSupport. Be extra careful to not put those pointers into
//       variables of the 'oop' type. There's extra GC verification around oops
//       that may fail when stale oops are being used.

template <typename RegisterMapT>
class FrameOopIterator : public OopIterator {
private:
  const frame& _f;
  const RegisterMapT* _map;

public:
  FrameOopIterator(const frame& f, const RegisterMapT* map)
    : _f(f),
      _map(map) {
  }

  virtual void oops_do(OopClosure* cl) override {
    if (_f.is_interpreted_frame()) {
      _f.oops_interpreted_do(cl, nullptr);
    } else {
      OopMapDo<OopClosure, DerivedOopClosure, IncludeAllValues> visitor(cl, nullptr);
      visitor.oops_do(&_f, _map, _f.oop_map());
    }
  }
};

class LockStackOopIterator : public OopIterator {
private:
  const stackChunkOop _chunk;
public:
  LockStackOopIterator(const stackChunkOop chunk) : _chunk(chunk) {}

  virtual void oops_do(OopClosure* cl) override {
    int cnt = _chunk->lockstack_size();
    oop* lockstack_start = (oop*)_chunk->start_address();
    for (int i = 0; i < cnt; i++) {
      cl->do_oop(&lockstack_start[i]);
    }
  }
};

frame stackChunkOopDesc::top_frame(RegisterMap* map) {
  assert(!is_empty(), "");
  StackChunkFrameStream<ChunkFrames::Mixed> fs(this);

  map->set_stack_chunk(this);
  fs.initialize_register_map(map);

  frame f = fs.to_frame();

  assert(to_offset(f.sp()) == sp(), "f.offset_sp(): %d sp(): %d async: %d", f.offset_sp(), sp(), map->is_async());
  relativize_frame(f);
  f.set_frame_index(0);
  return f;
}

frame stackChunkOopDesc::sender(const frame& f, RegisterMap* map) {
  assert(map->in_cont(), "");
  assert(!map->include_argument_oops(), "");
  assert(!f.is_empty(), "");
  assert(map->stack_chunk() == this, "");
  assert(!is_empty(), "");

  int index = f.frame_index(); // we need to capture the index before calling derelativize, which destroys it
  StackChunkFrameStream<ChunkFrames::Mixed> fs(this, derelativize(f));
  fs.next(map);

  if (!fs.is_done()) {
    frame sender = fs.to_frame();
    assert(is_usable_in_chunk(sender.unextended_sp()), "");
    relativize_frame(sender);

    sender.set_frame_index(index+1);
    return sender;
  }

  if (parent() != nullptr) {
    assert(!parent()->is_empty(), "");
    return parent()->top_frame(map);
  }

  return Continuation::continuation_parent_frame(map);
}

static int num_java_frames(nmethod* nm, address pc) {
  int count = 0;
  for (ScopeDesc* scope = nm->scope_desc_at(pc); scope != nullptr; scope = scope->sender()) {
    count++;
  }
  return count;
}

static int num_java_frames(const StackChunkFrameStream<ChunkFrames::Mixed>& f) {
  assert(f.is_interpreted()
         || (f.cb() != nullptr && f.cb()->is_nmethod() && f.cb()->as_nmethod()->is_java_method()), "");
  return f.is_interpreted() ? 1 : num_java_frames(f.cb()->as_nmethod(), f.orig_pc());
}

int stackChunkOopDesc::num_java_frames() const {
  int n = 0;
  for (StackChunkFrameStream<ChunkFrames::Mixed> f(const_cast<stackChunkOopDesc*>(this)); !f.is_done();
       f.next(SmallRegisterMap::instance())) {
    if (!f.is_stub()) {
      n += ::num_java_frames(f);
    }
  }
  return n;
}

template <stackChunkOopDesc::BarrierType barrier>
class DoBarriersStackClosure {
  const stackChunkOop _chunk;

public:
  DoBarriersStackClosure(stackChunkOop chunk) : _chunk(chunk) {}

  template <ChunkFrames frame_kind, typename RegisterMapT>
  bool do_frame(const StackChunkFrameStream<frame_kind>& f, const RegisterMapT* map) {
    _chunk->do_barriers0<barrier>(f, map);
    return true;
  }
};

template <stackChunkOopDesc::BarrierType barrier>
void stackChunkOopDesc::do_barriers() {
  DoBarriersStackClosure<barrier> closure(this);
  iterate_stack(&closure);
}

template void stackChunkOopDesc::do_barriers<stackChunkOopDesc::BarrierType::Load> ();
template void stackChunkOopDesc::do_barriers<stackChunkOopDesc::BarrierType::Store>();

class DerivedPointersSupport {
public:
  static void relativize(derived_base* base_loc, derived_pointer* derived_loc) {
    // The base oop could be stale from the GC's point-of-view. Treat it as an
    // uintptr_t to stay clear of the oop verification code in oopsHierarcy.hpp.
    uintptr_t base = *(uintptr_t*)base_loc;
    if (base == 0) {
      return;
    }
    assert(!UseCompressedOops || !CompressedOops::is_base((void*)base), "");

    // This is always a full derived pointer
    uintptr_t derived_int_val = *(uintptr_t*)derived_loc;

    // Make the pointer an offset (relativize) and store it at the same location
    uintptr_t offset = derived_int_val - base;
    *(uintptr_t*)derived_loc = offset;
  }

  static void derelativize(derived_base* base_loc, derived_pointer* derived_loc) {
    uintptr_t base = *(uintptr_t*)base_loc;
    if (base == 0) {
      return;
    }
    assert(!UseCompressedOops || !CompressedOops::is_base((void*)base), "");

    // All derived pointers should have been relativized into offsets
    uintptr_t offset = *(uintptr_t*)derived_loc;

    // Restore the original derived pointer
    *(uintptr_t*)derived_loc = base + offset;
  }

  struct RelativizeClosure : public DerivedOopClosure {
    virtual void do_derived_oop(derived_base* base_loc, derived_pointer* derived_loc) override {
      DerivedPointersSupport::relativize(base_loc, derived_loc);
    }
  };

  struct DerelativizeClosure : public DerivedOopClosure {
    virtual void do_derived_oop(derived_base* base_loc, derived_pointer* derived_loc) override {
      DerivedPointersSupport::derelativize(base_loc, derived_loc);
    }
  };
};

template <typename DerivedPointerClosureType>
class EncodeGCModeConcurrentFrameClosure {
  stackChunkOop _chunk;
  DerivedPointerClosureType* _cl;

public:
  EncodeGCModeConcurrentFrameClosure(stackChunkOop chunk, DerivedPointerClosureType* cl)
    : _chunk(chunk),
      _cl(cl) {
  }

  template <ChunkFrames frame_kind, typename RegisterMapT>
  bool do_frame(const StackChunkFrameStream<frame_kind>& f, const RegisterMapT* map) {
    f.iterate_derived_pointers(_cl, map);

    BarrierSetStackChunk* bs_chunk = BarrierSet::barrier_set()->barrier_set_stack_chunk();
    frame fr = f.to_frame();
    FrameOopIterator<RegisterMapT> iterator(fr, map);
    bs_chunk->encode_gc_mode(_chunk, &iterator);

    return true;
  }

  bool do_lockstack() {
    BarrierSetStackChunk* bs_chunk = BarrierSet::barrier_set()->barrier_set_stack_chunk();
    LockStackOopIterator iterator(_chunk);
    bs_chunk->encode_gc_mode(_chunk, &iterator);

    return true;
  }
};

bool stackChunkOopDesc::try_acquire_relativization() {
  for (;;) {
    // We use an acquiring load when reading the flags to ensure that if we leave this
    // function thinking that relativization is finished, we know that if another thread
    // did the relativization, we will still be able to observe the relativized derived
    // pointers, which is important as subsequent modifications of derived pointers must
    // happen after relativization.
    uint8_t flags_before = flags_acquire();
    if ((flags_before & FLAG_GC_MODE) != 0) {
      // Terminal state - relativization is ensured
      return false;
    }

    if ((flags_before & FLAG_CLAIM_RELATIVIZE) != 0) {
      // Someone else has claimed relativization - wait for completion
      MonitorLocker ml(ContinuationRelativize_lock, Mutex::_no_safepoint_check_flag);
      uint8_t flags_under_lock = flags_acquire();
      if ((flags_under_lock & FLAG_GC_MODE) != 0) {
        // Terminal state - relativization is ensured
        return false;
      }

      if ((flags_under_lock & FLAG_NOTIFY_RELATIVIZE) != 0) {
        // Relativization is claimed by another thread, and it knows it needs to notify
        ml.wait();
      } else if (try_set_flags(flags_under_lock, flags_under_lock | FLAG_NOTIFY_RELATIVIZE)) {
        // Relativization is claimed by another thread, and it knows it needs to notify
        ml.wait();
      }
      // Retry - rerun the loop
      continue;
    }

    if (try_set_flags(flags_before, flags_before | FLAG_CLAIM_RELATIVIZE)) {
      // Claimed relativization - let's do it
      return true;
    }
  }
}

void stackChunkOopDesc::release_relativization() {
  for (;;) {
    uint8_t flags_before = flags();
    if ((flags_before & FLAG_NOTIFY_RELATIVIZE) != 0) {
      MonitorLocker ml(ContinuationRelativize_lock, Mutex::_no_safepoint_check_flag);
      // No need to CAS the terminal state; nobody else can be racingly mutating here
      // as both claim and notify flags are already set (and monotonic)
      // We do however need to use a releasing store on the flags, to ensure that
      // the reader of that value (using load_acquire) will be able to observe
      // the relativization of the derived pointers
      uint8_t flags_under_lock = flags();
      release_set_flags(flags_under_lock | FLAG_GC_MODE);
      ml.notify_all();
      return;
    }

    if (try_set_flags(flags_before, flags_before | FLAG_GC_MODE)) {
      // Successfully set the terminal state; we are done
      return;
    }
  }
}

void stackChunkOopDesc::relativize_derived_pointers_concurrently() {
  if (!try_acquire_relativization()) {
    // Already relativized
    return;
  }

  DerivedPointersSupport::RelativizeClosure derived_cl;
  EncodeGCModeConcurrentFrameClosure<decltype(derived_cl)> frame_cl(this, &derived_cl);
  iterate_stack(&frame_cl);
  frame_cl.do_lockstack();

  release_relativization();
}

class TransformStackChunkClosure {
  stackChunkOop _chunk;

public:
  TransformStackChunkClosure(stackChunkOop chunk) : _chunk(chunk) { }

  template <ChunkFrames frame_kind, typename RegisterMapT>
  bool do_frame(const StackChunkFrameStream<frame_kind>& f, const RegisterMapT* map) {
    DerivedPointersSupport::RelativizeClosure derived_cl;
    f.iterate_derived_pointers(&derived_cl, map);

    BarrierSetStackChunk* bs_chunk = BarrierSet::barrier_set()->barrier_set_stack_chunk();
    frame fr = f.to_frame();
    FrameOopIterator<RegisterMapT> iterator(fr, map);
    bs_chunk->encode_gc_mode(_chunk, &iterator);

    return true;
  }

  bool do_lockstack() {
    BarrierSetStackChunk* bs_chunk = BarrierSet::barrier_set()->barrier_set_stack_chunk();
    LockStackOopIterator iterator(_chunk);
    bs_chunk->encode_gc_mode(_chunk, &iterator);

    return true;
  }
};

void stackChunkOopDesc::transform() {
  assert(!is_gc_mode(), "Should only be called once per chunk");
  set_gc_mode(true);

  assert(!has_bitmap(), "Should only be set once");
  set_has_bitmap(true);
  bitmap().clear();

  TransformStackChunkClosure closure(this);
  iterate_stack(&closure);
  closure.do_lockstack();
}

template <stackChunkOopDesc::BarrierType barrier, bool compressedOopsWithBitmap>
class BarrierClosure: public OopClosure {
  NOT_PRODUCT(intptr_t* _sp;)

public:
  BarrierClosure(intptr_t* sp) NOT_PRODUCT(: _sp(sp)) {}

  virtual void do_oop(oop* p)       override { compressedOopsWithBitmap ? do_oop_work((narrowOop*)p) : do_oop_work(p); }
  virtual void do_oop(narrowOop* p) override { do_oop_work(p); }

  template <class T> inline void do_oop_work(T* p) {
    oop value = (oop)HeapAccess<>::oop_load(p);
    if (barrier == stackChunkOopDesc::BarrierType::Store) {
      HeapAccess<>::oop_store(p, value);
    }
  }
};

template <stackChunkOopDesc::BarrierType barrier, ChunkFrames frame_kind, typename RegisterMapT>
void stackChunkOopDesc::do_barriers0(const StackChunkFrameStream<frame_kind>& f, const RegisterMapT* map) {
  // We need to invoke the write barriers so as not to miss oops in old chunks that haven't yet been concurrently scanned
  assert (!f.is_done(), "");

  if (f.is_interpreted()) {
    Method* m = f.to_frame().interpreter_frame_method();
    // Class redefinition support
    m->record_gc_epoch();
  } else if (f.is_compiled()) {
    nmethod* nm = f.cb()->as_nmethod();
    // The entry barrier takes care of having the right synchronization
    // when keeping the nmethod alive during concurrent execution.
    nm->run_nmethod_entry_barrier();
    // There is no need to mark the Method, as class redefinition will walk the
    // CodeCache, noting their Methods
  }

  if (has_bitmap() && UseCompressedOops) {
    BarrierClosure<barrier, true> oops_closure(f.sp());
    f.iterate_oops(&oops_closure, map);
  } else {
    BarrierClosure<barrier, false> oops_closure(f.sp());
    f.iterate_oops(&oops_closure, map);
  }
}

template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Load> (const StackChunkFrameStream<ChunkFrames::Mixed>& f, const RegisterMap* map);
template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Store>(const StackChunkFrameStream<ChunkFrames::Mixed>& f, const RegisterMap* map);
template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Load> (const StackChunkFrameStream<ChunkFrames::CompiledOnly>& f, const RegisterMap* map);
template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Store>(const StackChunkFrameStream<ChunkFrames::CompiledOnly>& f, const RegisterMap* map);
template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Load> (const StackChunkFrameStream<ChunkFrames::Mixed>& f, const SmallRegisterMap* map);
template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Store>(const StackChunkFrameStream<ChunkFrames::Mixed>& f, const SmallRegisterMap* map);
template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Load> (const StackChunkFrameStream<ChunkFrames::CompiledOnly>& f, const SmallRegisterMap* map);
template void stackChunkOopDesc::do_barriers0<stackChunkOopDesc::BarrierType::Store>(const StackChunkFrameStream<ChunkFrames::CompiledOnly>& f, const SmallRegisterMap* map);

template <typename RegisterMapT>
void stackChunkOopDesc::fix_thawed_frame(const frame& f, const RegisterMapT* map) {
  if (!(is_gc_mode() || requires_barriers())) {
    return;
  }

  BarrierSetStackChunk* bs_chunk = BarrierSet::barrier_set()->barrier_set_stack_chunk();
  FrameOopIterator<RegisterMapT> iterator(f, map);
  bs_chunk->decode_gc_mode(this, &iterator);

  if (f.is_compiled_frame() && f.oop_map()->has_derived_oops()) {
    DerivedPointersSupport::DerelativizeClosure derived_closure;
    OopMapDo<OopClosure, DerivedPointersSupport::DerelativizeClosure, SkipNullValue> visitor(nullptr, &derived_closure);
    visitor.oops_do(&f, map, f.oop_map());
  }
}

template void stackChunkOopDesc::fix_thawed_frame(const frame& f, const RegisterMap* map);
template void stackChunkOopDesc::fix_thawed_frame(const frame& f, const SmallRegisterMap* map);

void stackChunkOopDesc::transfer_lockstack(oop* dst, bool requires_barriers) {
  const bool requires_gc_barriers = is_gc_mode() || requires_barriers;
  const bool requires_uncompress = has_bitmap() && UseCompressedOops;
  const auto load_and_clear_obj = [&](intptr_t* at) -> oop {
    if (requires_gc_barriers) {
      if (requires_uncompress) {
        oop value = HeapAccess<>::oop_load(reinterpret_cast<narrowOop*>(at));
        HeapAccess<>::oop_store(reinterpret_cast<narrowOop*>(at), nullptr);
        return value;
      } else {
        oop value = HeapAccess<>::oop_load(reinterpret_cast<oop*>(at));
        HeapAccess<>::oop_store(reinterpret_cast<oop*>(at), nullptr);
        return value;
      }
    } else {
      oop value = *reinterpret_cast<oop*>(at);
      return value;
    }
  };

  const int cnt = lockstack_size();
  intptr_t* lockstack_start = start_address();
  for (int i = 0; i < cnt; i++) {
    oop mon_owner = load_and_clear_obj(&lockstack_start[i]);
    assert(oopDesc::is_oop(mon_owner), "not an oop");
    dst[i] = mon_owner;
  }
}

void stackChunkOopDesc::print_on(bool verbose, outputStream* st) const {
  if (*((juint*)this) == badHeapWordVal) {
    st->print_cr("BAD WORD");
  } else {
    InstanceStackChunkKlass::print_chunk(const_cast<stackChunkOopDesc*>(this), verbose, st);
  }
}

#ifdef ASSERT

class StackChunkVerifyOopsClosure : public OopClosure {
  stackChunkOop _chunk;
  int _count;

public:
  StackChunkVerifyOopsClosure(stackChunkOop chunk)
    : _chunk(chunk), _count(0) {}

  void do_oop(oop* p) override { (_chunk->has_bitmap() && UseCompressedOops) ? do_oop_work((narrowOop*)p) : do_oop_work(p); }
  void do_oop(narrowOop* p) override { do_oop_work(p); }

  template <typename T> inline void do_oop_work(T* p) {
    _count++;
    oop obj = _chunk->load_oop(p);
    assert(obj == nullptr || dbg_is_good_oop(obj), "p: " PTR_FORMAT " obj: " PTR_FORMAT, p2i(p), p2i(obj));
    if (_chunk->has_bitmap()) {
      BitMap::idx_t index = _chunk->bit_index_for(p);
      assert(_chunk->bitmap().at(index), "Bit not set at index %zu corresponding to " PTR_FORMAT, index, p2i(p));
    }
  }

  int count() const { return _count; }
};

class VerifyStackChunkFrameClosure {
  stackChunkOop _chunk;

public:
  intptr_t* _sp;
  CodeBlob* _cb;
  bool _callee_interpreted;
  int _size;
  int _argsize;
  int _num_oops;
  int _num_frames;
  int _num_interpreted_frames;
  int _num_i2c;

  VerifyStackChunkFrameClosure(stackChunkOop chunk)
    : _chunk(chunk), _sp(nullptr), _cb(nullptr), _callee_interpreted(false),
      _size(0), _argsize(0), _num_oops(0), _num_frames(0), _num_interpreted_frames(0), _num_i2c(0) {}

  template <ChunkFrames frame_kind, typename RegisterMapT>
  bool do_frame(const StackChunkFrameStream<frame_kind>& f, const RegisterMapT* map) {
    _sp = f.sp();
    _cb = f.cb();

    int fsize = f.frame_size() - ((f.is_interpreted() == _callee_interpreted) ? _argsize : 0);
    int num_oops = f.num_oops();
    assert(num_oops >= 0, "");

    _argsize   = f.stack_argsize() + frame::metadata_words_at_top;
    _size     += fsize;
    _num_oops += num_oops;
    if (f.is_interpreted()) {
      _num_interpreted_frames++;
    }

    log_develop_trace(continuations)("debug_verify_stack_chunk frame: %d sp: " INTPTR_FORMAT " pc: " PTR_FORMAT " interpreted: %d size: %d argsize: %d oops: %d", _num_frames, f.sp() - _chunk->start_address(), p2i(f.pc()), f.is_interpreted(), fsize, _argsize, num_oops);
    LogTarget(Trace, continuations) lt;
    if (lt.develop_is_enabled()) {
      LogStream ls(lt);
      f.print_on(&ls);
    }
    assert(f.pc() != nullptr,
           "young: %d num_frames: %d sp: " PTR_FORMAT " start: " PTR_FORMAT " end: " PTR_FORMAT,
           !_chunk->requires_barriers(), _num_frames, p2i(f.sp()), p2i(_chunk->start_address()), p2i(_chunk->bottom_address()));

    if (_num_frames == 0) {
      assert(f.pc() == _chunk->pc(), "");
    }

    if (_num_frames > 0 && !_callee_interpreted && f.is_interpreted()) {
      log_develop_trace(continuations)("debug_verify_stack_chunk i2c");
      _num_i2c++;
    }

    StackChunkVerifyOopsClosure oops_closure(_chunk);
    f.iterate_oops(&oops_closure, map);
    assert(oops_closure.count() == num_oops, "oops: %d oopmap->num_oops(): %d", oops_closure.count(), num_oops);

    _callee_interpreted = f.is_interpreted();
    _num_frames++;
    return true;
  }
};

template <typename T>
class StackChunkVerifyBitmapClosure : public BitMapClosure {
  stackChunkOop _chunk;

public:
  int _count;

  StackChunkVerifyBitmapClosure(stackChunkOop chunk) : _chunk(chunk), _count(0) {}

  bool do_bit(BitMap::idx_t index) override {
    T* p = _chunk->address_for_bit<T>(index);
    _count++;

    oop obj = _chunk->load_oop(p);
    assert(obj == nullptr || dbg_is_good_oop(obj),
           "p: " PTR_FORMAT " obj: " PTR_FORMAT " index: %zu",
           p2i(p), p2i((oopDesc*)obj), index);

    return true; // continue processing
  }
};

bool stackChunkOopDesc::verify(size_t* out_size, int* out_oops, int* out_frames, int* out_interpreted_frames) {
  DEBUG_ONLY(if (!VerifyContinuations) return true;)

  assert(oopDesc::is_oop(this), "");

  assert(stack_size() >= 0, "");
  assert(!has_bitmap() || is_gc_mode(), "");

  if (is_empty()) {
    assert(max_thawing_size() == 0, "");
  } else {
    assert(argsize() >= 0, "");
  }

  assert(oopDesc::is_oop_or_null(parent()), "");

  const bool concurrent = !Thread::current()->is_Java_thread();

  // If argsize == 0 and the chunk isn't mixed, the chunk contains the metadata (pc, fp -- frame::sender_sp_offset)
  // for the top frame (below sp), and *not* for the bottom frame.
  int size = bottom() - sp();
  assert(size >= 0, "");
  assert((size == 0) == is_empty(), "");

  const StackChunkFrameStream<ChunkFrames::Mixed> first(this);

  VerifyStackChunkFrameClosure closure(this);
  iterate_stack(&closure);

  assert(!is_empty() || closure._cb == nullptr, "");
  if (closure._cb != nullptr && closure._cb->is_nmethod()) {
    assert(argsize() ==
      (closure._cb->as_nmethod()->num_stack_arg_slots()*VMRegImpl::stack_slot_size) >>LogBytesPerWord,
      "chunk argsize: %d bottom frame argsize: %d", argsize(),
      (closure._cb->as_nmethod()->num_stack_arg_slots()*VMRegImpl::stack_slot_size) >>LogBytesPerWord);
  }

  assert(closure._num_interpreted_frames == 0 || has_mixed_frames(), "");

  if (!concurrent) {
    assert(closure._size <= size + (stack_size() - bottom()),
           "size: %d bottom: %d closure.size: %d end sp: " PTR_FORMAT " start sp: %d chunk size: %d",
           size, bottom(), closure._size, closure._sp - start_address(), sp(), stack_size());
    if (closure._num_frames > 0) {
      assert(closure._argsize >= frame::metadata_words_at_top, "should be set up");
      assert(argsize() == closure._argsize - frame::metadata_words_at_top,
             "argsize(): %d closure.argsize: %d closure.callee_interpreted: %d",
             argsize(), closure._argsize, closure._callee_interpreted);
    }

    int calculated_max_size = closure._size
                              + closure._num_i2c * frame::align_wiggle
                              + closure._num_interpreted_frames * frame::align_wiggle;
    assert(max_thawing_size() == calculated_max_size,
           "max_size(): %d calculated_max_size: %d argsize: %d num_i2c: %d",
           max_thawing_size(), calculated_max_size, closure._argsize, closure._num_i2c);

    if (out_size   != nullptr) *out_size   += size;
    if (out_oops   != nullptr) *out_oops   += closure._num_oops;
    if (out_frames != nullptr) *out_frames += closure._num_frames;
    if (out_interpreted_frames != nullptr) *out_interpreted_frames += closure._num_interpreted_frames;
  } else {
    assert(out_size == nullptr, "");
    assert(out_oops == nullptr, "");
    assert(out_frames == nullptr, "");
    assert(out_interpreted_frames == nullptr, "");
  }

  if (has_bitmap()) {
    assert(bitmap().size() == InstanceStackChunkKlass::bitmap_size_in_bits(stack_size()),
           "bitmap().size(): %zu stack_size: %d",
           bitmap().size(), stack_size());

    int oop_count;
    if (UseCompressedOops) {
      StackChunkVerifyBitmapClosure<narrowOop> bitmap_closure(this);
      bitmap().iterate(&bitmap_closure,
                       bit_index_for((narrowOop*)(sp_address() - frame::metadata_words_at_bottom)),
                       bit_index_for((narrowOop*)end_address()));
      oop_count = bitmap_closure._count;
    } else {
      StackChunkVerifyBitmapClosure<oop> bitmap_closure(this);
      bitmap().iterate(&bitmap_closure,
                       bit_index_for((oop*)(sp_address() - frame::metadata_words_at_bottom)),
                       bit_index_for((oop*)end_address()));
      oop_count = bitmap_closure._count;
    }
    assert(oop_count == closure._num_oops,
           "bitmap_closure._count: %d closure._num_oops: %d", oop_count, closure._num_oops);
  }

  return true;
}
#endif