base/lazy_instance.h

// Copyright (c) 2012 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

// The LazyInstance<Type, Traits> class manages a single instance of Type,
// which will be lazily created on the first time it's accessed.  This class is
// useful for places you would normally use a function-level static, but you
// need to have guaranteed thread-safety.  The Type constructor will only ever
// be called once, even if two threads are racing to create the object.  Get()
// and Pointer() will always return the same, completely initialized instance.
// When the instance is constructed it is registered with AtExitManager.  The
// destructor will be called on program exit.
//
// LazyInstance is completely thread safe, assuming that you create it safely.
// The class was designed to be POD initialized, so it shouldn't require a
// static constructor.  It really only makes sense to declare a LazyInstance as
// a global variable using the LAZY_INSTANCE_INITIALIZER initializer.
//
// LazyInstance is similar to Singleton, except it does not have the singleton
// property.  You can have multiple LazyInstance's of the same type, and each
// will manage a unique instance.  It also preallocates the space for Type, as
// to avoid allocating the Type instance on the heap.  This may help with the
// performance of creating the instance, and reducing heap fragmentation.  This
// requires that Type be a complete type so we can determine the size.
//
// Example usage:
//   static LazyInstance<MyClass> my_instance = LAZY_INSTANCE_INITIALIZER;
//   void SomeMethod() {
//     my_instance.Get().SomeMethod();  // MyClass::SomeMethod()
//
//     MyClass* ptr = my_instance.Pointer();
//     ptr->DoDoDo();  // MyClass::DoDoDo
//   }

#ifndef BASE_LAZY_INSTANCE_H_
#define BASE_LAZY_INSTANCE_H_
#pragma once

#include <new>  // For placement new.

#include "base/atomicops.h"
#include "base/base_export.h"
#include "base/basictypes.h"
#include "base/logging.h"
#include "base/third_party/dynamic_annotations/dynamic_annotations.h"
#include "base/threading/thread_restrictions.h"

// LazyInstance uses its own struct initializer-list style static
// initialization, as base's LINKER_INITIALIZED requires a constructor and on
// some compilers (notably gcc 4.4) this still ends up needing runtime
// initialization.
#define LAZY_INSTANCE_INITIALIZER {0}

namespace base {

template <typename Type>
struct DefaultLazyInstanceTraits {
  static const bool kRegisterOnExit = true;
  static const bool kAllowedToAccessOnNonjoinableThread = false;

  static Type* New(void* instance) {
    DCHECK_EQ(reinterpret_cast<uintptr_t>(instance) % sizeof(instance), 0u)
        << ": Bad boy, the buffer passed to placement new is not aligned!\n"
        "This may break some stuff like SSE-based optimizations assuming the "
        "<Type> objects are word aligned.";
    // Use placement new to initialize our instance in our preallocated space.
    // The parenthesis is very important here to force POD type initialization.
    return new (instance) Type();
  }
  static void Delete(Type* instance) {
    // Explicitly call the destructor.
    instance->~Type();
  }
};

// Use LazyInstance<T>::Leaky for a less-verbose call-site typedef; e.g.:
// base::LazyInstance<T>::Leaky my_leaky_lazy_instance;
// instead of:
// base::LazyInstance<T, LeakyLazyInstanceTraits<T> > my_leaky_lazy_instance;
// (especially when T is MyLongTypeNameImplClientHolderFactory).
template <typename Type>
struct LeakyLazyInstanceTraits {
  static const bool kRegisterOnExit = false;
  static const bool kAllowedToAccessOnNonjoinableThread = true;

  static Type* New(void* instance) {
    return DefaultLazyInstanceTraits<Type>::New(instance);
  }
  static void Delete(Type* instance) {
  }
};

// We pull out some of the functionality into non-templated functions, so we
// can implement the more complicated pieces out of line in the .cc file.
namespace internal {

// Our AtomicWord doubles as a spinlock, where a value of
// kBeingCreatedMarker means the spinlock is being held for creation.
static const subtle::AtomicWord kLazyInstanceStateCreating = 1;

// Check if instance needs to be created. If so return true otherwise
// if another thread has beat us, wait for instance to be created and
// return false.
BASE_EXPORT bool NeedsLazyInstance(subtle::AtomicWord* state);

// After creating an instance, call this to register the dtor to be called
// at program exit and to update the atomic state to hold the |new_instance|
BASE_EXPORT void CompleteLazyInstance(subtle::AtomicWord* state,
                                      subtle::AtomicWord new_instance,
                                      void* lazy_instance,
                                      void (*dtor)(void*));

}  // namespace internal

template <typename Type, typename Traits = DefaultLazyInstanceTraits<Type> >
class LazyInstance {
 public:
  // Do not define a destructor, as doing so makes LazyInstance a
  // non-POD-struct. We don't want that because then a static initializer will
  // be created to register the (empty) destructor with atexit() under MSVC, for
  // example. We handle destruction of the contained Type class explicitly via
  // the OnExit member function, where needed.
  // ~LazyInstance() {}

  // Convenience typedef to avoid having to repeat Type for leaky lazy
  // instances.
  typedef LazyInstance<Type, LeakyLazyInstanceTraits<Type> > Leaky;

  Type& Get() {
    return *Pointer();
  }

  Type* Pointer() {
#ifndef NDEBUG
    // Avoid making TLS lookup on release builds.
    if (!Traits::kAllowedToAccessOnNonjoinableThread)
      ThreadRestrictions::AssertSingletonAllowed();
#endif
    // If any bit in the created mask is true, the instance has already been
    // fully constructed.
    static const subtle::AtomicWord kLazyInstanceCreatedMask =
        ~internal::kLazyInstanceStateCreating;

    // We will hopefully have fast access when the instance is already created.
    // Since a thread sees private_instance_ == 0 or kLazyInstanceStateCreating
    // at most once, the load is taken out of NeedsInstance() as a fast-path.
    // The load has acquire memory ordering as a thread which sees
    // private_instance_ > creating needs to acquire visibility over
    // the associated data (private_buf_). Pairing Release_Store is in
    // CompleteLazyInstance().
    subtle::AtomicWord value = subtle::Acquire_Load(&private_instance_);
    if (!(value & kLazyInstanceCreatedMask) &&
        internal::NeedsLazyInstance(&private_instance_)) {
      // Create the instance in the space provided by |private_buf_|.
      value = reinterpret_cast<subtle::AtomicWord>(Traits::New(private_buf_));
      internal::CompleteLazyInstance(&private_instance_, value, this,
                                     Traits::kRegisterOnExit ? OnExit : NULL);
    }

    // This annotation helps race detectors recognize correct lock-less
    // synchronization between different threads calling Pointer().
    // We suggest dynamic race detection tool that "Traits::New" above
    // and CompleteLazyInstance(...) happens before "return instance()" below.
    // See the corresponding HAPPENS_BEFORE in CompleteLazyInstance(...).
    ANNOTATE_HAPPENS_AFTER(&private_instance_);
    return instance();
  }

  bool operator==(Type* p) {
    switch (subtle::NoBarrier_Load(&private_instance_)) {
      case 0:
        return p == NULL;
      case internal::kLazyInstanceStateCreating:
        return static_cast<int8*>(static_cast<void*>(p)) == private_buf_;
      default:
        return p == instance();
    }
  }

  // Effectively private: member data is only public to allow the linker to
  // statically initialize it. DO NOT USE FROM OUTSIDE THIS CLASS.

  // Note this must use AtomicWord, not Atomic32, to ensure correct alignment
  // of |private_buf_| on 64 bit architectures. (This member must be first to
  // allow the syntax used in LAZY_INSTANCE_INITIALIZER to work correctly.)
  subtle::AtomicWord private_instance_;
  int8 private_buf_[sizeof(Type)];  // Preallocated space for the Type instance.

 private:
  Type* instance() {
    return reinterpret_cast<Type*>(subtle::NoBarrier_Load(&private_instance_));
  }

  // Adapter function for use with AtExit.  This should be called single
  // threaded, so don't synchronize across threads.
  // Calling OnExit while the instance is in use by other threads is a mistake.
  static void OnExit(void* lazy_instance) {
    LazyInstance<Type, Traits>* me =
        reinterpret_cast<LazyInstance<Type, Traits>*>(lazy_instance);
    Traits::Delete(me->instance());
    subtle::Release_Store(&me->private_instance_, 0);
  }
};

}  // namespace base

#endif  // BASE_LAZY_INSTANCE_H_