runtime/base/mutex.h

/*
 * Copyright (C) 2011 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#ifndef ART_RUNTIME_BASE_MUTEX_H_
#define ART_RUNTIME_BASE_MUTEX_H_

#include <pthread.h>
#include <stdint.h>
#include <unistd.h>  // for pid_t

#include <iosfwd>
#include <string>

#include <android-base/logging.h>

#include "base/aborting.h"
#include "base/atomic.h"
#include "base/globals.h"
#include "base/macros.h"

#if defined(__linux__)
#define ART_USE_FUTEXES 1
#else
#define ART_USE_FUTEXES 0
#endif

// Currently Darwin doesn't support locks with timeouts.
#if !defined(__APPLE__)
#define HAVE_TIMED_RWLOCK 1
#else
#define HAVE_TIMED_RWLOCK 0
#endif

namespace art {

class SHARED_LOCKABLE ReaderWriterMutex;
class SHARED_LOCKABLE MutatorMutex;
class ScopedContentionRecorder;
class Thread;
class Mutex;

// LockLevel is used to impose a lock hierarchy [1] where acquisition of a Mutex at a higher or
// equal level to a lock a thread holds is invalid. The lock hierarchy achieves a cycle free
// partial ordering and thereby cause deadlock situations to fail checks.
//
// [1] http://www.drdobbs.com/parallel/use-lock-hierarchies-to-avoid-deadlock/204801163
enum LockLevel : uint8_t {
  kLoggingLock = 0,
  kSwapMutexesLock,
  kUnexpectedSignalLock,
  kThreadSuspendCountLock,
  kAbortLock,
  kNativeDebugInterfaceLock,
  kSignalHandlingLock,
  // A generic lock level for mutexs that should not allow any additional mutexes to be gained after
  // acquiring it.
  kGenericBottomLock,
  kJdwpAdbStateLock,
  kJdwpSocketLock,
  kRegionSpaceRegionLock,
  kMarkSweepMarkStackLock,
  kJitCodeCacheLock,
  kRosAllocGlobalLock,
  kRosAllocBracketLock,
  kRosAllocBulkFreeLock,
  kTaggingLockLevel,
  kTransactionLogLock,
  kCustomTlsLock,
  kJniFunctionTableLock,
  kJniWeakGlobalsLock,
  kJniGlobalsLock,
  kReferenceQueueSoftReferencesLock,
  kReferenceQueuePhantomReferencesLock,
  kReferenceQueueFinalizerReferencesLock,
  kReferenceQueueWeakReferencesLock,
  kReferenceQueueClearedReferencesLock,
  kReferenceProcessorLock,
  kJitDebugInterfaceLock,
  kAllocSpaceLock,
  kBumpPointerSpaceBlockLock,
  kArenaPoolLock,
  kInternTableLock,
  kOatFileSecondaryLookupLock,
  kHostDlOpenHandlesLock,
  kVerifierDepsLock,
  kOatFileManagerLock,
  kTracingUniqueMethodsLock,
  kTracingStreamingLock,
  kClassLoaderClassesLock,
  kDefaultMutexLevel,
  kDexLock,
  kMarkSweepLargeObjectLock,
  kJdwpObjectRegistryLock,
  kModifyLdtLock,
  kAllocatedThreadIdsLock,
  kMonitorPoolLock,
  kClassLinkerClassesLock,  // TODO rename.
  kDexToDexCompilerLock,
  kCHALock,
  kSubtypeCheckLock,
  kBreakpointLock,
  kMonitorLock,
  kMonitorListLock,
  kJniLoadLibraryLock,
  kThreadListLock,
  kAllocTrackerLock,
  kDeoptimizationLock,
  kProfilerLock,
  kJdwpShutdownLock,
  kJdwpEventListLock,
  kJdwpAttachLock,
  kJdwpStartLock,
  kRuntimeShutdownLock,
  kTraceLock,
  kHeapBitmapLock,
  kMutatorLock,
  kUserCodeSuspensionLock,
  kInstrumentEntrypointsLock,
  kZygoteCreationLock,

  // The highest valid lock level. Use this if there is code that should only be called with no
  // other locks held. Since this is the highest lock level we also allow it to be held even if the
  // runtime or current thread is not fully set-up yet (for example during thread attach). Note that
  // this lock also has special behavior around the mutator_lock_. Since the mutator_lock_ is not
  // really a 'real' lock we allow this to be locked when the mutator_lock_ is held exclusive.
  // Furthermore, the mutator_lock_ may not be acquired in any form when a lock of this level is
  // held. Since the mutator_lock_ being held strong means that all other threads are suspended this
  // will prevent deadlocks while still allowing this lock level to function as a "highest" level.
  kTopLockLevel,

  kLockLevelCount  // Must come last.
};
std::ostream& operator<<(std::ostream& os, const LockLevel& rhs);

constexpr bool kDebugLocking = kIsDebugBuild;

// Record Log contention information, dumpable via SIGQUIT.
#ifdef ART_USE_FUTEXES
// To enable lock contention logging, set this to true.
constexpr bool kLogLockContentions = false;
#else
// Keep this false as lock contention logging is supported only with
// futex.
constexpr bool kLogLockContentions = false;
#endif
constexpr size_t kContentionLogSize = 4;
constexpr size_t kContentionLogDataSize = kLogLockContentions ? 1 : 0;
constexpr size_t kAllMutexDataSize = kLogLockContentions ? 1 : 0;

// Base class for all Mutex implementations
class BaseMutex {
 public:
  const char* GetName() const {
    return name_;
  }

  virtual bool IsMutex() const { return false; }
  virtual bool IsReaderWriterMutex() const { return false; }
  virtual bool IsMutatorMutex() const { return false; }

  virtual void Dump(std::ostream& os) const = 0;

  static void DumpAll(std::ostream& os);

  bool ShouldRespondToEmptyCheckpointRequest() const {
    return should_respond_to_empty_checkpoint_request_;
  }

  void SetShouldRespondToEmptyCheckpointRequest(bool value) {
    should_respond_to_empty_checkpoint_request_ = value;
  }

  virtual void WakeupToRespondToEmptyCheckpoint() = 0;

 protected:
  friend class ConditionVariable;

  BaseMutex(const char* name, LockLevel level);
  virtual ~BaseMutex();
  void RegisterAsLocked(Thread* self);
  void RegisterAsUnlocked(Thread* self);
  void CheckSafeToWait(Thread* self);

  friend class ScopedContentionRecorder;

  void RecordContention(uint64_t blocked_tid, uint64_t owner_tid, uint64_t nano_time_blocked);
  void DumpContention(std::ostream& os) const;

  const char* const name_;

  // A log entry that records contention but makes no guarantee that either tid will be held live.
  struct ContentionLogEntry {
    ContentionLogEntry() : blocked_tid(0), owner_tid(0) {}
    uint64_t blocked_tid;
    uint64_t owner_tid;
    AtomicInteger count;
  };
  struct ContentionLogData {
    ContentionLogEntry contention_log[kContentionLogSize];
    // The next entry in the contention log to be updated. Value ranges from 0 to
    // kContentionLogSize - 1.
    AtomicInteger cur_content_log_entry;
    // Number of times the Mutex has been contended.
    AtomicInteger contention_count;
    // Sum of time waited by all contenders in ns.
    Atomic<uint64_t> wait_time;
    void AddToWaitTime(uint64_t value);
    ContentionLogData() : wait_time(0) {}
  };
  ContentionLogData contention_log_data_[kContentionLogDataSize];

  const LockLevel level_;  // Support for lock hierarchy.
  bool should_respond_to_empty_checkpoint_request_;

 public:
  bool HasEverContended() const {
    if (kLogLockContentions) {
      return contention_log_data_->contention_count.load(std::memory_order_seq_cst) > 0;
    }
    return false;
  }
};

// A Mutex is used to achieve mutual exclusion between threads. A Mutex can be used to gain
// exclusive access to what it guards. A Mutex can be in one of two states:
// - Free - not owned by any thread,
// - Exclusive - owned by a single thread.
//
// The effect of locking and unlocking operations on the state is:
// State     | ExclusiveLock | ExclusiveUnlock
// -------------------------------------------
// Free      | Exclusive     | error
// Exclusive | Block*        | Free
// * Mutex is not reentrant and so an attempt to ExclusiveLock on the same thread will result in
//   an error. Being non-reentrant simplifies Waiting on ConditionVariables.
std::ostream& operator<<(std::ostream& os, const Mutex& mu);
class LOCKABLE Mutex : public BaseMutex {
 public:
  explicit Mutex(const char* name, LockLevel level = kDefaultMutexLevel, bool recursive = false);
  ~Mutex();

  virtual bool IsMutex() const { return true; }

  // Block until mutex is free then acquire exclusive access.
  void ExclusiveLock(Thread* self) ACQUIRE();
  void Lock(Thread* self) ACQUIRE() {  ExclusiveLock(self); }

  // Returns true if acquires exclusive access, false otherwise.
  bool ExclusiveTryLock(Thread* self) TRY_ACQUIRE(true);
  bool TryLock(Thread* self) TRY_ACQUIRE(true) { return ExclusiveTryLock(self); }

  // Release exclusive access.
  void ExclusiveUnlock(Thread* self) RELEASE();
  void Unlock(Thread* self) RELEASE() {  ExclusiveUnlock(self); }

  // Is the current thread the exclusive holder of the Mutex.
  ALWAYS_INLINE bool IsExclusiveHeld(const Thread* self) const;

  // Assert that the Mutex is exclusively held by the current thread.
  ALWAYS_INLINE void AssertExclusiveHeld(const Thread* self) const ASSERT_CAPABILITY(this);
  ALWAYS_INLINE void AssertHeld(const Thread* self) const ASSERT_CAPABILITY(this);

  // Assert that the Mutex is not held by the current thread.
  void AssertNotHeldExclusive(const Thread* self) ASSERT_CAPABILITY(!*this) {
    if (kDebugLocking && (gAborting == 0)) {
      CHECK(!IsExclusiveHeld(self)) << *this;
    }
  }
  void AssertNotHeld(const Thread* self) ASSERT_CAPABILITY(!*this) {
    AssertNotHeldExclusive(self);
  }

  // Id associated with exclusive owner. No memory ordering semantics if called from a thread other
  // than the owner.
  pid_t GetExclusiveOwnerTid() const;

  // Returns how many times this Mutex has been locked, it is better to use AssertHeld/NotHeld.
  unsigned int GetDepth() const {
    return recursion_count_;
  }

  virtual void Dump(std::ostream& os) const;

  // For negative capabilities in clang annotations.
  const Mutex& operator!() const { return *this; }

  void WakeupToRespondToEmptyCheckpoint() OVERRIDE;

 private:
#if ART_USE_FUTEXES
  // 0 is unheld, 1 is held.
  AtomicInteger state_;
  // Exclusive owner.
  Atomic<pid_t> exclusive_owner_;
  // Number of waiting contenders.
  AtomicInteger num_contenders_;
#else
  pthread_mutex_t mutex_;
  Atomic<pid_t> exclusive_owner_;  // Guarded by mutex_. Asynchronous reads are OK.
#endif

  unsigned int recursion_count_;
  const bool recursive_;  // Can the lock be recursively held?

  friend class ConditionVariable;
  DISALLOW_COPY_AND_ASSIGN(Mutex);
};

// A ReaderWriterMutex is used to achieve mutual exclusion between threads, similar to a Mutex.
// Unlike a Mutex a ReaderWriterMutex can be used to gain exclusive (writer) or shared (reader)
// access to what it guards. A flaw in relation to a Mutex is that it cannot be used with a
// condition variable. A ReaderWriterMutex can be in one of three states:
// - Free - not owned by any thread,
// - Exclusive - owned by a single thread,
// - Shared(n) - shared amongst n threads.
//
// The effect of locking and unlocking operations on the state is:
//
// State     | ExclusiveLock | ExclusiveUnlock | SharedLock       | SharedUnlock
// ----------------------------------------------------------------------------
// Free      | Exclusive     | error           | SharedLock(1)    | error
// Exclusive | Block         | Free            | Block            | error
// Shared(n) | Block         | error           | SharedLock(n+1)* | Shared(n-1) or Free
// * for large values of n the SharedLock may block.
std::ostream& operator<<(std::ostream& os, const ReaderWriterMutex& mu);
class SHARED_LOCKABLE ReaderWriterMutex : public BaseMutex {
 public:
  explicit ReaderWriterMutex(const char* name, LockLevel level = kDefaultMutexLevel);
  ~ReaderWriterMutex();

  virtual bool IsReaderWriterMutex() const { return true; }

  // Block until ReaderWriterMutex is free then acquire exclusive access.
  void ExclusiveLock(Thread* self) ACQUIRE();
  void WriterLock(Thread* self) ACQUIRE() {  ExclusiveLock(self); }

  // Release exclusive access.
  void ExclusiveUnlock(Thread* self) RELEASE();
  void WriterUnlock(Thread* self) RELEASE() {  ExclusiveUnlock(self); }

  // Block until ReaderWriterMutex is free and acquire exclusive access. Returns true on success
  // or false if timeout is reached.
#if HAVE_TIMED_RWLOCK
  bool ExclusiveLockWithTimeout(Thread* self, int64_t ms, int32_t ns)
      EXCLUSIVE_TRYLOCK_FUNCTION(true);
#endif

  // Block until ReaderWriterMutex is shared or free then acquire a share on the access.
  void SharedLock(Thread* self) ACQUIRE_SHARED() ALWAYS_INLINE;
  void ReaderLock(Thread* self) ACQUIRE_SHARED() { SharedLock(self); }

  // Try to acquire share of ReaderWriterMutex.
  bool SharedTryLock(Thread* self) SHARED_TRYLOCK_FUNCTION(true);

  // Release a share of the access.
  void SharedUnlock(Thread* self) RELEASE_SHARED() ALWAYS_INLINE;
  void ReaderUnlock(Thread* self) RELEASE_SHARED() { SharedUnlock(self); }

  // Is the current thread the exclusive holder of the ReaderWriterMutex.
  ALWAYS_INLINE bool IsExclusiveHeld(const Thread* self) const;

  // Assert the current thread has exclusive access to the ReaderWriterMutex.
  ALWAYS_INLINE void AssertExclusiveHeld(const Thread* self) const ASSERT_CAPABILITY(this);
  ALWAYS_INLINE void AssertWriterHeld(const Thread* self) const ASSERT_CAPABILITY(this);

  // Assert the current thread doesn't have exclusive access to the ReaderWriterMutex.
  void AssertNotExclusiveHeld(const Thread* self) ASSERT_CAPABILITY(!this) {
    if (kDebugLocking && (gAborting == 0)) {
      CHECK(!IsExclusiveHeld(self)) << *this;
    }
  }
  void AssertNotWriterHeld(const Thread* self) ASSERT_CAPABILITY(!this) {
    AssertNotExclusiveHeld(self);
  }

  // Is the current thread a shared holder of the ReaderWriterMutex.
  bool IsSharedHeld(const Thread* self) const;

  // Assert the current thread has shared access to the ReaderWriterMutex.
  ALWAYS_INLINE void AssertSharedHeld(const Thread* self) ASSERT_SHARED_CAPABILITY(this) {
    if (kDebugLocking && (gAborting == 0)) {
      // TODO: we can only assert this well when self != null.
      CHECK(IsSharedHeld(self) || self == nullptr) << *this;
    }
  }
  ALWAYS_INLINE void AssertReaderHeld(const Thread* self) ASSERT_SHARED_CAPABILITY(this) {
    AssertSharedHeld(self);
  }

  // Assert the current thread doesn't hold this ReaderWriterMutex either in shared or exclusive
  // mode.
  ALWAYS_INLINE void AssertNotHeld(const Thread* self) ASSERT_SHARED_CAPABILITY(!this) {
    if (kDebugLocking && (gAborting == 0)) {
      CHECK(!IsSharedHeld(self)) << *this;
    }
  }

  // Id associated with exclusive owner. No memory ordering semantics if called from a thread other
  // than the owner. Returns 0 if the lock is not held. Returns either 0 or -1 if it is held by
  // one or more readers.
  pid_t GetExclusiveOwnerTid() const;

  virtual void Dump(std::ostream& os) const;

  // For negative capabilities in clang annotations.
  const ReaderWriterMutex& operator!() const { return *this; }

  void WakeupToRespondToEmptyCheckpoint() OVERRIDE;

 private:
#if ART_USE_FUTEXES
  // Out-of-inline path for handling contention for a SharedLock.
  void HandleSharedLockContention(Thread* self, int32_t cur_state);

  // -1 implies held exclusive, +ve shared held by state_ many owners.
  AtomicInteger state_;
  // Exclusive owner. Modification guarded by this mutex.
  Atomic<pid_t> exclusive_owner_;
  // Number of contenders waiting for a reader share.
  AtomicInteger num_pending_readers_;
  // Number of contenders waiting to be the writer.
  AtomicInteger num_pending_writers_;
#else
  pthread_rwlock_t rwlock_;
  Atomic<pid_t> exclusive_owner_;  // Writes guarded by rwlock_. Asynchronous reads are OK.
#endif
  DISALLOW_COPY_AND_ASSIGN(ReaderWriterMutex);
};

// MutatorMutex is a special kind of ReaderWriterMutex created specifically for the
// Locks::mutator_lock_ mutex. The behaviour is identical to the ReaderWriterMutex except that
// thread state changes also play a part in lock ownership. The mutator_lock_ will not be truly
// held by any mutator threads. However, a thread in the kRunnable state is considered to have
// shared ownership of the mutator lock and therefore transitions in and out of the kRunnable
// state have associated implications on lock ownership. Extra methods to handle the state
// transitions have been added to the interface but are only accessible to the methods dealing
// with state transitions. The thread state and flags attributes are used to ensure thread state
// transitions are consistent with the permitted behaviour of the mutex.
//
// *) The most important consequence of this behaviour is that all threads must be in one of the
// suspended states before exclusive ownership of the mutator mutex is sought.
//
std::ostream& operator<<(std::ostream& os, const MutatorMutex& mu);
class SHARED_LOCKABLE MutatorMutex : public ReaderWriterMutex {
 public:
  explicit MutatorMutex(const char* name, LockLevel level = kDefaultMutexLevel)
    : ReaderWriterMutex(name, level) {}
  ~MutatorMutex() {}

  virtual bool IsMutatorMutex() const { return true; }

  // For negative capabilities in clang annotations.
  const MutatorMutex& operator!() const { return *this; }

 private:
  friend class Thread;
  void TransitionFromRunnableToSuspended(Thread* self) UNLOCK_FUNCTION() ALWAYS_INLINE;
  void TransitionFromSuspendedToRunnable(Thread* self) SHARED_LOCK_FUNCTION() ALWAYS_INLINE;

  DISALLOW_COPY_AND_ASSIGN(MutatorMutex);
};

// ConditionVariables allow threads to queue and sleep. Threads may then be resumed individually
// (Signal) or all at once (Broadcast).
class ConditionVariable {
 public:
  ConditionVariable(const char* name, Mutex& mutex);
  ~ConditionVariable();

  void Broadcast(Thread* self);
  void Signal(Thread* self);
  // TODO: No thread safety analysis on Wait and TimedWait as they call mutex operations via their
  //       pointer copy, thereby defeating annotalysis.
  void Wait(Thread* self) NO_THREAD_SAFETY_ANALYSIS;
  bool TimedWait(Thread* self, int64_t ms, int32_t ns) NO_THREAD_SAFETY_ANALYSIS;
  // Variant of Wait that should be used with caution. Doesn't validate that no mutexes are held
  // when waiting.
  // TODO: remove this.
  void WaitHoldingLocks(Thread* self) NO_THREAD_SAFETY_ANALYSIS;

 private:
  const char* const name_;
  // The Mutex being used by waiters. It is an error to mix condition variables between different
  // Mutexes.
  Mutex& guard_;
#if ART_USE_FUTEXES
  // A counter that is modified by signals and broadcasts. This ensures that when a waiter gives up
  // their Mutex and another thread takes it and signals, the waiting thread observes that sequence_
  // changed and doesn't enter the wait. Modified while holding guard_, but is read by futex wait
  // without guard_ held.
  AtomicInteger sequence_;
  // Number of threads that have come into to wait, not the length of the waiters on the futex as
  // waiters may have been requeued onto guard_. Guarded by guard_.
  volatile int32_t num_waiters_;
#else
  pthread_cond_t cond_;
#endif
  DISALLOW_COPY_AND_ASSIGN(ConditionVariable);
};

// Scoped locker/unlocker for a regular Mutex that acquires mu upon construction and releases it
// upon destruction.
class SCOPED_CAPABILITY MutexLock {
 public:
  MutexLock(Thread* self, Mutex& mu) ACQUIRE(mu) : self_(self), mu_(mu) {
    mu_.ExclusiveLock(self_);
  }

  ~MutexLock() RELEASE() {
    mu_.ExclusiveUnlock(self_);
  }

 private:
  Thread* const self_;
  Mutex& mu_;
  DISALLOW_COPY_AND_ASSIGN(MutexLock);
};

// Scoped locker/unlocker for a ReaderWriterMutex that acquires read access to mu upon
// construction and releases it upon destruction.
class SCOPED_CAPABILITY ReaderMutexLock {
 public:
  ALWAYS_INLINE ReaderMutexLock(Thread* self, ReaderWriterMutex& mu) ACQUIRE(mu);

  ALWAYS_INLINE ~ReaderMutexLock() RELEASE();

 private:
  Thread* const self_;
  ReaderWriterMutex& mu_;
  DISALLOW_COPY_AND_ASSIGN(ReaderMutexLock);
};

// Scoped locker/unlocker for a ReaderWriterMutex that acquires write access to mu upon
// construction and releases it upon destruction.
class SCOPED_CAPABILITY WriterMutexLock {
 public:
  WriterMutexLock(Thread* self, ReaderWriterMutex& mu) EXCLUSIVE_LOCK_FUNCTION(mu) :
      self_(self), mu_(mu) {
    mu_.ExclusiveLock(self_);
  }

  ~WriterMutexLock() UNLOCK_FUNCTION() {
    mu_.ExclusiveUnlock(self_);
  }

 private:
  Thread* const self_;
  ReaderWriterMutex& mu_;
  DISALLOW_COPY_AND_ASSIGN(WriterMutexLock);
};

// For StartNoThreadSuspension and EndNoThreadSuspension.
class CAPABILITY("role") Role {
 public:
  void Acquire() ACQUIRE() {}
  void Release() RELEASE() {}
  const Role& operator!() const { return *this; }
};

class Uninterruptible : public Role {
};

// Global mutexes corresponding to the levels above.
class Locks {
 public:
  static void Init();
  static void InitConditions() NO_THREAD_SAFETY_ANALYSIS;  // Condition variables.

  // Destroying various lock types can emit errors that vary depending upon
  // whether the client (art::Runtime) is currently active.  Allow the client
  // to set a callback that is used to check when it is acceptable to call
  // Abort.  The default behavior is that the client *is not* able to call
  // Abort if no callback is established.
  using ClientCallback = bool();
  static void SetClientCallback(ClientCallback* is_safe_to_call_abort_cb) NO_THREAD_SAFETY_ANALYSIS;
  // Checks for whether it is safe to call Abort() without using locks.
  static bool IsSafeToCallAbortRacy() NO_THREAD_SAFETY_ANALYSIS;

  // Add a mutex to expected_mutexes_on_weak_ref_access_.
  static void AddToExpectedMutexesOnWeakRefAccess(BaseMutex* mutex, bool need_lock = true);
  // Remove a mutex from expected_mutexes_on_weak_ref_access_.
  static void RemoveFromExpectedMutexesOnWeakRefAccess(BaseMutex* mutex, bool need_lock = true);
  // Check if the given mutex is in expected_mutexes_on_weak_ref_access_.
  static bool IsExpectedOnWeakRefAccess(BaseMutex* mutex);

  // Guards allocation entrypoint instrumenting.
  static Mutex* instrument_entrypoints_lock_;

  // Guards code that deals with user-code suspension. This mutex must be held when suspending or
  // resuming threads with SuspendReason::kForUserCode. It may be held by a suspended thread, but
  // only if the suspension is not due to SuspendReason::kForUserCode.
  static Mutex* user_code_suspension_lock_ ACQUIRED_AFTER(instrument_entrypoints_lock_);

  // A barrier is used to synchronize the GC/Debugger thread with mutator threads. When GC/Debugger
  // thread wants to suspend all mutator threads, it needs to wait for all mutator threads to pass
  // a barrier. Threads that are already suspended will get their barrier passed by the GC/Debugger
  // thread; threads in the runnable state will pass the barrier when they transit to the suspended
  // state. GC/Debugger thread will be woken up when all mutator threads are suspended.
  //
  // Thread suspension:
  // mutator thread                                | GC/Debugger
  //   .. running ..                               |   .. running ..
  //   .. running ..                               | Request thread suspension by:
  //   .. running ..                               |   - acquiring thread_suspend_count_lock_
  //   .. running ..                               |   - incrementing Thread::suspend_count_ on
  //   .. running ..                               |     all mutator threads
  //   .. running ..                               |   - releasing thread_suspend_count_lock_
  //   .. running ..                               | Block wait for all threads to pass a barrier
  // Poll Thread::suspend_count_ and enter full    |   .. blocked ..
  // suspend code.                                 |   .. blocked ..
  // Change state to kSuspended (pass the barrier) | Wake up when all threads pass the barrier
  // x: Acquire thread_suspend_count_lock_         |   .. running ..
  // while Thread::suspend_count_ > 0              |   .. running ..
  //   - wait on Thread::resume_cond_              |   .. running ..
  //     (releases thread_suspend_count_lock_)     |   .. running ..
  //   .. waiting ..                               | Request thread resumption by:
  //   .. waiting ..                               |   - acquiring thread_suspend_count_lock_
  //   .. waiting ..                               |   - decrementing Thread::suspend_count_ on
  //   .. waiting ..                               |     all mutator threads
  //   .. waiting ..                               |   - notifying on Thread::resume_cond_
  //    - re-acquire thread_suspend_count_lock_    |   - releasing thread_suspend_count_lock_
  // Release thread_suspend_count_lock_            |  .. running ..
  // Change to kRunnable                           |  .. running ..
  //  - this uses a CAS operation to ensure the    |  .. running ..
  //    suspend request flag isn't raised as the   |  .. running ..
  //    state is changed                           |  .. running ..
  //  - if the CAS operation fails then goto x     |  .. running ..
  //  .. running ..                                |  .. running ..
  static MutatorMutex* mutator_lock_ ACQUIRED_AFTER(user_code_suspension_lock_);

  // Allow reader-writer mutual exclusion on the mark and live bitmaps of the heap.
  static ReaderWriterMutex* heap_bitmap_lock_ ACQUIRED_AFTER(mutator_lock_);

  // Guards shutdown of the runtime.
  static Mutex* runtime_shutdown_lock_ ACQUIRED_AFTER(heap_bitmap_lock_);

  // Guards background profiler global state.
  static Mutex* profiler_lock_ ACQUIRED_AFTER(runtime_shutdown_lock_);

  // Guards trace (ie traceview) requests.
  static Mutex* trace_lock_ ACQUIRED_AFTER(profiler_lock_);

  // Guards debugger recent allocation records.
  static Mutex* alloc_tracker_lock_ ACQUIRED_AFTER(trace_lock_);

  // Guards updates to instrumentation to ensure mutual exclusion of
  // events like deoptimization requests.
  // TODO: improve name, perhaps instrumentation_update_lock_.
  static Mutex* deoptimization_lock_ ACQUIRED_AFTER(alloc_tracker_lock_);

  // Guards Class Hierarchy Analysis (CHA).
  static Mutex* cha_lock_ ACQUIRED_AFTER(deoptimization_lock_);

  // Guard the update of the SubtypeCheck data stores in each Class::status_ field.
  // This lock is used in SubtypeCheck methods which are the interface for
  // any SubtypeCheck-mutating methods.
  // In Class::IsSubClass, the lock is not required since it does not update the SubtypeCheck data.
  static Mutex* subtype_check_lock_ ACQUIRED_AFTER(cha_lock_);

  // The thread_list_lock_ guards ThreadList::list_. It is also commonly held to stop threads
  // attaching and detaching.
  static Mutex* thread_list_lock_ ACQUIRED_AFTER(subtype_check_lock_);

  // Signaled when threads terminate. Used to determine when all non-daemons have terminated.
  static ConditionVariable* thread_exit_cond_ GUARDED_BY(Locks::thread_list_lock_);

  // Guards maintaining loading library data structures.
  static Mutex* jni_libraries_lock_ ACQUIRED_AFTER(thread_list_lock_);

  // Guards breakpoints.
  static ReaderWriterMutex* breakpoint_lock_ ACQUIRED_AFTER(jni_libraries_lock_);

  // Guards lists of classes within the class linker.
  static ReaderWriterMutex* classlinker_classes_lock_ ACQUIRED_AFTER(breakpoint_lock_);

  // When declaring any Mutex add DEFAULT_MUTEX_ACQUIRED_AFTER to use annotalysis to check the code
  // doesn't try to hold a higher level Mutex.
  #define DEFAULT_MUTEX_ACQUIRED_AFTER ACQUIRED_AFTER(art::Locks::classlinker_classes_lock_)

  static Mutex* allocated_monitor_ids_lock_ ACQUIRED_AFTER(classlinker_classes_lock_);

  // Guard the allocation/deallocation of thread ids.
  static Mutex* allocated_thread_ids_lock_ ACQUIRED_AFTER(allocated_monitor_ids_lock_);

  // Guards modification of the LDT on x86.
  static Mutex* modify_ldt_lock_ ACQUIRED_AFTER(allocated_thread_ids_lock_);

  static ReaderWriterMutex* dex_lock_ ACQUIRED_AFTER(modify_ldt_lock_);

  // Guards opened oat files in OatFileManager.
  static ReaderWriterMutex* oat_file_manager_lock_ ACQUIRED_AFTER(dex_lock_);

  // Guards extra string entries for VerifierDeps.
  static ReaderWriterMutex* verifier_deps_lock_ ACQUIRED_AFTER(oat_file_manager_lock_);

  // Guards dlopen_handles_ in DlOpenOatFile.
  static Mutex* host_dlopen_handles_lock_ ACQUIRED_AFTER(verifier_deps_lock_);

  // Guards intern table.
  static Mutex* intern_table_lock_ ACQUIRED_AFTER(host_dlopen_handles_lock_);

  // Guards reference processor.
  static Mutex* reference_processor_lock_ ACQUIRED_AFTER(intern_table_lock_);

  // Guards cleared references queue.
  static Mutex* reference_queue_cleared_references_lock_ ACQUIRED_AFTER(reference_processor_lock_);

  // Guards weak references queue.
  static Mutex* reference_queue_weak_references_lock_ ACQUIRED_AFTER(reference_queue_cleared_references_lock_);

  // Guards finalizer references queue.
  static Mutex* reference_queue_finalizer_references_lock_ ACQUIRED_AFTER(reference_queue_weak_references_lock_);

  // Guards phantom references queue.
  static Mutex* reference_queue_phantom_references_lock_ ACQUIRED_AFTER(reference_queue_finalizer_references_lock_);

  // Guards soft references queue.
  static Mutex* reference_queue_soft_references_lock_ ACQUIRED_AFTER(reference_queue_phantom_references_lock_);

  // Guard accesses to the JNI Global Reference table.
  static ReaderWriterMutex* jni_globals_lock_ ACQUIRED_AFTER(reference_queue_soft_references_lock_);

  // Guard accesses to the JNI Weak Global Reference table.
  static Mutex* jni_weak_globals_lock_ ACQUIRED_AFTER(jni_globals_lock_);

  // Guard accesses to the JNI function table override.
  static Mutex* jni_function_table_lock_ ACQUIRED_AFTER(jni_weak_globals_lock_);

  // Guard accesses to the Thread::custom_tls_. We use this to allow the TLS of other threads to be
  // read (the reader must hold the ThreadListLock or have some other way of ensuring the thread
  // will not die in that case though). This is useful for (eg) the implementation of
  // GetThreadLocalStorage.
  static Mutex* custom_tls_lock_ ACQUIRED_AFTER(jni_function_table_lock_);

  // When declaring any Mutex add BOTTOM_MUTEX_ACQUIRED_AFTER to use annotalysis to check the code
  // doesn't try to acquire a higher level Mutex. NB Due to the way the annotalysis works this
  // actually only encodes the mutex being below jni_function_table_lock_ although having
  // kGenericBottomLock level is lower than this.
  #define BOTTOM_MUTEX_ACQUIRED_AFTER ACQUIRED_AFTER(art::Locks::custom_tls_lock_)

  // Have an exclusive aborting thread.
  static Mutex* abort_lock_ ACQUIRED_AFTER(custom_tls_lock_);

  // Allow mutual exclusion when manipulating Thread::suspend_count_.
  // TODO: Does the trade-off of a per-thread lock make sense?
  static Mutex* thread_suspend_count_lock_ ACQUIRED_AFTER(abort_lock_);

  // One unexpected signal at a time lock.
  static Mutex* unexpected_signal_lock_ ACQUIRED_AFTER(thread_suspend_count_lock_);

  // Guards the magic global variables used by native tools (e.g. libunwind).
  static Mutex* native_debug_interface_lock_ ACQUIRED_AFTER(unexpected_signal_lock_);

  // Have an exclusive logging thread.
  static Mutex* logging_lock_ ACQUIRED_AFTER(native_debug_interface_lock_);

  // List of mutexes that we expect a thread may hold when accessing weak refs. This is used to
  // avoid a deadlock in the empty checkpoint while weak ref access is disabled (b/34964016). If we
  // encounter an unexpected mutex on accessing weak refs,
  // Thread::CheckEmptyCheckpointFromWeakRefAccess will detect it.
  static std::vector<BaseMutex*> expected_mutexes_on_weak_ref_access_;
  static Atomic<const BaseMutex*> expected_mutexes_on_weak_ref_access_guard_;
  class ScopedExpectedMutexesOnWeakRefAccessLock;
};

class Roles {
 public:
  // Uninterruptible means that the thread may not become suspended.
  static Uninterruptible uninterruptible_;
};

}  // namespace art

#endif  // ART_RUNTIME_BASE_MUTEX_H_