boost.png (6897 bytes) Home Libraries People FAQ More

PrevUpHomeNext

Concepts

Mutexes

Mutexes

Note

Certain changes to the mutexes and lock concepts are currently under discussion. In particular, the combination of the multiple lock concepts into a single lock concept is likely, and the combination of the multiple mutex concepts into a single mutex concept is also possible.

A mutex (short for mutual-exclusion) object is used to serialize access to a resource shared between multiple threads. The Mutex concept, with TryMutex and TimedMutex refinements, formalize the requirements. A model that implements Mutex and its refinements has two states: locked and unlocked. Before using a shared resource, a thread locks a Boost.Threads mutex object (an object whose type is a model of Mutex or one of it's refinements), ensuring thread-safe access to the shared resource. When use of the shared resource is complete, the thread unlocks the mutex object, allowing another thread to acquire the lock and use the shared resource.

Traditional C thread APIs, like POSIX threads or the Windows thread APIs, expose functions to lock and unlock a mutex object. This is dangerous since it's easy to forget to unlock a locked mutex. When the flow of control is complex, with multiple return points, the likelihood of forgetting to unlock a mutex object becomes even greater. When exceptions are thrown, it becomes nearly impossible to ensure that the mutex object is unlocked properly when using these traditional API's. The result is deadlock.

Many C++ threading libraries use a pattern known as Scoped Locking[SchmidtStalRohnertBuschmann] to free the programmer from the need to explicitly lock and unlock mutex objects. With this pattern, a Lock concept is employed where the lock object's constructor locks the associated mutex object and the destructor automatically does the unlocking. The Boost.Threads library takes this pattern to the extreme in that Lock concepts are the only way to lock and unlock a mutex object: lock and unlock functions are not exposed by any Boost.Threads mutex objects. This helps to ensure safe usage patterns, especially when code throws exceptions.

Locking Strategies

Every mutex object follows one of several locking strategies. These strategies define the semantics for the locking operation when the calling thread already owns a lock on the mutex object.

Recursive Locking Strategy

With a recursive locking strategy, when a thread attempts to acquire a lock on the mutex object for which it already owns a lock, the operation is successful. Note the distinction between a thread, which may have multiple locks outstanding on a recursive mutex object, and a lock object, which even for a recursive mutex object cannot have any of its lock functions called multiple times without first calling unlock.

Internally a lock count is maintained and the owning thread must unlock the mutex object the same number of times that it locked it before the mutex object's state returns to unlocked. Since mutex objects in Boost.Threads expose locking functionality only through lock concepts, a thread will always unlock a mutex object the same number of times that it locked it. This helps to eliminate a whole set of errors typically found in traditional C style thread APIs.

Classes boost::recursive_mutex, boost::recursive_try_mutex and boost::recursive_timed_mutex use this locking strategy.

Checked Locking Strategy

With a checked locking strategy, when a thread attempts to acquire a lock on the mutex object for which the thread already owns a lock, the operation will fail with some sort of error indication. Further, attempts by a thread to unlock a mutex object that was not locked by the thread will also return some sort of error indication. In Boost.Threads, an exception of type boost::lock_error would be thrown in these cases.

Boost.Threads does not currently provide any mutex objects that use this strategy.

Unchecked Locking Strategy

With an unchecked locking strategy, when a thread attempts to acquire a lock on a mutex object for which the thread already owns a lock the operation will deadlock. In general this locking strategy is less safe than a checked or recursive strategy, but it's also a faster strategy and so is employed by many libraries.

Boost.Threads does not currently provide any mutex objects that use this strategy.

Unspecified Locking Strategy

With an unspecified locking strategy, when a thread attempts to acquire a lock on a mutex object for which the thread already owns a lock the operation results in undefined behavior.

In general a mutex object with an unspecified locking strategy is unsafe, and it requires programmer discipline to use the mutex object properly. However, this strategy allows an implementation to be as fast as possible with no restrictions on its implementation. This is especially true for portable implementations that wrap the native threading support of a platform. For this reason, the classes boost::mutex, boost::try_mutex and boost::timed_mutex use this locking strategy despite the lack of safety.

Scheduling Policies

Every mutex object follows one of several scheduling policies. These policies define the semantics when the mutex object is unlocked and there is more than one thread waiting to acquire a lock. In other words, the policy defines which waiting thread shall acquire the lock.

FIFO Scheduling Policy

With a FIFO ("First In First Out") scheduling policy, threads waiting for the lock will acquire it in a first-come-first-served order. This can help prevent a high priority thread from starving lower priority threads that are also waiting on the mutex object's lock.

Priority Driven Policy

With a Priority Driven scheduling policy, the thread with the highest priority acquires the lock. Note that this means that low-priority threads may never acquire the lock if the mutex object has high contention and there is always at least one high-priority thread waiting. This is known as thread starvation. When multiple threads of the same priority are waiting on the mutex object's lock one of the other scheduling priorities will determine which thread shall acquire the lock.

Unspecified Policy

The mutex object does not specify a scheduling policy. In order to ensure portability, all Boost.Threads mutex objects use an unspecified scheduling policy.

Mutex Concepts
Mutex Concept

A Mutex object has two states: locked and unlocked. Mutex object state can only be determined by a lock object meeting the appropriate lock concept requirements and constructed for the Mutex object.

A Mutex is NonCopyable.

For a Mutex type M and an object m of that type, the following expressions must be well-formed and have the indicated effects.

Table 12.1. Mutex Expressions

Expression Effects
M m;

Constructs a mutex object m.

Postcondition: m is unlocked.

(&m)->~M(); Precondition: m is unlocked. Destroys a mutex object m.
M::scoped_lock A model of ScopedLock
TryMutex Concept

A TryMutex is a refinement of Mutex. For a TryMutex type M and an object m of that type, the following expressions must be well-formed and have the indicated effects.

Table 12.2. TryMutex Expressions

Expression Effects
M::scoped_try_lock A model of ScopedTryLock
TimedMutex Concept

A TimedMutex is a refinement of TryMutex. For a TimedMutex type M and an object m of that type, the following expressions must be well-formed and have the indicated effects.

Table 12.3. TimedMutex Expressions

Expression Effects
M::scoped_timed_lock A model of ScopedTimedLock
Mutex Models

Boost.Threads currently supplies six models of Mutex and its refinements.

Lock Concepts

A lock object provides a safe means for locking and unlocking a mutex object (an object whose type is a model of Mutex or one of its refinements). In other words they are an implementation of the Scoped Locking[SchmidtStalRohnertBuschmann] pattern. The ScopedLock, ScopedTryLock, and ScopedTimedLock concepts formalize the requirements.

Lock objects are constructed with a reference to a mutex object and typically acquire ownership of the mutex object by setting its state to locked. They also ensure ownership is relinquished in the destructor. Lock objects also expose functions to query the lock status and to manually lock and unlock the mutex object.

Lock objects are meant to be short lived, expected to be used at block scope only. The lock objects are not thread-safe. Lock objects must maintain state to indicate whether or not they've been locked and this state is not protected by any synchronization concepts. For this reason a lock object should never be shared between multiple threads.

Lock Concept

For a Lock type L and an object lk and const object clk of that type, the following expressions must be well-formed and have the indicated effects.

Table 12.5. Lock Expressions

Expression Effects
(&lk)->~L(); if (locked()) unlock();
(&clk)->operator const void*() Returns type void*, non-zero if the associated mutex object has been locked by clk, otherwise 0.
clk.locked() Returns a bool, (&clk)->operator const void*() != 0
lk.lock()

Throws boost::lock_error if locked().

If the associated mutex object is already locked by some other thread, places the current thread in the Blocked state until the associated mutex is unlocked, after which the current thread is placed in the Ready state, eventually to be returned to the Running state. If the associated mutex object is already locked by the same thread the behavior is dependent on the locking strategy of the associated mutex object.

Postcondition: locked() == true

lk.unlock()

Throws boost::lock_error if !locked().

Unlocks the associated mutex.

Postcondition: !locked()

ScopedLock Concept

A ScopedLock is a refinement of Lock. For a ScopedLock type L and an object lk of that type, and an object m of a type meeting the Mutex requirements, and an object b of type bool, the following expressions must be well-formed and have the indicated effects.

Table 12.6. ScopedLock Expressions

Expression Effects
L lk(m); Constructs an object lk, and associates mutex object m with it, then calls lock()
L lk(m,b); Constructs an object lk, and associates mutex object m with it, then if b, calls lock()
TryLock Concept

A TryLock is a refinement of Lock. For a TryLock type L and an object lk of that type, the following expressions must be well-formed and have the indicated effects.

Table 12.7. TryLock Expressions

Expression Effects
lk.try_lock()

Throws boost::lock_error if locked().

Makes a non-blocking attempt to lock the associated mutex object, returning true if the lock attempt is successful, otherwise false. If the associated mutex object is already locked by the same thread the behavior is dependent on the locking strategy of the associated mutex object.

ScopedTryLock Concept

A ScopedTryLock is a refinement of TryLock. For a ScopedTryLock type L and an object lk of that type, and an object m of a type meeting the TryMutex requirements, and an object b of type bool, the following expressions must be well-formed and have the indicated effects.

Table 12.8. ScopedTryLock Expressions

Expression Effects
L lk(m); Constructs an object lk, and associates mutex object m with it, then calls try_lock()
L lk(m,b); Constructs an object lk, and associates mutex object m with it, then if b, calls lock()
TimedLock Concept

A TimedLock is a refinement of TryLock. For a TimedLock type L and an object lk of that type, and an object t of type boost::xtime, the following expressions must be well-formed and have the indicated effects.

Table 12.9. TimedLock Expressions

Expression Effects
lk.timed_lock(t)

Throws boost::lock_error if locked().

Makes a blocking attempt to lock the associated mutex object, and returns true if successful within the specified time t, otherwise false. If the associated mutex object is already locked by the same thread the behavior is dependent on the locking strategy of the associated mutex object.

ScopedTimedLock Concept

A ScopedTimedLock is a refinement of TimedLock. For a ScopedTimedLock type L and an object lk of that type, and an object m of a type meeting the TimedMutex requirements, and an object b of type bool, and an object t of type boost::xtime, the following expressions must be well-formed and have the indicated effects.

Table 12.10. ScopedTimedLock Expressions

Expression Effects
L lk(m,t); Constructs an object lk, and associates mutex object m with it, then calls timed_lock(t)
L lk(m,b); Constructs an object lk, and associates mutex object m with it, then if b, calls lock()
Lock Models

Boost.Threads currently supplies twelve models of Lock and its refinements.

Table 12.11. Lock Models

Concept Refines Models
Lock    
ScopedLock Lock

boost::mutex::scoped_lock

boost::recursive_mutex::scoped_lock

boost::try_mutex::scoped_lock

boost::recursive_try_mutex::scoped_lock

boost::timed_mutex::scoped_lock

boost::recursive_timed_mutex::scoped_lock

TryLock Lock  
ScopedTryLock TryLock

boost::try_mutex::scoped_try_lock

boost::recursive_try_mutex::scoped_try_lock

boost::timed_mutex::scoped_try_lock

boost::recursive_timed_mutex::scoped_try_lock

TimedLock TryLock  
ScopedTimedLock TimedLock

boost::timed_mutex::scoped_timed_lock

boost::recursive_timed_mutex::scoped_timed_lock

Last revised: October 16, 2005 at 14:37:34 GMT

Copyright © 2001-2003 William E. Kempf

PrevUpHomeNext