_spin: _spin_init, _spin_lock, _spin_trylock, _spin_unlock, _spin_destroy -- overview of spin lock routines


   cc [options] -Kthread file

#include <synch.h>

int _spin_init(spin_t *lock, void *arg);

void _spin_lock(spin_t *lock);

int _spin_trylock(spin_t *lock);

void _spin_unlock(spin_t *lock);

int _spin_destroy(spin_t *lock);


Spin locks, a type of mutual exclusion lock (mutex), are used to serialize the execution of threads. They are typically used to ensure that only one thread at a time accesses shared data. Note that these locks protect data only when the convention of acquiring the lock is faithfully followed before any access of the data.

The difference between spin locks and ordinary mutex locks is in their locking routines. When a mutex is already locked, the locking routine (mutex_lock(3synch)) will block the caller until the lock is available. When a ``spin'' lock is already locked, the locking routine (_spin_lock) will busy-wait, or ``spin'', in a loop, testing if the lock has become available. Such spinning wastes processor cycles and can slow processors doing useful work, including the processor holding the lock, by consuming communication bandwidth.

The Threads Library provides five routines to manipulate spin locks; these are outlined below and described in more detail on individual manual pages.


pointer to spin lock to be initialized, locked, unlocked, or destroyed

NULL (reserved for future use)


_spin_init initializes the spin lock pointed to by lock to the unlocked state.


_spin_lock locks the spin lock pointed to by lock. If lock is locked, the calling thread will busy-wait, or ``spin'', until lock is available. When _spin_lock returns, the caller has acquired lock.


_spin_trylock attempts once to lock the spin lock pointed to by lock.

If lock is available, _spin_trylock will return successfully with lock locked. If lock is already locked, _spin_trylock immediately returns EBUSY to the caller without acquiring lock or spinning.


_spin_unlock unlocks the spin lock pointed to by lock.

If one or more threads are waiting for the lock when _spin_unlock is called, one of the waiting threads will acquire the lock.


_spin_destroy destroys the spin lock pointed to by lock. This includes invalidating lock and freeing any associated implementation-allocated dynamic resources.


Because spin locks waste system resources, most applications should use mutexes instead of spin locks for mutual exclusion. However, spin locks are useful when:

Spin locks should only be used when there is a guarantee that the thread will not be preempted or blocked while holding a spin lock. It is the responsibility of each application to unlock all spin locks before calling sleep or blocking routines.


Spin locks must not be used on a single processor system. In the best case, a spin lock on a single processor system will waste resources, slowing down the owner of the lock; in the worst case, it will deadlock the processor.

Operations on spin locks are not recursive--a thread can deadlock if it attempts to relock a spin lock that it already has locked.


Intro(3synch), mutex(3synch), _spin_destroy(3synch), _spin_init(3synch), _spin_lock(3synch), _spin_trylock(3synch), _spin_unlock(3synch)
© 2004 The SCO Group, Inc. All rights reserved.
UnixWare 7 Release 7.1.4 - 25 April 2004