From 88101146f2ec7d53ffb793e365f05097ffd35fd3 Mon Sep 17 00:00:00 2001
From: marha pthread_mutex_init, pthread_mutex_lock, pthread_mutex_trylock,
-pthread_mutex_timedlock, pthread_mutex_unlock, pthread_mutex_destroy
-- operations on mutexes
+pthread_mutex_timedlock, pthread_mutex_unlock,
+pthread_mutex_consistent, pthread_mutex_destroy - operations on
+mutexes
#include <pthread.h>
#include <time.h>POSIX Threads for Windows – REFERENCE - Pthreads-w32
+POSIX Threads for Windows – REFERENCE -
+Pthreads-w32
-Name
+Name
Synopsis
+Synopsis
int pthread_mutex_unlock(pthread_mutex_t *mutex);
+int pthread_mutex_consistent(pthread_mutex_t *mutex); +
int pthread_mutex_destroy(pthread_mutex_t *mutex);
-A mutex is a MUTual EXclusion device, and is useful for protecting shared data structures from concurrent modifications, and implementing critical sections and monitors. @@ -76,16 +93,22 @@ for more information on mutex attributes. statically, using the constants PTHREAD_MUTEX_INITIALIZER (for normal “fast” mutexes), PTHREAD_RECURSIVE_MUTEX_INITIALIZER (for recursive mutexes), and PTHREAD_ERRORCHECK_MUTEX_INITIALIZER -(for error checking mutexes). In -the Pthreads-w32 implementation, an application should still -call pthread_mutex_destroy at some point to ensure that any +(for error checking mutexes). In +the Pthreads-w32 implementation, +an application should still call pthread_mutex_destroy +at some point to ensure that any resources consumed by the mutex are released.
+Any mutex type can be +initialized as a robust mutex. +See pthread_mutexattr_init(3) +for more information as well as the +section Robust Mutexes +below.
pthread_mutex_lock locks the given mutex. If the mutex is currently unlocked, it becomes locked and owned by the calling thread, and pthread_mutex_lock returns immediately. If the mutex is already locked by another thread, pthread_mutex_lock -suspends the calling thread until the mutex is unlocked. -
+suspends the calling thread until the mutex is unlocked.If the mutex is already locked by the calling thread, the behavior of pthread_mutex_lock depends on the type of the mutex. If the mutex is of the “normal” type, the calling thread is suspended @@ -101,15 +124,18 @@ be performed before the mutex returns to the unlocked state.
pthread_mutex_trylock behaves identically to pthread_mutex_lock, except that it does not block the calling thread if the mutex is already locked by another thread (or by the -calling thread in the case of a “normal” mutex). Instead, -pthread_mutex_trylock returns immediately with the error code -EBUSY. +calling thread in the case of a “normal” or “errorcheck” +mutex). Instead, pthread_mutex_trylock returns immediately +with the error code EBUSY.
pthread_mutex_timedlock behaves identically to pthread_mutex_lock, except that if it cannot acquire the lock before the abs_timeout time, the call returns with the error code ETIMEDOUT. If the mutex can be locked immediately it is, and the abs_timeout parameter is ignored.
+pthread_mutex_consistent may only be called for +PTHREAD_MUTEX_ROBUST mutexes. It simply marks the mutex as +consistent. See Robust Mutexes below.
pthread_mutex_unlock unlocks the given mutex. The mutex is assumed to be locked and owned by the calling thread on entrance to pthread_mutex_unlock. If the mutex is of the “normal” @@ -117,19 +143,53 @@ type, pthread_mutex_unlock always returns it to the unlocked state. If it is of the ‘‘recursive’’ type, it decrements the locking count of the mutex (number of pthread_mutex_lock operations performed on it by the calling thread), and only when this -count reaches zero is the mutex actually unlocked. -
+count reaches zero is the mutex actually unlocked. In Pthreads-win32, +non-robust normal or default mutex types do not check the owner of +the mutex. For all types of robust mutexes the owner is checked and +an error code is returned if the calling thread does not own the +mutex.On ‘‘error checking’’ mutexes, pthread_mutex_unlock actually checks at run-time that the mutex is locked on entrance, and that it was locked by the same thread that is now calling pthread_mutex_unlock. If these conditions are not met, an error code is returned and the mutex remains unchanged. ‘‘Normal’’ -mutexes perform no such checks, thus allowing a locked mutex to be -unlocked by a thread other than its owner. This is non-portable -behavior and is not meant to be used as a feature.
+[non-robust] mutexes perform no such checks, thus allowing a locked +mutex to be unlocked by a thread other than its owner. This is +non-portable behavior and is not meant to be used as a feature.pthread_mutex_destroy destroys a mutex object, freeing the resources it might hold. The mutex must be unlocked on entrance.
-If the mutex is PTHREAD_MUTEX_ROBUST and the owning thread +terminates without unlocking the mutex the implementation will wake +one waiting thread, if any. The next thread to acquire the mutex will +receive the error code EOWNERDEAD, +in which case that thread should if possible ensure that the state +protected by the mutex is consistent and then call +pthread_mutex_consistent before +unlocking. The mutex may then be used normally from then on.
+If the thread cannot recover the +state then it must call pthread_mutex_unlock +without calling pthread_mutex_consistent. +This will mark the mutex as unusable and wake all currently waiting +threads with the return code ENOTRECOVERABLE. +The error indicates that the mutex is no longer usable and any +threads that receive this error code from any lock operation have not +acquired the mutex. The mutex can be made consistent by calling +pthread_mutex_destroy to +uninitialize the mutex, and calling pthread_mutex_int +to reinitialize the mutex. However, +the state that was protected by the mutex remains inconsistent and +some form of application recovery is required.
+If a thread that receives the +EOWNERDEAD error code +itself terminates without unlocking the mutex then this behaviour +repeats for the next acquiring thread.
+Applications must ensure that +they check the return values from all calls targeting robust mutexes.
+Robust mutexes are slower because they +require some additional overhead, however they are not very much +slower than the non-robust recursive type.
+None of the mutex functions is a cancellation point, not even pthread_mutex_lock, in spite of the fact that it can suspend a thread for arbitrary durations. This way, the status of mutexes at @@ -138,46 +198,56 @@ unlock precisely those mutexes that need to be unlocked before the thread stops executing. Consequently, threads using deferred cancellation should never hold a mutex for extended periods of time.
-The mutex functions are not async-signal safe. What this means is that they should not be called from a signal handler. In particular, calling pthread_mutex_lock or pthread_mutex_unlock from a signal handler may deadlock the calling thread.
-pthread_mutex_init always returns 0. The other mutex functions return 0 on success and a non-zero error code on error.
-The pthread_mutex_lock function returns the following error code on error:
-The pthread_mutex_trylock function returns the following error -codes on error: -
-The pthread_mutex_timedlock function returns the following @@ -185,15 +255,22 @@ error codes on error:
The pthread_mutex_unlock function returns the following error @@ -201,12 +278,12 @@ code on error:
Xavier Leroy <Xavier.Leroy@inria.fr>
Modified by Ross Johnson for use with Pthreads-w32.
-pthread_mutexattr_init(3) , pthread_mutexattr_settype(3) , pthread_cancel(3) .
-A shared global variable x can be protected by a mutex as follows:
-int x; -pthread_mutex_t mut = PTHREAD_MUTEX_INITIALIZER;
+int x; +pthread_mutex_t mut = PTHREAD_MUTEX_INITIALIZER;All accesses and modifications to x should be bracketed by calls to pthread_mutex_lock and pthread_mutex_unlock as follows:-pthread_mutex_lock(&mut); +-pthread_mutex_lock(&mut); /* operate on x */ pthread_mutex_unlock(&mut);
-Table of -Contents+Table +of Contents