Hallo, dies ist ein Test.

PWD: /www/data-lst1/unixsoft/unixsoft/kaempfer/.public_html

Running in File Mode
Relative path: ./../../.././../../../usr/man/man9f/mutex.9f
Real path: /usr/share/man/man9f/mutex.9f
Zurück

'\" te .\" Copyright (c) 2008, 2020, Oracle and/or its affiliates. .TH mutex 9F "9 Nov 2010" "Oracle Solaris 11.4" "Kernel Functions" .SH NAME mutex, mutex_enter, mutex_exit, mutex_init, mutex_destroy, mutex_owned, mutex_tryenter \- mutual exclusion lock routines .SH SYNOPSIS .LP .nf #include <sys/ksynch.h> \fBvoid\fR \fBmutex_init\fR(\fBkmutex_t *\fR\fImp\fR, \fBchar *\fR\fIname\fR, \fBkmutex_type_t\fR \fItype\fR, \fBvoid *\fR\fIarg\fR); .fi .LP .nf \fBvoid\fR \fBmutex_destroy\fR(\fBkmutex_t *\fR\fImp\fR); .fi .LP .nf \fBvoid\fR \fBmutex_enter\fR(\fBkmutex_t *\fR\fImp\fR); .fi .LP .nf \fBvoid\fR \fBmutex_exit\fR(\fBkmutex_t *\fR\fImp\fR); .fi .LP .nf \fBint\fR \fBmutex_owned\fR(\fBkmutex_t *\fR\fImp\fR); .fi .LP .nf \fBint\fR \fBmutex_tryenter\fR(\fBkmutex_t *\fR\fImp\fR); .fi .SH INTERFACE LEVEL .sp .LP Solaris DDI specific (Solaris DDI). .SH PARAMETERS .sp .ne 2 .mk .na \fB\fImp\fR\fR .ad .RS 8n .rt Pointer to a kernel mutex lock (\fBkmutex_t\fR). .RE .sp .ne 2 .mk .na \fB\fIname\fR\fR .ad .RS 8n .rt Descriptive string. This is obsolete and should be \fINULL\fR. (Non-\fINULL\fR strings are legal, but they are a waste of kernel memory.) .RE .sp .ne 2 .mk .na \fB\fItype\fR\fR .ad .RS 8n .rt Type of mutex lock. .RE .sp .ne 2 .mk .na \fB\fIarg\fR\fR .ad .RS 8n .rt Type-specific argument for initialization routine. .RE .SH DESCRIPTION .sp .LP A mutex enforces a policy of mutual exclusion. Only one thread at a time may hold a particular mutex. Threads trying to lock a held mutex will block until the mutex is unlocked. .sp .LP Mutexes are strictly bracketing and may not be recursively locked, meaning that mutexes should be exited in the opposite order they were entered, and cannot be reentered before exiting. .sp .LP \fBmutex_init()\fR initializes a mutex. It is an error to initialize a mutex more than once. .sp .LP The \fItype\fR argument specifies the type of mutex lock. The basic mutex type is \fBMUTEX_ADAPTIVE\fR, which is expected to be used in almost all of the kernel. \fBMUTEX_SPIN\fR provides interrupt blocking and must be used in interrupt handlers above \fBLOCK_LEVEL\fR. The \fIiblock\fR cookie argument to \fBmutex_init()\fR encodes the interrupt level to block. The \fIiblock\fR cookie must be \fINULL\fR for adaptive locks. .sp .LP \fBMUTEX_DEFAULT\fR is the type usually specified (except in drivers) to \fBmutex_init()\fR. It is identical to \fBMUTEX_ADAPTIVE\fR. .sp .LP \fBMUTEX_DRIVER\fR is always used by drivers. \fBmutex_init()\fR converts this to either \fBMUTEX_ADAPTIVE\fR or \fBMUTEX_SPIN\fR, depending on the \fIiblock\fR cookie. .sp .LP The \fIarg\fR argument provides type-specific information for a given variant type of mutex. When \fBmutex_init()\fR is called for driver mutexes, if the mutex is used by the interrupt handler, the \fIarg\fR should be the interrupt priority returned from \fBddi_intr_get_pri\fR(9F) or \fBddi_intr_get_softint_pri\fR(9F). Note that \fIarg\fR should be the value of the interrupt priority cast by calling the \fBDDI_INTR_PRI\fR macro. If the mutex is never used inside an interrupt handler, the argument should be \fINULL\fR. .sp .LP \fBmutex_enter()\fR is used to acquire a mutex. If the mutex is already held, then the caller blocks. After returning, the calling thread is the owner of the mutex. If the mutex is already held by the calling thread, a panic ensues. .sp .LP \fBmutex_owned()\fR should only be used in \fBASSERT()\fR and may be enforced by not being defined unless the preprocessor symbol \fBDEBUG\fR is defined. Its return value is non-zero if the current thread (or, if that cannot be determined, at least some thread) holds the mutex pointed to by \fImp\fR. .sp .LP \fBmutex_tryenter()\fR is very similar to \fBmutex_enter()\fR except that it doesn't block when the mutex is already held. \fBmutex_tryenter()\fR returns non-zero when it acquired the mutex and 0 when the mutex is already held. .sp .LP \fBmutex_exit()\fR releases a mutex and will unblock another thread if any are blocked on the mutex. .sp .LP \fBmutex_destroy()\fR releases any resources that might have been allocated by \fBmutex_init()\fR. \fBmutex_destroy()\fR must be called before freeing the memory containing the mutex with the mutex unheld (not owned by any thread). The caller must be sure that no thread attempts to use the mutex. It is an error to reference mutex after it was destroyed except re-initializing it using \fBmutex_init()\fR. .SH RETURN VALUES .sp .LP \fBmutex_tryenter()\fR returns a non-zero value on success and zero on failure. .sp .LP \fBmutex_owned()\fR returns a non-zero value if the calling thread currently holds the mutex pointed to by \fImp\fR, or when that cannot be determined, if any thread holds the mutex. Otherwise \fBmutex_owned()\fR returns zero. .SH CONTEXT .sp .LP These functions can be called from user, kernel, or high-level interrupt context, except for \fBmutex_init()\fR and \fBmutex_destroy()\fR, which can be called from user or kernel context only. .SH EXAMPLES .LP \fBExample 1\fR Initializing a Mutex .sp .LP A driver might do this to initialize a mutex that is part of its unit structure and used in its interrupt routine: .sp .in +2 .nf ddi_intr_get_pri(hdlp, &pri); mutex_init(&un->un_lock, NULL, MUTEX_DRIVER, DDI_INTR_PRI(pri)); ddi_intr_add_handler(hdlp, xxintr, (caddr_t)un, NULL); .fi .in -2 .sp .LP \fBExample 2\fR Calling a Routine with a Lock .sp .LP A routine that expects to be called with a certain lock held might have the following ASSERT: .sp .in +2 .nf xxstart(struct xxunit *un) { ASSERT(mutex_owned(&un->un_lock)); \&... .fi .in -2 .sp .SH SEE ALSO .sp .LP \fBlockstat\fR(4D), \fBlockstat\fR(8), \fBIntro\fR(9F), \fBcondvar\fR(9F), \fBddi_intr_add_handler\fR(9F), \fBddi_intr_alloc\fR(9F), \fBddi_intr_get_pri\fR(9F), \fBddi_intr_get_softint_pri\fR(9F), \fBrwlock\fR(9F), \fBsemaphore\fR(9F) .sp .LP \fIWriting Device Drivers in Oracle Solaris 11.4\fR .SH NOTES .sp .LP Compiling with \fB_LOCKTEST\fR or \fB_MPSTATS\fR defined has no effect. To gather lock statistics, see \fBlockstat\fR(8). Mutex statistics can be gathered on the fly, without rebooting or recompiling the kernel, via the \fBlockstat\fR(4D) driver. .sp .LP The address of a \fBkmutex_t\fR lock must be aligned on an 8-byte boundary for 64-bit kernels, or a 4-byte boundary for 32-bit kernels. Violation of this requirement will result in undefined behavior, including, but not limited to, failure of mutual exclusion or a system panic. .sp .LP To write scalable, responsive drivers that do not hang, panic or deadlock the system, follow these guidelines: .br .in +2 Never return from a driver entry point with a mutex held. .in -2 .br .in +2 Never hold a mutex when calling a service that may block, for example \fBkmem_alloc\fR(9F) with KM_SLEEP or \fBdelay\fR(9F). .in -2 .br .in +2 Always acquire mutexes in a consistent order. If a critical section acquires mutex A followed by B, and elsewhere in the driver mutex B is acquired before A, the driver can deadlock with one thread holding A and waiting for B and another thread holding B while waiting for A. .in -2 .br .in +2 Always use a mutex to enforce exclusive access to data, not instruction paths. .in -2 .br .in +2 Acquiring a lock in user context that is also acquired in interrupt context means that, as long as that lock is held, the driver instance holding the lock is subject to all the rules and limitations of interrupt context. .in -2 .br .in +2 In most cases, a mutex can and should be acquired and released within the same function. .in -2 .br .in +2 Liberal use of debugging aids like ASSERT(mutex_owned(&mutex)) can help find callers of a function which should be holding a mutex but are not. This means you need to test your driver compiled with DEBUG. .in -2 .br .in +2 Do not use a mutex to set driver state. However, you should use a mutex to protect driver state data. .in -2 .br .in +2 Use per-instance and automatic data where possible to reduce the amount of shared data. Per-instance data can be protected by a per-instance lock to improve scalability and reduce contention with multiple hardware instances. .in -2 .br .in +2 Avoid global data and global mutexes whenever possible. .in -2 .br .in +2 .in -2