Copyright (c) 2007, 2009 The NetBSD Foundation, Inc.
All rights reserved.
This code is derived from software contributed to The NetBSD Foundation
by Andrew Doran.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
1. Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
.Dd December 8, 2017 .Dt MUTEX 9 .Os .Sh NAME .Nm mutex , .Nm mutex_init , .Nm mutex_destroy , .Nm mutex_enter , .Nm mutex_exit , .Nm mutex_ownable , .Nm mutex_owned , .Nm mutex_spin_enter , .Nm mutex_spin_exit , .Nm mutex_tryenter .Nd mutual exclusion primitives .Sh SYNOPSIS n sys/mutex.h .Ft void .Fn mutex_init "kmutex_t *mtx" "kmutex_type_t type" "int ipl" .Ft void .Fn mutex_destroy "kmutex_t *mtx" .Ft void .Fn mutex_enter "kmutex_t *mtx" .Ft void .Fn mutex_exit "kmutex_t *mtx" .Ft int .Fn mutex_ownable "kmutex_t *mtx" .Ft int .Fn mutex_owned "kmutex_t *mtx" .Ft void .Fn mutex_spin_enter "kmutex_t *mtx" .Ft void .Fn mutex_spin_exit "kmutex_t *mtx" .Ft int .Fn mutex_tryenter "kmutex_t *mtx"
p .Cd "options DIAGNOSTIC" .Cd "options LOCKDEBUG" .Sh DESCRIPTION Mutexes are used in the kernel to implement mutual exclusion among LWPs (lightweight processes) and interrupt handlers.
p The .Vt kmutex_t type provides storage for the mutex object. This should be treated as an opaque object and not examined directly by consumers.
p Mutexes replace the .Xr spl 9 system traditionally used to provide synchronization between interrupt handlers and LWPs. .Sh OPTIONS l -tag -width abcd t Cd "options DIAGNOSTIC"
p Kernels compiled with the .Dv DIAGNOSTIC option perform basic sanity checks on mutex operations. t Cd "options LOCKDEBUG"
p Kernels compiled with the .Dv LOCKDEBUG option perform potentially CPU intensive sanity checks on mutex operations. .El .Sh FUNCTIONS l -tag -width abcd t Fn mutex_init "mtx" "type" "ipl"
p Dynamically initialize a mutex for use.
p No other operations can be performed on a mutex until it has been initialized. Once initialized, all types of mutex are manipulated using the same interface. Note that .Fn mutex_init may block in order to allocate memory.
p The .Fa type argument must be given as .Dv MUTEX_DEFAULT . Other constants are defined but are for low-level system use and are not an endorsed, stable part of the interface.
p The type of mutex returned depends on the .Fa ipl argument: l -tag -width abcd t IPL_NONE, or one of the IPL_SOFT* constants
p An adaptive mutex will be returned. Adaptive mutexes provide mutual exclusion between LWPs, and between LWPs and soft interrupt handlers.
p Adaptive mutexes cannot be acquired from a hardware interrupt handler. An LWP may either sleep or busy-wait when attempting to acquire an adaptive mutex that is already held. t IPL_VM, IPL_SCHED, IPL_HIGH
p A spin mutex will be returned. Spin mutexes provide mutual exclusion between LWPs, and between LWPs and interrupt handlers.
p The .Fa ipl argument is used to pass a system interrupt priority level (IPL) that will block all interrupt handlers that may try to acquire the mutex.
p LWPs that own spin mutexes may not sleep, and therefore must not try to acquire adaptive mutexes or other sleep locks.
p A processor will always busy-wait when attempting to acquire a spin mutex that is already held. .El
p See .Xr spl 9 for further information on interrupt priority levels (IPLs). t Fn mutex_destroy "mtx"
p Release resources used by a mutex. The mutex may not be used after it has been destroyed. .Fn mutex_destroy may block in order to free memory. t Fn mutex_enter "mtx"
p Acquire a mutex. If the mutex is already held, the caller will block and not return until the mutex is acquired.
p All loads and stores after .Fn mutex_enter will not be reordered before it or served from a prior cache, and hence will .Em happen after any prior .Fn mutex_exit to release the mutex even on another CPU or in an interrupt. Thus, there is a global total ordering on all loads and stores under the same mutex.
p Mutexes and other types of locks must always be acquired in a consistent order with respect to each other. Otherwise, the potential for system deadlock exists.
p Adaptive mutexes and other types of lock that can sleep may not be acquired while a spin mutex is held by the caller.
p When acquiring a spin mutex, the IPL of the current CPU will be raised to the level set in .Fn mutex_init if it is not already equal or higher. t Fn mutex_exit "mtx"
p Release a mutex. The mutex must have been previously acquired by the caller. Mutexes may be released out of order as needed.
p All loads and stores before .Fn mutex_exit will not be reordered after it or delayed in a write buffer, and hence will .Fn happen before any subsequent .Fn mutex_enter to acquire the mutex even on another CPU or in an interrupt. Thus, there is a global total ordering on all loads and stores under the same mutex. t Fn mutex_ownable "mtx"
p When compiled with .Dv LOCKDEBUG (see .Xr options 4 ) , ensure that the current process can successfully acquire .Ar mtx . If .Ar mtx is already owned by the current process, the system will panic with a "locking against myself" error.
p This function is needed because .Fn mutex_owned does not differentiate if a spin mutex is owned by the current process vs owned by another process. .Fn mutex_ownable is reasonably heavy-weight, and should only be used with .Xr KDASSERT 9 . t Fn mutex_owned "mtx"
p For adaptive mutexes, return non-zero if the current LWP holds the mutex. For spin mutexes, return non-zero if the mutex is held, potentially by the current processor. Otherwise, return zero.
p .Fn mutex_owned is provided for making diagnostic checks to verify that a lock is held. For example: d -literal KASSERT(mutex_owned(&driver_lock)); .Ed
p It should not be used to make locking decisions at run time. For spin mutexes, it must not be used to verify that a lock is not held. t Fn mutex_spin_enter "mtx"
p Equivalent to .Fn mutex_enter , but may only be used when it is known that .Ar mtx is a spin mutex. Implies the same memory ordering as .Fn mutex_enter . On some architectures, this can substantially reduce the cost of acquiring a spin mutex. t Fn mutex_spin_exit "mtx"
p Equivalent to .Fn mutex_exit , but may only be used when it is known that .Ar mtx is a spin mutex. Implies the same memory ordering as .Fn mutex_exit . On some architectures, this can substantially reduce the cost of releasing a spin mutex. t Fn mutex_tryenter "mtx"
p Try to acquire a mutex, but do not block if the mutex is already held. Returns non-zero if the mutex was acquired, or zero if the mutex was already held.
p .Fn mutex_tryenter can be used as an optimization when acquiring locks in the wrong order. For example, in a setting where the convention is that .Dv first_lock must be acquired before .Dv second_lock , the following can be used to optimistically lock in reverse order: d -literal /* We hold second_lock, but not first_lock. */ KASSERT(mutex_owned(&second_lock)); if (!mutex_tryenter(&first_lock)) { /* Failed to get it - lock in the correct order. */ mutex_exit(&second_lock); mutex_enter(&first_lock); mutex_enter(&second_lock); /* * We may need to recheck any conditions the code * path depends on, as we released second_lock * briefly. */ } .Ed .El .Sh CODE REFERENCES The core of the mutex implementation is in
a sys/kern/kern_mutex.c .
p The header file
a sys/sys/mutex.h describes the public interface, and interfaces that machine-dependent code must provide to support mutexes. .Sh SEE ALSO .Xr atomic_ops 3 , .Xr membar_ops 3 , .Xr lockstat 8 , .Xr condvar 9 , .Xr kpreempt 9 , .Xr rwlock 9 , .Xr spl 9
p .Rs .%A Jim Mauro .%A Richard McDougall .%T Solaris Internals: Core Kernel Architecture .%I Prentice Hall .%D 2001 .%O ISBN 0-13-022496-0 .Re .Sh HISTORY The mutex primitives first appeared in .Nx 5.0 .
Indexes created Thu Oct 02 14:10:14 GMT 2025