2 .\" Copyright (c) 1998 Berkeley Software Design, Inc. All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
12 .\" 3. Berkeley Software Design Inc's name may not be used to endorse or
13 .\" promote products derived from this software without specific prior
14 .\" written permission.
16 .\" THIS SOFTWARE IS PROVIDED BY BERKELEY SOFTWARE DESIGN INC ``AS IS'' AND
17 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 .\" ARE DISCLAIMED. IN NO EVENT SHALL BERKELEY SOFTWARE DESIGN INC BE LIABLE
20 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
28 .\" from BSDI $Id: mutex.4,v 1.1.2.3 1998/04/27 22:53:13 ewv Exp $
41 .Nm mtx_lock_spin_flags ,
43 .Nm mtx_trylock_flags ,
46 .Nm mtx_unlock_flags ,
47 .Nm mtx_unlock_spin_flags ,
54 .Nd kernel synchronization primitives
60 .Fn mtx_init "struct mtx *mutex" "const char *name" "const char *type" "int opts"
62 .Fn mtx_destroy "struct mtx *mutex"
64 .Fn mtx_lock "struct mtx *mutex"
66 .Fn mtx_lock_spin "struct mtx *mutex"
68 .Fn mtx_lock_flags "struct mtx *mutex" "int flags"
70 .Fn mtx_lock_spin_flags "struct mtx *mutex" "int flags"
72 .Fn mtx_trylock "struct mtx *mutex"
74 .Fn mtx_trylock_flags "struct mtx *mutex" "int flags"
76 .Fn mtx_unlock "struct mtx *mutex"
78 .Fn mtx_unlock_spin "struct mtx *mutex"
80 .Fn mtx_unlock_flags "struct mtx *mutex" "int flags"
82 .Fn mtx_unlock_spin_flags "struct mtx *mutex" "int flags"
84 .Fn mtx_sleep "void *chan" "struct mtx *mtx" "int priority" "const char *wmesg" "int timo"
86 .Fn mtx_initialized "const struct mtx *mutex"
88 .Fn mtx_owned "const struct mtx *mutex"
90 .Fn mtx_recursed "const struct mtx *mutex"
92 .Cd "options INVARIANTS"
93 .Cd "options INVARIANT_SUPPORT"
95 .Fn mtx_assert "const struct mtx *mutex" "int what"
97 .Fn MTX_SYSINIT "name" "struct mtx *mtx" "const char *description" "int opts"
99 Mutexes are the most basic and primary method of thread synchronization.
100 The major design considerations for mutexes are:
103 Acquiring and releasing uncontested mutexes should be as cheap
106 They must have the information and storage space to support
107 priority propagation.
109 A thread must be able to recursively acquire a mutex,
110 provided that the mutex is initialized to support recursion.
113 There are currently two flavors of mutexes, those that context switch
114 when they block and those that do not.
118 mutexes will context switch when they are already held.
120 they may spin for some amount
121 of time before context switching.
122 It is important to remember that since a thread may be preempted at any time,
123 the possible context switch introduced by acquiring a mutex is guaranteed
124 to not break anything that is not already broken.
126 Mutexes which do not context switch are
129 These should only be used to protect data shared with primary interrupt
131 This includes interrupt filters and low level scheduling code.
132 In all architectures both acquiring and releasing of a
133 uncontested spin mutex is more expensive than the same operation
135 In order to protect an interrupt service routine from blocking
136 against itself all interrupts are either blocked or deferred on a processor
137 while holding a spin lock.
138 It is permissible to hold multiple spin mutexes.
140 Once a spin mutex has been acquired it is not permissible to acquire a
143 The storage needed to implement a mutex is provided by a
145 In general this should be treated as an opaque object and
146 referenced only with the mutex primitives.
150 function must be used to initialize a mutex
151 before it can be passed to any of the other mutex functions.
154 option is used to identify the lock in debugging output etc.
157 option is used by the witness code to classify a mutex when doing checks
164 is used in its place.
165 The pointer passed in as
169 is saved rather than the data it points to.
170 The data pointed to must remain stable
171 until the mutex is destroyed.
174 argument is used to set the type of mutex.
175 It may contain either
180 See below for additional initialization options.
181 It is not permissible to pass the same
185 multiple times without intervening calls to
192 mutual exclusion lock
193 on behalf of the currently running kernel thread.
194 If another kernel thread is holding the mutex,
195 the caller will be disconnected from the CPU
196 until the mutex is available
197 (i.e., it will block).
203 mutual exclusion lock
204 on behalf of the currently running kernel thread.
205 If another kernel thread is holding the mutex,
206 the caller will spin until the mutex becomes available.
207 Interrupts are disabled during the spin and remain disabled
208 following the acquiring of the lock.
210 It is possible for the same thread to recursively acquire a mutex
211 with no ill effects, provided that the
215 during the initialization of the mutex.
220 .Fn mtx_lock_spin_flags
225 lock, respectively, and also accept a
228 In both cases, the only flags presently available for lock acquires are
234 bit is turned on in the
238 tracing is being done,
239 it will be silenced during the lock acquire.
242 bit is turned on in the
244 argument, then the mutex can be acquired recursively.
248 attempts to acquire the
252 If the mutex cannot be immediately acquired
255 otherwise the mutex will be acquired
256 and a non-zero value will be returned.
259 .Fn mtx_trylock_flags
260 function has the same behavior as
262 but should be used when the caller desires to pass in a
265 Presently, the only valid value in the
269 and its effects are identical to those described for
277 mutual exclusion lock.
278 The current thread may be preempted if a higher priority thread is waiting
285 mutual exclusion lock.
290 .Fn mtx_unlock_spin_flags
291 functions behave in exactly the same way as do the standard mutex
292 unlock routines above, while also allowing a
294 argument which may specify
298 is identical to its behavior in the mutex lock routines.
302 function is used to destroy
304 so the data associated with it may be freed
305 or otherwise overwritten.
306 Any mutex which is destroyed
307 must previously have been initialized with
309 It is permissible to have a single hold count
310 on a mutex when it is destroyed.
311 It is not permissible to hold the mutex recursively,
312 or have another thread blocked on the mutex
313 when it is destroyed.
317 function is used to atomically release
319 while waiting for an event.
320 For more details on the parameters to this function,
326 function returns non-zero if
328 has been initialized and zero otherwise.
332 function returns non-zero
333 if the current thread holds
335 If the current thread does not hold
341 function returns non-zero if the
344 This check should only be made if the running thread already owns
349 function allows assertions specified in
353 If the assertions are not true and the kernel is compiled with
354 .Cd "options INVARIANTS"
356 .Cd "options INVARIANT_SUPPORT" ,
357 the kernel will panic.
358 Currently the following assertions are supported:
359 .Bl -tag -width MA_NOTRECURSED
361 Assert that the current thread
363 pointed to by the first argument.
365 Assert that the current thread
366 does not hold the mutex
367 pointed to by the first argument.
369 Assert that the current thread has recursed on the mutex
370 pointed to by the first argument.
371 This assertion is only valid in conjunction with
373 .It Dv MA_NOTRECURSED
374 Assert that the current thread has not recursed on the mutex
375 pointed to by the first argument.
376 This assertion is only valid in conjunction with
382 macro is used to generate a call to the
384 routine at system startup in order to initialize a given mutex lock.
385 The parameters are the same as
387 but with an additional argument,
389 that is used in generating unique variable names for the related structures associated with the lock and the sysinit routine.
390 .Ss The Default Mutex Type
391 Most kernel code should use the default lock type,
393 The default lock type will allow the thread
394 to be disconnected from the CPU
395 if the lock is already held by another thread.
397 may treat the lock as a short term spin lock
398 under some circumstances.
399 However, it is always safe to use these forms of locks
400 in an interrupt thread
401 without fear of deadlock
402 against an interrupted thread on the same CPU.
403 .Ss The Spin Mutex Type
406 mutex will not relinquish the CPU
407 when it cannot immediately get the requested lock,
408 but will loop, waiting for the mutex to be released by another CPU.
409 This could result in deadlock
410 if another thread interrupted the thread which held a mutex
411 and then tried to acquire the mutex.
412 For this reason spin locks disable all interrupts on the local CPU.
414 Spin locks are fairly specialized locks
415 that are intended to be held for very short periods of time.
416 Their primary purpose is to protect portions of the code
417 that implement other synchronization primitives such as default mutexes,
418 thread scheduling, and interrupt threads.
419 .Ss Initialization Options
420 The options passed in the
424 specify the mutex type.
429 options is required and only one of those two options may be specified.
430 The possibilities are:
431 .Bl -tag -width MTX_NOWITNESS
434 will always allow the current thread to be suspended
435 to avoid deadlock conditions against interrupt threads.
436 The implementation of this lock type
437 may spin for a while before suspending the current thread.
440 will never relinquish the CPU.
441 All interrupts are disabled on the local CPU
442 while any spin lock is held.
444 Specifies that the initialized mutex is allowed to recurse.
445 This bit must be present if the mutex is permitted to recurse.
447 Do not log any mutex operations for this lock.
453 Witness should not log messages about duplicate locks being acquired.
455 Do not profile this lock.
457 .Ss Lock and Unlock Flags
458 The flags passed to the
460 .Fn mtx_lock_spin_flags ,
461 .Fn mtx_unlock_flags ,
463 .Fn mtx_unlock_spin_flags
464 functions provide some basic options to the caller,
465 and are often used only under special circumstances to modify lock or
467 Standard locking and unlocking should be performed with the
474 Only if a flag is required should the corresponding
475 flags-accepting routines be used.
477 Options that modify mutex behavior:
478 .Bl -tag -width MTX_QUIET
480 This option is used to quiet logging messages during individual mutex
482 This can be used to trim superfluous logging messages for debugging purposes.
487 must be acquired, it must be acquired prior to acquiring
489 Put another way: it is impossible to acquire
491 non-recursively while
492 holding another mutex.
493 It is possible to acquire other mutexes while holding
495 and it is possible to acquire
497 recursively while holding other mutexes.
499 Sleeping while holding a mutex (except for
502 and should be avoided.
503 There are numerous assertions which will fail if this is attempted.
504 .Ss Functions Which Access Memory in Userspace
505 No mutexes should be held (except for
507 across functions which
508 access memory in userspace, such as
514 No locks are needed when calling these functions.
517 .Xr LOCK_PROFILING 9 ,
527 functions appeared in