diff options
author | glebius <glebius@FreeBSD.org> | 2006-02-01 19:39:25 +0000 |
---|---|---|
committer | glebius <glebius@FreeBSD.org> | 2006-02-01 19:39:25 +0000 |
commit | 70f2cce3e4099fe4831681b01b584822c8aed8e8 (patch) | |
tree | 274cb075d73dd0b4021d615b2e23881125367dd8 | |
parent | 16c9ed73d6b803cc7c25cdeec1d82956eb5bc231 (diff) | |
download | FreeBSD-src-70f2cce3e4099fe4831681b01b584822c8aed8e8.zip FreeBSD-src-70f2cce3e4099fe4831681b01b584822c8aed8e8.tar.gz |
Document read/write locks.
Reviewed by: jhb, ru
-rw-r--r-- | share/man/man9/Makefile | 1 | ||||
-rw-r--r-- | share/man/man9/rwlock.9 | 200 |
2 files changed, 201 insertions, 0 deletions
diff --git a/share/man/man9/Makefile b/share/man/man9/Makefile index 3ecff81..ed0eed1 100644 --- a/share/man/man9/Makefile +++ b/share/man/man9/Makefile @@ -196,6 +196,7 @@ MAN= accept_filter.9 \ rtalloc.9 \ rtentry.9 \ runqueue.9 \ + rwlock.9 \ sbuf.9 \ scheduler.9 \ securelevel_gt.9 \ diff --git a/share/man/man9/rwlock.9 b/share/man/man9/rwlock.9 new file mode 100644 index 0000000..7dae9d8 --- /dev/null +++ b/share/man/man9/rwlock.9 @@ -0,0 +1,200 @@ +.\" Copyright (c) 2006 Gleb Smirnoff <glebius@FreeBSD.org> +.\" All rights reserved. +.\" +.\" 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 AUTHOR 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 AUTHOR 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. +.\" +.\" $FreeBSD$ +.\" +.Dd January 30, 2006 +.Dt RWLOCK 9 +.Os +.Sh NAME +.Nm rwlock +.Nm rw_init , +.Nm rw_rlock , +.Nm rw_wlock , +.Nm rw_assert , +.Nm rw_runlock , +.Nm rw_wunlock , +.Nm rw_initialized , +.Nm rw_destroy , +.Nm RW_SYSINIT , +.Nd kernel synchronization primitives +.Sh SYNOPSIS +.In sys/param.h +.In sys/lock.h +.In sys/rwlock.h +.Ft void +.Fn rw_init "struct rwlock *rwlock" "const char *name" +.Ft void +.Fn rw_rlock "struct rwlock *rwlock" +.Ft void +.Fn rw_wlock "struct rwlock *rwlock" +.Ft void +.Fn rw_runlock "struct rwlock *rwlock" +.Ft void +.Fn rw_wunlock "struct rwlock *rwlock" +.Ft int +.Fn rw_initialized "struct rwlock *rwlock" +.Ft void +.Fn rw_destroy "struct rwlock *rwlock" +.Pp +.Cd "options INVARIANTS" +.Cd "options INVARIANT_SUPPORT" +.Ft void +.Fn rw_assert "struct rwlock *rwlock" "int what" +.In sys/kernel.h +.Fn RW_SYSINIT "name" "struct rwlock *rwlock" "const char *description" +.Sh DESCRIPTION +Read/write locks are a method of thread synchronization, which +allows several threads have shared access to protected data, or +one thread have exclusive access. +The threads that share access are known as +.Em readers +since they should only read the protected data, while thread +with exclusive access is known as a +.Em writer +since it can modify protected data. +.Pp +Although the description of read/write locks looks very similar +to description of +.Xr sx 9 +locks, their usage pattern is different. +The read/write locks can be treated as mutexes (see +.Xr mutex 9 ) +with shared/exclusive semantics. +Unlike +.Xr sx 9 , +an +.Nm +can be locked while holding a non-spin mutex; +.Nm +cannot be held while sleeping. +The +.Nm +locks have priority propagation like mutexes, but priority +can be propagated only to exclusive holder. +This limitation comes from the fact that shared owners +are anonymous. +Another important property is that shared holder of +.Nm +can recurse on it. +.Ss Macros and Functions +.Bl -tag -width indent +.It Fn rw_init "struct rwlock *rwlock" "const char *name" +Initialize structure located at +.Fa rwlock +as read/write lock, described by name +.Fa name . +The description is used solely for debugging purposes. +This function must be used before any other manipulations +with the lock. +.It Fn rw_rlock "struct rwlock *rwlock" +Lock the +.Fa rwlock +as reader. +If any thread holds this lock exclusively, the current thread blocks, +and its priority is propagated to exclusive holder. +The +.Fn rw_rlock +function can be called when the thread has already acquired reader +access on +.Fa rwlock . +This is called +.Dq "recursing on a lock" . +.It Fn rw_wlock "struct rwlock *rwlock" +Lock the +.Fa rwlock +as writer. +If there are any shared owners of the lock, the current thread blocks. +The +.Fn rw_wlock +function cannot be called recursively. +.It Fn rw_runlock "struct rwlock *rwlock" +This function releases shared lock, previously acquired by +.Fn rw_rlock . +.It Fn rw_wunlock "struct rwlock *rwlock" +This function releases exclusive lock, previously acquired by +.Fn rw_wlock . +.It Fn rw_initialized "struct rwlock *rwlock" +This function returns non-zero if the +.Fa rwlock +has been initialized, and zero otherwise. +.It Fn rw_destroy "struct rwlock *rwlock" +This functions destroys a lock previously initialized with +.Fn rw_init . +The +.Fa rwlock +must be unlocked. +.It Fn rw_assert "struct rwlock *rwlock" "int what" +This function allows assertions specified in +.Fa what +to be made about +.Fa rwlock . +If the assertions are not true and the kernel is compiled +with +.Cd "options INVARIANTS" +and +.Cd "options INVARIANT_SUPPORT" , +the kernel will panic. +Currently the following assertions are supported: +.Bl -tag -width ".Dv RA_UNLOCKED" +.It Dv RA_LOCKED +Assert that current thread is either shared or exclusive owner +of the +.Nm +pointed to by the first argument. +.It Dv RA_RLOCKED +Assert that current thread is shared owner of the +.Nm +pointed +to by the first argument. +.It Dv RA_WLOCKED +Assert that current thread is exclusive owner of the +.Nm +pointed +to by the first argument. +.It Dv RA_UNLOCKED +Assert that current thread is neither shared nor exclusive owner +of the +.Nm +pointed to by the first argument. +.El +.El +.Sh SEE ALSO +.Xr condvar 9 , +.Xr mutex 9 , +.Xr panic 9 , +.Xr sema 9 , +.Xr sx 9 +.Sh HISTORY +These +functions appeared in +.Fx 7.0 . +.Sh AUTHORS +.An -nosplit +The +.Nm +facility was written by +.An "John Baldwin" . +This manual page was written by +.An "Gleb Smirnoff" . |