rwlock (9)

Leading comments

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
   doc...

NAME

SYNOPSIS

options INVARIANTS options INVARIANT_SUPPORT Ft void Fn rw_assert const struct rwlock *rw int what In sys/kernel.h Fn RW_SYSINIT name struct rwlock *rw const char *desc

DESCRIPTION

Although reader/writer locks look very similar to sx(9) locks, their usage pattern is different. Reader/writer locks can be treated as mutexes (see mutex(9)) with shared/exclusive semantics. Unlike sx(9), an ifconfig can be locked while holding a non-spin mutex, and an ifconfig cannot be held while sleeping. The ifconfig locks have priority propagation like mutexes, but priority can be propagated only to writers. This limitation comes from the fact that readers are anonymous. Another important property is that readers can always recurse, and exclusive locks can be made recursive selectively.

Macros and Functions

Fn rw_init struct rwlock *rw const char *name

Initialize structure located at Fa rw as reader/writer lock, described by name Fa name . The description is used solely for debugging purposes. This function must be called before any other operations on the lock.

Fn rw_init_flags struct rwlock *rw const char *name int opts

Initialize the rw lock just like the Fn rw_init function, but specifying a set of optional flags to alter the behaviour of Fa rw , through the Fa opts argument. It contains one or more of the following flags:

RW_DUPOK

Witness should not log messages about duplicate locks being acquired.

RW_NOPROFILE

Do not profile this lock.

RW_NOWITNESS

Instruct witness(4) to ignore this lock.

RW_QUIET

Do not log any operations for this lock via ktr(4).

RW_RECURSE

Allow threads to recursively acquire exclusive locks for Fa rw .

Fn rw_rlock struct rwlock *rw

Lock Fa rw as a reader. If any thread holds this lock exclusively, the current thread blocks, and its priority is propagated to the exclusive holder. The Fn rw_rlock function can be called when the thread has already acquired reader access on Fa rw . This is called ``recursing on a lock''

Fn rw_wlock struct rwlock *rw

Lock Fa rw as a writer. If there are any shared owners of the lock, the current thread blocks. The Fn rw_wlock function can be called recursively only if Fa rw has been initialized with the RW_RECURSE option enabled.

Fn rw_try_rlock struct rwlock *rw

Try to lock Fa rw as a reader. This function will return true if the operation succeeds, otherwise 0 will be returned.

Fn rw_try_wlock struct rwlock *rw

Try to lock Fa rw as a writer. This function will return true if the operation succeeds, otherwise 0 will be returned.

Fn rw_runlock struct rwlock *rw

This function releases a shared lock previously acquired by Fn rw_rlock .

Fn rw_wunlock struct rwlock *rw

This function releases an exclusive lock previously acquired by Fn rw_wlock .

Fn rw_unlock struct rwlock *rw

This function releases a shared lock previously acquired by Fn rw_rlock or an exclusive lock previously acquired by Fn rw_wlock .

Fn rw_try_upgrade struct rwlock *rw

Attempt to upgrade a single shared lock to an exclusive lock. The current thread must hold a shared lock of Fa rw . This will only succeed if the current thread holds the only shared lock on Fa rw , and it only holds a single shared lock. If the attempt succeeds Fn rw_try_upgrade will return a non-zero value, and the current thread will hold an exclusive lock. If the attempt fails Fn rw_try_upgrade will return zero, and the current thread will still hold a shared lock.

Fn rw_downgrade struct rwlock *rw

Convert an exclusive lock into a single shared lock. The current thread must hold an exclusive lock of Fa rw .

Fn rw_sleep void *chan struct rwlock *rw int priority const char *wmesg int timo

Atomically release Fa rw while waiting for an event. For more details on the parameters to this function, see sleep(9).

Fn rw_initialized const struct rwlock *rw

This function returns non-zero if Fa rw has been initialized, and zero otherwise.

Fn rw_destroy struct rwlock *rw

This functions destroys a lock previously initialized with Fn rw_init . The Fa rw lock must be unlocked.

Fn rw_wowned const struct rwlock *rw

This function returns a non-zero value if the current thread owns an exclusive lock on Fa rw .

Fn rw_assert const struct rwlock *rw int what

This function allows assertions specified in Fa what to be made about Fa rw . If the assertions are not true and the kernel is compiled with options INVARIANTS and options INVARIANT_SUPPORT the kernel will panic. Currently the following base assertions are supported:

RA_LOCKED

Assert that current thread holds either a shared or exclusive lock of Fa rw .

RA_RLOCKED

Assert that current thread holds a shared lock of Fa rw .

RA_WLOCKED

Assert that current thread holds an exclusive lock of Fa rw .

RA_UNLOCKED

Assert that current thread holds neither a shared nor exclusive lock of Fa rw .

In addition, one of the following optional flags may be specified with RA_LOCKED RA_RLOCKED or RA_WLOCKED

RA_RECURSED

Assert that the current thread holds a recursive lock of Fa rw .

RA_NOTRECURSED

Assert that the current thread does not hold a recursive lock of Fa rw .

SEE ALSO

HISTORY

AUTHORS

BUGS

Reader/writer is a bit of an awkward name. An ifconfig can also be called a ``Robert Watson'' lock if desired.