Copyright (c) 2005 Pedro Martelletto <pedro@ambientworks.net>
All rights reserved.
Permission to use, copy, modify, and distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
.Dd $Mdocdate: April 25 2009 $ .Dt MUTEX 9 .Os .Sh NAME .Nm mutex , .Nm mtx_init , .Nm mtx_enter , .Nm mtx_leave .Nd interface to CPU mutexes .Sh SYNOPSIS .Fd #include <sys/mutex.h> .Ft void .Fn mtx_init "struct mutex *mtxp" "int wantipl" .Ft void .Fn mtx_enter "struct mutex *mtxp" .Ft void .Fn mtx_leave "struct mutex *mtxp" .Sh DESCRIPTION The .Nm set of functions provides a non-recursive, interrupt-aware spinning mechanism to ensure mutual exclusion between different CPUs.
p The .Fn mtx_init function is used to initiate the mutex pointed to by .Fa mtxp . When acquired, the mutex will cause the processor interrupt level to be raised to .Fa wantipl if necessary.
p The .Fn mtx_enter function acquires a mutex, spinning if necessary.
p The .Fn mtx_leave function releases a mutex. In case the acquisition of the mutex caused the interrupt level to be changed, it is then restored. .Sh SEE ALSO .Xr lockmgr 9 , .Xr rwlock 9 , .Xr spl 9 .Sh HISTORY The .Nm functions first appeared in .Ox 3.6 . .Sh AUTHORS The .Nm functions were written by .An Artur Grabowski .Aq art@openbsd.org . .Sh CAVEATS As these are spinning locks, don't sleep while holding one.
p If using a mutex in an interrupt handler, locks must be initialised with the correct IPL or deadlock will occur.
p Multiple mutexes may be nested, but not interleaved. This is okay: d -literal -offset indent mtx_enter(foo); mtx_enter(bar); mtx_leave(bar); mtx_leave(foo); .Ed
p While this is .Fa not : d -literal -offset indent mtx_enter(foo); mtx_enter(bar); mtx_leave(foo); mtx_leave(bar); .Ed