NetBSD/share/man/man9/sleep.9
2001-10-07 09:47:45 +00:00

214 lines
6.4 KiB
Groff

.\" $NetBSD: sleep.9,v 1.17 2001/10/07 09:47:45 jdolecek Exp $
.\"
.\" Copyright (c) 1996 The NetBSD Foundation, Inc.
.\" All rights reserved.
.\"
.\" This code is derived from software contributed to The NetBSD Foundation
.\" by Paul Kranenburg.
.\"
.\" 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.
.\" 3. All advertising materials mentioning features or use of this software
.\" must display the following acknowledgement:
.\" This product includes software developed by the NetBSD
.\" Foundation, Inc. and its contributors.
.\" 4. Neither the name of The NetBSD Foundation nor the names of its
.\" contributors may be used to endorse or promote products derived
.\" from this software without specific prior written permission.
.\"
.\" 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 June 23, 1996
.Dt SLEEP 9
.Os
.Sh NAME
.Nm ltsleep ,
.Nm sleep ,
.Nm tsleep ,
.Nm wakeup
.Nd process context sleep and and wakeup
.Sh SYNOPSIS
.Fd #include <sys/proc.h>
.Ft int
.Fn "ltsleep" "void *ident" "int priority" "const char *wmesg" "int timo" "__volatile struct simplelock *slock"
.Ft int
.Fn "tsleep" "void *ident" "int priority" "const char *wmesg" "int timo"
.Ft void
.Fn "sleep" "void *ident" "int priority"
.Ft void
.Fn "wakeup" "void *ident"
.Sh DESCRIPTION
These functions implement voluntary context switching.
.Fn ltsleep ,
.Fn tsleep
and
.Fn sleep
are used throughout the kernel whenever processing in the current context
can not continue for any of the following reasons:
.Bl -bullet -offset indent
.It
The current process needs to await the results of a pending I/O operation.
.It
The current process needs resources
.Pq e.g. memory
which are temporarily unavailable.
.It
The current process wants access to data-structures which are locked by
other processes.
.El
.Pp
The function
.Fn wakeup
is used to notify sleeping processes of possible changes to the condition
that caused them to go to sleep.
Typically, an awakened process will -- after it has acquired a context
again -- retry the action that blocked its operation to see if the
.Dq blocking
condition has cleared.
.Pp
The
.Fn ltsleep
function takes the following arguments:
.Bl -tag -width priority
.It Fa ident
An identifier of the
.Dq wait channel
representing the resource for which the current process needs to wait.
This typically is the virtual address of some kernel data-structure related
to the resource for which the process is contending.
The same identifier must be used in a call to
.Fn wakeup
to get the process going again.
.Fa ident
should not be
.Dv NULL .
.It Fa priority
The process priority to be used when the process is awakened and put on
the queue of runnable processes.
This mechanism is used to optimize
.Dq throughput
of processes executing in kernel mode.
If the flag
.Dv PCATCH
is OR'ed into
.Fa priority
the process checks for posted signals before and after sleeping.
If the flag
.Dv PNORELOCK
is OR'ed into
.Fa priority ,
.Fa slock
is NOT re-locked after process resume.
.It Fa wmesg
A pointer to a character string indicating the reason a process is sleeping.
The kernel does not use the string, but makes it available
.Pq through the process structure field Li p_wmesg
for user level utilities such as
.Xr ps 1 .
.It Fa timo
If non-zero, the process will sleep for at most
.Li timo/hz
seconds.
If this amount of time elapses and no
.Fn wakeup "ident"
has occurred, and no signal
.Pq if Dv PCATCH No was set
was posted,
.Fn tsleep
will return
.Er EWOULDBLOCK .
.It Fa slock
If not NULL, the
.Fa slock
interlock is unlocked once the scheduler lock is acquired. Unless
.Dv PNORELOCK
was set,
.Fa slock
is locked again once
the process is resumed from sleep. This provides wakeup-before-sleep
condition protection facility.
.El
.Pp
The
.Fn tsleep
macro is functionally equivalent to:
.Bd -literal -offset indent
ltsleep(ident, priority, wmesg, timo, NULL)
.Ed
.Pp
The
.Fn sleep
function puts the process in an uninterruptible sleep.
It is functionally equivalent to:
.Bd -literal -offset indent
ltsleep(ident, priority & PRIMASK, 0, 0, NULL)
.Ed
.Pp
The
.Fn wakeup
function will mark all processes which are currently sleeping on the identifier
.Fa ident
as runnable.
Eventually, each of the processes will resume execution in the kernel
context, causing a return from
.Fn [t]sleep .
Note that processes returning from sleep should always re-evaluate the
conditions that blocked them, since a call to
.Fn wakeup
merely signals a
.Em possible
change to the blocking conditions.
For example, when two or more processes are waiting for an exclusive-access
lock
.Pq see Xr lock 9 ,
only one of them will succeed in acquiring the lock when it is released.
All others will have to go back to sleep and wait for the next opportunity.
.Sh RETURN VALUES
.Fn ltsleep
returns 0 if it returns as a result of a
.Fn wakeup .
If a
.Fn ltsleep
returns as a result of a signal, the return value is
.Er ERESTART
if the signal has the
.Dv SA_RESTART
property
.Pq see Xr sigaction 2 ,
and
.Er EINTR
otherwise.
If
.Fn ltsleep
returns because of a timeout it returns
.Er EWOULDBLOCK .
.Sh HISTORY
The sleep/wakeup process synchronization mechanism is very old. It
appeared in a very early version of Unix.
.Fn tsleep
appeared in
.Bx 4.4 .
.Fn ltsleep
appeared in
.Nx 1.5 .
.Sh SEE ALSO
.Xr sigaction 2 ,
.Xr hz 9 ,
.Xr lock 9