xref: /freebsd/share/man/man9/mi_switch.9 (revision b52b9d56d4e96089873a75f9e29062eec19fabba)
1.\"	$NetBSD: ctxsw.9,v 1.2 1996/12/02 00:11:31 tls Exp $
2.\"
3.\" Copyright (c) 1996 The NetBSD Foundation, Inc.
4.\" All rights reserved.
5.\"
6.\" This code is derived from software contributed to The NetBSD Foundation
7.\" by Paul Kranenburg.
8.\"
9.\" Redistribution and use in source and binary forms, with or without
10.\" modification, are permitted provided that the following conditions
11.\" are met:
12.\" 1. Redistributions of source code must retain the above copyright
13.\"    notice, this list of conditions and the following disclaimer.
14.\" 2. Redistributions in binary form must reproduce the above copyright
15.\"    notice, this list of conditions and the following disclaimer in the
16.\"    documentation and/or other materials provided with the distribution.
17.\" 3. All advertising materials mentioning features or use of this software
18.\"    must display the following acknowledgement:
19.\"        This product includes software developed by the NetBSD
20.\"        Foundation, Inc. and its contributors.
21.\" 4. Neither the name of The NetBSD Foundation nor the names of its
22.\"    contributors may be used to endorse or promote products derived
23.\"    from this software without specific prior written permission.
24.\"
25.\" THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
26.\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
27.\" TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
28.\" PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE
29.\" LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
30.\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
31.\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
32.\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
33.\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
34.\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
35.\" POSSIBILITY OF SUCH DAMAGE.
36.\"
37.\" $FreeBSD$
38.\"
39.Dd November 24, 1996
40.Dt MI_SWITCH 9
41.Os
42.Sh NAME
43.Nm mi_switch ,
44.Nm cpu_switch ,
45.Nm cpu_throw
46.Nd switch to another thread context
47.Sh SYNOPSIS
48.In sys/param.h
49.In sys/proc.h
50.Ft void
51.Fn mi_switch "void"
52.Ft void
53.Fn cpu_switch "void"
54.Ft void
55.Fn cpu_throw "void"
56.Sh DESCRIPTION
57The
58.Fn mi_switch
59function implements the machine independent prelude to a thread context
60switch.
61It is called from only a few distinguished places in the kernel
62code as a result of the principle of non-preemtable kernel mode execution.
63The various major uses of
64.Nm
65can be enumerated as follows:
66.Bl -enum -offset indent
67.It
68from within
69.Xr sleep 9
70,
71.Xr tsleep 9
72and
73.Xr msleep 9
74when the current thread
75voluntarily relinquishes the CPU to wait for some resource to become
76available.
77.It
78after handling a trap
79(e.g. a system call, device interrupt)
80when the kernel prepares a return to user-mode execution.
81This case is
82typically handled by machine dependent trap-handling code after detection
83of a change in the signal disposition of the current process, or when a
84higher priority thread might be available to run.
85The latter event is
86communicated by the machine independent scheduling routines by calling
87the machine defined
88.Fn need_resched .
89.It
90in the signal handling code
91(see
92.Xr issignal 9 )
93if a signal is delivered that causes a process to stop.
94.It
95When a thread dies in
96.Xr thread_exit 9
97and control of the processor can be passed to the next runnable thread.
98.It
99In
100.Xr thread_suspend_check 9
101where a thread needs to stop execution due to the suspension state of
102the process as a whole.
103.El
104.Pp
105.Fn mi_switch
106records the amount of time the current thread has been running in the
107process structures and checks this value against the CPU time limits
108allocated to the process
109(see
110.Xr getrlimit 2 ) .
111Exceeding the soft limit results in a
112.Dv SIGXCPU
113signal to be posted to the process, while exceeding the hard limit will
114cause a
115.Dv SIGKILL .
116.Pp
117If the thread is still in
118.Em TDS_RUNNING
119state mi_switch will put it back onto the run queue, assuming that
120it will want to run again soon. If it is in one of the other
121states  and KSE threading is enabled, the associated
122.Em KSE
123will be made available to to any higher priority threads from the same
124group, to allow them to be scheduled next.
125.Pp
126After these administrative tasks are done,
127.Fn mi_switch
128hands over control to the machine dependent routine
129.Fn cpu_switch ,
130which will perform the actual thread context switch.
131.Pp
132.Fn cpu_switch
133first saves the context of the current thread.
134Next, it calls
135.Fn choosethread
136to determine which thread to run next.
137Finally, it reads in the saved context of the new thread and starts to
138execute the new thread.
139.Pp
140.Fn cpu_throw
141is similar to
142.Fn cpu_switch
143except that it does not save the context of the old thread.
144This function is useful when the kernel does not have an old thread
145context to save, such as when CPUs other than the boot CPU perform their
146first task switch, or when the kernel does not care about the state of the
147old thread, such as in
148.Fn thread_exit
149when the kernel terminates the current thread and switches into a new
150thread.
151.Pp
152To protect the
153.Xr runqueue 9 ,
154all of these functions must be called with the
155.Va sched_lock
156mutex held.
157.Sh SEE ALSO
158.Xr issignal 9 ,
159.Xr mutex 9 ,
160.Xr runqueue 9 ,
161.Xr tsleep 9 ,
162.Xr wakeup 9
163