xref: /openbsd-src/share/man/man9/mi_switch.9 (revision 84eb796f31be5cb26d74c92cc2583b24360b30b0)
1.\"	$OpenBSD: mi_switch.9,v 1.7 2017/09/05 03:49:56 jmatthew Exp $
2.\"	$NetBSD: ctxsw.9,v 1.9 1999/03/06 22:09:29 mycroft Exp $
3.\"
4.\" Copyright (c) 1996 The NetBSD Foundation, Inc.
5.\" All rights reserved.
6.\"
7.\" This code is derived from software contributed to The NetBSD Foundation
8.\" by Paul Kranenburg.
9.\"
10.\" Redistribution and use in source and binary forms, with or without
11.\" modification, are permitted provided that the following conditions
12.\" are met:
13.\" 1. Redistributions of source code must retain the above copyright
14.\"    notice, this list of conditions and the following disclaimer.
15.\" 2. Redistributions in binary form must reproduce the above copyright
16.\"    notice, this list of conditions and the following disclaimer in the
17.\"    documentation and/or other materials provided with the distribution.
18.\"
19.\" THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
20.\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
21.\" TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
22.\" PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
23.\" BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
24.\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
25.\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
26.\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
27.\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
28.\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
29.\" POSSIBILITY OF SUCH DAMAGE.
30.\"
31.Dd $Mdocdate: September 5 2017 $
32.Dt MI_SWITCH 9
33.Os
34.Sh NAME
35.Nm mi_switch ,
36.Nm cpu_switchto
37.Nd switch to another process context
38.Sh SYNOPSIS
39.In sys/param.h
40.In sys/proc.h
41.Ft void
42.Fn mi_switch "void"
43.Ft void
44.Fn cpu_switchto "struct proc *old" "struct proc *new"
45.Sh DESCRIPTION
46The
47.Fn mi_switch
48function implements the machine-independent prelude to a process context
49switch.
50It is called from only a few distinguished places in the kernel code as a
51result of the principle of non-preemptable kernel mode execution.
52The three major uses of
53.Fn mi_switch
54can be enumerated as follows:
55.Bl -enum -offset indent
56.It
57From within functions like
58.Xr tsleep 9 ,
59when the current process sleeps to wait for some resource to become
60available, or from within functions like
61.Fn yield ,
62when it relinquishes the CPU to allow other processes to make progress.
63.It
64After handling a trap
65.Pq e.g., a system call or device interrupt
66when the kernel prepares a return to user-mode execution.
67This case is typically handled by machine-dependent trap-handling code after
68detection of a change in the signal disposition of the current process, or
69when a higher priority process might be available to run.
70The latter event is communicated by the machine-independent scheduling
71routines by calling the machine-dependent
72.Fn need_resched "void" .
73.It
74In the signal handling code
75.Pq see Pa sys/kern/kern_sig.c
76if a signal is delivered that causes a process to stop.
77.El
78.Pp
79.Fn mi_switch
80records the amount of time the current process has been running in the
81process structure and checks this value against the CPU time limits
82allocated to the process
83.Pq see Xr getrlimit 2 .
84Exceeding the soft limit results in a
85.Dv SIGXCPU
86signal to be posted to the process, while exceeding the hard limit will
87cause a
88.Dv SIGKILL .
89After these administrative tasks are done,
90.Fn mi_switch
91chooses the next process to run and hands over control to the machine
92dependent routine
93.Fn cpu_switchto ,
94which will perform the actual process context switch.
95.Pp
96.Fn cpu_switchto
97will save the context of the old process and switch to the new one.
98A special case is when the old process is
99.Dv NULL
100which means that the old process has exited and doesn't need to be
101saved.
102.Pp
103Note that
104.Fn mi_switch
105and thus
106.Fn cpu_switchto
107must be called while holding the scheduler lock.
108.Sh SEE ALSO
109.Xr spl 9 ,
110.Xr tsleep 9 ,
111.Xr wakeup 9
112