CTXSW(9) BSD Kernel Manual CTXSW(9)NAME
mi_switch, cpu_switch - switch to another process context
SYNOPSIS
#include <sys/param.h>
#include <sys/proc.h>
void
mi_switch(void);
void
cpu_switch(void);
DESCRIPTION
The mi_switch() function implements the machine-independent prelude to a
process context switch. It is called from only a few distinguished places
in the kernel code as a result of the principle of non-preemptable kernel
mode execution. The three major uses of mi_switch() can be enumerated as
follows:
1. From within sleep(9) and tsleep(9) when the current process
voluntarily relinquishes the CPU to wait for some resource to
become available.
2. After handling a trap (e.g., a system call or device
interrupt) when the kernel prepares a return to user-mode exe-
cution. This case is typically handled by machine-dependent
trap-handling code after detection of a change in the signal
disposition of the current process, or when a higher priority
process might be available to run. The latter event is commun-
icated by the machine-independent scheduling routines by cal-
ling the machine-dependent need_resched(void).
3. In the signal handling code (see issignal(9)) if a signal is
delivered that causes a process to stop.
mi_switch() records the amount of time the current process has been run-
ning in the process structure and checks this value against the CPU time
limits allocated to the process (see getrlimit(2)). Exceeding the soft
limit results in a SIGXCPU signal to be posted to the process, while
exceeding the hard limit will cause a SIGKILL. For a process which accu-
mulated longer than 10 minutes of CPU time, its nice level is raised to
4. After these administrative tasks are done, mi_switch() hands over con-
trol to the machine dependent routine cpu_switch(void), which will per-
form the actual process context switch.
cpu_switch() will make a choice amongst the processes which are ready to
run from a priority queue data-structure. The priority queue consists of
an array qs[NQS] of queue header structures each of which identifies a
list of runnable processes of equal priority (see <sys/proc.h>). A single
word whichqs containing a bit mask identifying non-empty queues assists
in selecting a process quickly. cpu_switch() must remove the first pro-
cess from the list on the queue with the highest priority (lower indices
in qs indicate higher priority), and assign the address of its process
structure to the global variable curproc. If no processes are available
on the run queues, cpu_switch() shall go into an "idle" loop. The idle
loop must allow interrupts to be taken that will eventually cause
processes to appear again on the run queues. The variable curproc should
be NULL while cpu_switch() waits for this to happen.
Note that mi_switch() and thus cpu_switch() should be called at
splhigh(9).
SEE ALSOspl(9), tsleep(9), wakeup(9)MirOS BSD #10-current November 24, 1996 1