man9: Add man page for kern_yield() and friends
The page is fairly simple and will be referenced by mi_switch(9). Provide some usage notes so that the broader implications of how/when to use these functions are understood. Reviewed by: kib, markj MFC after: 1 week Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D38182
This commit is contained in:
parent
dc9b13736f
commit
30cd6fd75d
@ -196,6 +196,7 @@ MAN= accept_filter.9 \
|
|||||||
KASSERT.9 \
|
KASSERT.9 \
|
||||||
kern_reboot.9 \
|
kern_reboot.9 \
|
||||||
kern_testfrwk.9 \
|
kern_testfrwk.9 \
|
||||||
|
kern_yield.9 \
|
||||||
kernacc.9 \
|
kernacc.9 \
|
||||||
kernel_mount.9 \
|
kernel_mount.9 \
|
||||||
khelp.9 \
|
khelp.9 \
|
||||||
@ -1332,6 +1333,8 @@ MLINKS+=intr_event.9 intr_event_add_handler.9 \
|
|||||||
intr_event.9 intr_event_handle.9 \
|
intr_event.9 intr_event_handle.9 \
|
||||||
intr_event.9 intr_event_remove_handler.9 \
|
intr_event.9 intr_event_remove_handler.9 \
|
||||||
intr_event.9 intr_priority.9
|
intr_event.9 intr_priority.9
|
||||||
|
MLINKS+=kern_yield.9 maybe_yield.9 \
|
||||||
|
kern_yield.9 should_yield.9
|
||||||
MLINKS+=kernacc.9 useracc.9
|
MLINKS+=kernacc.9 useracc.9
|
||||||
MLINKS+=kernel_mount.9 free_mntarg.9 \
|
MLINKS+=kernel_mount.9 free_mntarg.9 \
|
||||||
kernel_mount.9 mount_arg.9 \
|
kernel_mount.9 mount_arg.9 \
|
||||||
|
134
share/man/man9/kern_yield.9
Normal file
134
share/man/man9/kern_yield.9
Normal file
@ -0,0 +1,134 @@
|
|||||||
|
.\"
|
||||||
|
.\" Copyright (c) 2023 The FreeBSD Foundation
|
||||||
|
.\"
|
||||||
|
.\" This documentation was written by Mitchell Horne <mhorne@FreeBSD.org> under
|
||||||
|
.\" sponsorship from the FreeBSD Foundation.
|
||||||
|
.\"
|
||||||
|
.\" 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.
|
||||||
|
.\"
|
||||||
|
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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 January 30, 2023
|
||||||
|
.Dt KERN_YIELD 9
|
||||||
|
.Os
|
||||||
|
.Sh NAME
|
||||||
|
.Nm kern_yield ,
|
||||||
|
.Nm maybe_yield ,
|
||||||
|
.Nm should_yield
|
||||||
|
.Nd "functions for yielding execution of the current thread"
|
||||||
|
.Sh SYNOPSIS
|
||||||
|
.Ft void
|
||||||
|
.Fn kern_yield "int prio"
|
||||||
|
.Ft void
|
||||||
|
.Fn maybe_yield
|
||||||
|
.Ft bool
|
||||||
|
.Fn should_yield
|
||||||
|
.Sh DESCRIPTION
|
||||||
|
The
|
||||||
|
.Fn kern_yield
|
||||||
|
function causes the currently running thread to voluntarily, but
|
||||||
|
unconditionally, surrender its execution to the scheduler.
|
||||||
|
The
|
||||||
|
.Va prio
|
||||||
|
argument specifies the scheduling priority to be assigned before the context
|
||||||
|
switch, which has an influence on when execution will resume.
|
||||||
|
Note that the requested priority will take effect until the thread returns to
|
||||||
|
usermode, after which its base user priority will be restored.
|
||||||
|
Valid values for
|
||||||
|
.Va prio
|
||||||
|
are any of the
|
||||||
|
.Dv PRI_*
|
||||||
|
values defined in
|
||||||
|
.In sys/priority.h ,
|
||||||
|
as well as the following special values:
|
||||||
|
.Bl -tag -width "PRI_UNCHANGED"
|
||||||
|
.It Dv PRI_USER
|
||||||
|
Schedule the thread with its base user priority; the value corresponding to
|
||||||
|
.Xr setpriority 2 /
|
||||||
|
.Xr nice 3 .
|
||||||
|
.It Dv PRI_UNCHANGED
|
||||||
|
Yield the thread without changing its priority.
|
||||||
|
.El
|
||||||
|
.Pp
|
||||||
|
The
|
||||||
|
.Fn should_yield
|
||||||
|
function checks if sufficient time has passed since the thread's last voluntary
|
||||||
|
context switch that yielding would be a useful service to other threads.
|
||||||
|
It returns
|
||||||
|
.Va true
|
||||||
|
when this is the case.
|
||||||
|
See
|
||||||
|
.Sx USAGE NOTES
|
||||||
|
for an elaboration of what this means.
|
||||||
|
.Pp
|
||||||
|
The
|
||||||
|
.Fn maybe_yield
|
||||||
|
function is a helper function for the common task of optionally yielding the
|
||||||
|
processor.
|
||||||
|
Internally,
|
||||||
|
.Fn kern_yield "PRI_USER"
|
||||||
|
will be called if
|
||||||
|
.Fn should_yield
|
||||||
|
returns
|
||||||
|
.Va true .
|
||||||
|
.Sh USAGE NOTES
|
||||||
|
Although the kernel supports preemption, this is usually reserved for
|
||||||
|
high-priority realtime or interrupt threads.
|
||||||
|
Kernel worker threads and timesharing threads are not guaranteed to preempt
|
||||||
|
each another.
|
||||||
|
Thus, threads executing in the kernel are expected to behave cooperatively
|
||||||
|
with respect to other threads in the system.
|
||||||
|
The yield functions are mostly intended to be used by threads which perform a
|
||||||
|
lot of work inside the kernel.
|
||||||
|
For example:
|
||||||
|
.Fn maybe_yield
|
||||||
|
is called by the
|
||||||
|
.Dv vlnru
|
||||||
|
process each time it reclaims a vnode.
|
||||||
|
.Pp
|
||||||
|
The scheduler aims to identify threads which monopolize the CPU, and will
|
||||||
|
schedule them with decreased priority.
|
||||||
|
Threads which regularly yield the processor will be given the chance to run
|
||||||
|
more often.
|
||||||
|
The possibly surprising effect of this is that, depending on the disposition of
|
||||||
|
other threads on the CPU's runqueue, a call to
|
||||||
|
.Fn kern_yield
|
||||||
|
does not guarantee that the yielding thread will be taken off the CPU.
|
||||||
|
.Pp
|
||||||
|
With the above considerations in mind, it is advised that code written using
|
||||||
|
.Fn kern_yield
|
||||||
|
be measured to confirm that its use has a positive effect on relevant
|
||||||
|
performance or responsiveness metrics.
|
||||||
|
Switching thread contexts has a non-zero cost, and thus yielding the processor
|
||||||
|
too eagerly could have a negative impact on performance.
|
||||||
|
.Pp
|
||||||
|
Additionally, since yielding is a cooperative action, it is advised that the
|
||||||
|
yielding thread release any locks that it may be holding, when possible.
|
||||||
|
Otherwise, threads which have been given the chance to run could end up waiting
|
||||||
|
on the yielding thread to release the lock, largely defeating the purpose of
|
||||||
|
the yield.
|
||||||
|
.Sh SEE ALSO
|
||||||
|
.Xr setpriority 2 ,
|
||||||
|
.Xr nice 3 ,
|
||||||
|
.Xr mi_switch 9
|
||||||
|
.Sh AUTHORS
|
||||||
|
.An -nosplit
|
||||||
|
This manual page was written by
|
||||||
|
.An Mitchell Horne Aq Mt mhorne@FreeBSD.org .
|
Loading…
Reference in New Issue
Block a user