freebsd-src/usr.sbin/powerd/powerd.8

148 lines
4.4 KiB
Groff
Raw Normal View History

.\" Copyright (c) 2005 Nate Lawson
.\" All rights reserved.
.\"
.\" 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 REGENTS 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 REGENTS 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.
.\"
.\" $FreeBSD$
.\"
.Dd December 21, 2009
.Dt POWERD 8
.Os
.Sh NAME
.Nm powerd
2005-06-16 20:01:43 +00:00
.Nd "system power control utility"
.Sh SYNOPSIS
.Nm
2005-06-16 20:01:43 +00:00
.Op Fl a Ar mode
.Op Fl b Ar mode
.Op Fl i Ar percent
.Op Fl m Ar freq
.Op Fl M Ar freq
2005-06-16 20:01:43 +00:00
.Op Fl n Ar mode
.Op Fl p Ar ival
.Op Fl P Ar pidfile
2005-06-16 20:01:43 +00:00
.Op Fl r Ar percent
.Op Fl v
.Sh DESCRIPTION
The
.Nm
utility monitors the system state and sets various power control options
accordingly.
It offers four modes (maximum, minimum, adaptive and hiadaptive) that can be
individually selected while on AC power or batteries.
Set of powerd enchancements: 1. Make it more SMP polite. Previous version uses average CPU load that often leads to load underestimation. It make powerd with default configuration unusable on systems with more then 2 CPUs. I propose to use summary load instead of average one. IMO this is the best we can do without specially tuned scheduler. Also as soon as measuring total load on SMP systems is more useful then total idle, I have switched to it. 2. Make powerd's operation independent from number and size of frequency levels. I have added internal frequency counter which translated into real frequencies only on a last stage and only as good as gone. Some systems may have only several power levels, while others - many of them, so adaptation time with previous approach was completely different. 3. As part of previous I have changed adaptive mode to rise frequency on demand up to 2 times and fall on 1/8 per time internal. 4. For desktop (AC-powered) systems I have added one more mode - "hiadaptive". It rises frequency twice faster, drops it 4 times slower, prefers twice lower CPU load and has additional delay before leaving the highest frequency after the period of maximum load. This mode was specially made to improve interactivity of the systems where operation capabilities are more significant then power consumption, but keeping maximum frequency all the time is not needed. 5. I have reduced default polling interval from 1/2 to 1/4 of second. It is not so important for algorithm math now, but gives better system interactivity. Discussed on: mobile@
2008-11-18 13:24:38 +00:00
The modes maximum, minimum, adaptive and hiadaptive may be abbreviated
max, min, adp, hadp.
.Pp
Maximum mode chooses the highest performance values.
Minimum mode selects the lowest performance values to get the most power
savings.
Adaptive mode attempts to strike a balance by degrading performance when
the system appears idle and increasing it when the system is busy.
It offers a good balance between a small performance loss for greatly
increased power savings.
Hiadaptive mode is like adaptive mode, but tuned for systems where
2010-03-15 16:06:52 +00:00
performance and interactivity are more important than power consumption.
It increases frequency faster, reduces the frequency less aggressively and
will maintain full frequency for longer.
Set of powerd enchancements: 1. Make it more SMP polite. Previous version uses average CPU load that often leads to load underestimation. It make powerd with default configuration unusable on systems with more then 2 CPUs. I propose to use summary load instead of average one. IMO this is the best we can do without specially tuned scheduler. Also as soon as measuring total load on SMP systems is more useful then total idle, I have switched to it. 2. Make powerd's operation independent from number and size of frequency levels. I have added internal frequency counter which translated into real frequencies only on a last stage and only as good as gone. Some systems may have only several power levels, while others - many of them, so adaptation time with previous approach was completely different. 3. As part of previous I have changed adaptive mode to rise frequency on demand up to 2 times and fall on 1/8 per time internal. 4. For desktop (AC-powered) systems I have added one more mode - "hiadaptive". It rises frequency twice faster, drops it 4 times slower, prefers twice lower CPU load and has additional delay before leaving the highest frequency after the period of maximum load. This mode was specially made to improve interactivity of the systems where operation capabilities are more significant then power consumption, but keeping maximum frequency all the time is not needed. 5. I have reduced default polling interval from 1/2 to 1/4 of second. It is not so important for algorithm math now, but gives better system interactivity. Discussed on: mobile@
2008-11-18 13:24:38 +00:00
The default mode is adaptive for battery power and hiadaptive for the rest.
.Pp
The
.Nm
utility recognizes the following runtime options:
2005-06-16 20:01:43 +00:00
.Bl -tag -width ".Fl r Ar percent"
.It Fl a Ar mode
Selects the
.Ar mode
to use while on AC power.
.It Fl b Ar mode
Selects the
.Ar mode
to use while on battery power.
.It Fl i Ar percent
Set of powerd enchancements: 1. Make it more SMP polite. Previous version uses average CPU load that often leads to load underestimation. It make powerd with default configuration unusable on systems with more then 2 CPUs. I propose to use summary load instead of average one. IMO this is the best we can do without specially tuned scheduler. Also as soon as measuring total load on SMP systems is more useful then total idle, I have switched to it. 2. Make powerd's operation independent from number and size of frequency levels. I have added internal frequency counter which translated into real frequencies only on a last stage and only as good as gone. Some systems may have only several power levels, while others - many of them, so adaptation time with previous approach was completely different. 3. As part of previous I have changed adaptive mode to rise frequency on demand up to 2 times and fall on 1/8 per time internal. 4. For desktop (AC-powered) systems I have added one more mode - "hiadaptive". It rises frequency twice faster, drops it 4 times slower, prefers twice lower CPU load and has additional delay before leaving the highest frequency after the period of maximum load. This mode was specially made to improve interactivity of the systems where operation capabilities are more significant then power consumption, but keeping maximum frequency all the time is not needed. 5. I have reduced default polling interval from 1/2 to 1/4 of second. It is not so important for algorithm math now, but gives better system interactivity. Discussed on: mobile@
2008-11-18 13:24:38 +00:00
Specifies the CPU load percent level when adaptive
mode should begin to degrade performance to save power.
Set of powerd enchancements: 1. Make it more SMP polite. Previous version uses average CPU load that often leads to load underestimation. It make powerd with default configuration unusable on systems with more then 2 CPUs. I propose to use summary load instead of average one. IMO this is the best we can do without specially tuned scheduler. Also as soon as measuring total load on SMP systems is more useful then total idle, I have switched to it. 2. Make powerd's operation independent from number and size of frequency levels. I have added internal frequency counter which translated into real frequencies only on a last stage and only as good as gone. Some systems may have only several power levels, while others - many of them, so adaptation time with previous approach was completely different. 3. As part of previous I have changed adaptive mode to rise frequency on demand up to 2 times and fall on 1/8 per time internal. 4. For desktop (AC-powered) systems I have added one more mode - "hiadaptive". It rises frequency twice faster, drops it 4 times slower, prefers twice lower CPU load and has additional delay before leaving the highest frequency after the period of maximum load. This mode was specially made to improve interactivity of the systems where operation capabilities are more significant then power consumption, but keeping maximum frequency all the time is not needed. 5. I have reduced default polling interval from 1/2 to 1/4 of second. It is not so important for algorithm math now, but gives better system interactivity. Discussed on: mobile@
2008-11-18 13:24:38 +00:00
The default is 50% or lower.
.It Fl m Ar freq
Specifies the minimum frequency to throttle down to.
.It Fl M Ar freq
Specifies the maximum frequency to throttle up to.
.It Fl n Ar mode
Selects the
.Ar mode
to use normally when the AC line state is unknown.
.It Fl p Ar ival
Specifies a different polling interval (in milliseconds) for AC line state
and system idle levels.
Set of powerd enchancements: 1. Make it more SMP polite. Previous version uses average CPU load that often leads to load underestimation. It make powerd with default configuration unusable on systems with more then 2 CPUs. I propose to use summary load instead of average one. IMO this is the best we can do without specially tuned scheduler. Also as soon as measuring total load on SMP systems is more useful then total idle, I have switched to it. 2. Make powerd's operation independent from number and size of frequency levels. I have added internal frequency counter which translated into real frequencies only on a last stage and only as good as gone. Some systems may have only several power levels, while others - many of them, so adaptation time with previous approach was completely different. 3. As part of previous I have changed adaptive mode to rise frequency on demand up to 2 times and fall on 1/8 per time internal. 4. For desktop (AC-powered) systems I have added one more mode - "hiadaptive". It rises frequency twice faster, drops it 4 times slower, prefers twice lower CPU load and has additional delay before leaving the highest frequency after the period of maximum load. This mode was specially made to improve interactivity of the systems where operation capabilities are more significant then power consumption, but keeping maximum frequency all the time is not needed. 5. I have reduced default polling interval from 1/2 to 1/4 of second. It is not so important for algorithm math now, but gives better system interactivity. Discussed on: mobile@
2008-11-18 13:24:38 +00:00
The default is 250 ms.
.It Fl P Ar pidfile
Specifies an alternative file in which the process ID should be stored.
The default is
.Pa /var/run/powerd.pid .
.It Fl r Ar percent
Set of powerd enchancements: 1. Make it more SMP polite. Previous version uses average CPU load that often leads to load underestimation. It make powerd with default configuration unusable on systems with more then 2 CPUs. I propose to use summary load instead of average one. IMO this is the best we can do without specially tuned scheduler. Also as soon as measuring total load on SMP systems is more useful then total idle, I have switched to it. 2. Make powerd's operation independent from number and size of frequency levels. I have added internal frequency counter which translated into real frequencies only on a last stage and only as good as gone. Some systems may have only several power levels, while others - many of them, so adaptation time with previous approach was completely different. 3. As part of previous I have changed adaptive mode to rise frequency on demand up to 2 times and fall on 1/8 per time internal. 4. For desktop (AC-powered) systems I have added one more mode - "hiadaptive". It rises frequency twice faster, drops it 4 times slower, prefers twice lower CPU load and has additional delay before leaving the highest frequency after the period of maximum load. This mode was specially made to improve interactivity of the systems where operation capabilities are more significant then power consumption, but keeping maximum frequency all the time is not needed. 5. I have reduced default polling interval from 1/2 to 1/4 of second. It is not so important for algorithm math now, but gives better system interactivity. Discussed on: mobile@
2008-11-18 13:24:38 +00:00
Specifies the CPU load percent level where adaptive
mode should consider the CPU running and increase performance.
The default is 75% or higher.
.It Fl v
Verbose mode.
Messages about power changes will be printed to stdout and
.Nm
will operate in the foreground.
.El
.Sh SEE ALSO
.Xr acpi 4 ,
.Xr apm 4 ,
.Xr cpufreq 4
.Sh HISTORY
The
.Nm
utility first appeared in
.Fx 6.0 .
.Sh AUTHORS
2005-06-16 20:01:43 +00:00
.An -nosplit
.An Colin Percival
first wrote
2005-06-16 20:01:43 +00:00
.Nm estctrl ,
the utility that
.Nm
is based on.
.An Nate Lawson
then updated it for
.Xr cpufreq 4 ,
2005-02-27 15:49:07 +00:00
added features, and wrote this manual page.
.Sh BUGS
The
.Nm
utility should also power down idle disks and other components besides the CPU.
.Pp
If
.Nm
2005-06-16 20:01:43 +00:00
is used with
.Pa power_profile ,
they may override each other.
.Pp
2005-06-16 20:01:43 +00:00
The
.Nm
2005-06-16 20:01:43 +00:00
utility
should probably use the
.Xr devctl 4
interface instead of polling for AC line state.