rpms/cpuspeed/devel cpuspeed.8,1.1,1.2
Jarod Wilson
jwilson at fedoraproject.org
Thu Oct 9 00:26:51 UTC 2008
Author: jwilson
Update of /cvs/pkgs/rpms/cpuspeed/devel
In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv10115
Modified Files:
cpuspeed.8
Log Message:
Update man page for 1.5.x (and upstream submission)
Index: cpuspeed.8
===================================================================
RCS file: /cvs/pkgs/rpms/cpuspeed/devel/cpuspeed.8,v
retrieving revision 1.1
retrieving revision 1.2
diff -u -r1.1 -r1.2
--- cpuspeed.8 9 Jan 2007 15:51:31 -0000 1.1
+++ cpuspeed.8 9 Oct 2008 00:26:20 -0000 1.2
@@ -1,12 +1,12 @@
.\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.36.
-.TH CPUSPEED "8" "January 09, 2005" "cpuspeed v1.2.1" "Program Options"
+.TH CPUSPEED "8" "October 08, 2008" "cpuspeed v1.5.1" "Program Options"
.SH NAME
-cpuspeed \- user-space cpu frequency scaling program, v1.2.1
+cpuspeed \- user-space cpu frequency scaling program, v1.5.1
.SH SYNOPSIS
.B cpuspeed
[\fIOptions\fR]
.SH DESCRIPTION
-cpuspeed v1.2.1
+cpuspeed v1.5.1
.PP
This program monitors the system's idle percentage and reduces or raises the
CPUs' clock speeds and voltages accordingly to minimize power consumption
@@ -31,22 +31,28 @@
changes in tenths of a second (default is 20).
.HP
\fB\-p\fR <fast up> <threshold>
-.TP
-Sets the CPU idle percentage thresholds.
-<fast up> is the idle
.IP
-percentage below which a CPU will be set to the highest possible
-speed. <threshold> is the idle percentage above which a CPU's
-speed will be decreased and below which a CPU's speed will be
+Sets the CPU core idle percentage thresholds. <fast up> is the idle
+percentage below which a CPU core will be set to the highest possible
+speed. <threshold> is the idle percentage above which a CPU core's
+speed will be decreased and below which a CPU core's speed will be
increased (defaults are 10 and 25).
.HP
\fB\-m\fR <minimum speed>
.IP
-Sets the minimum speed in KHz below which a CPU will not be set.
+Sets the minimum speed in KHz below which a CPU core will not be set.
.HP
\fB\-M\fR <maximum speed>
.IP
-Sets the maximum speed in KHz above which a CPU will not be set.
+Sets the maximum speed in KHz above which a CPU core will not be set.
+.HP
+\fB\-n\fR
+.IP
+Do not treat niced programs as idle time.
+.HP
+\fB\-w\fR
+.IP
+Do not treat time waiting for IO as idle time.
.HP
\fB\-t\fR <temp file> <maxtemp>
.IP
@@ -56,7 +62,7 @@
\fB\-T\fR <interval>
.IP
Sets the interval at which the temperature will be polled in
-tenths of a second (default is 10).
+tenths of a second (default is 10).
(Requires the '\-t' option above.)
.HP
\fB\-a\fR <AC file>
@@ -85,24 +91,29 @@
.IP
Restores previous speed on program exit.
.HP
-\fB\-s\fR <CPU>
-.TP
-Manage only a single CPU.
-<CPU> specifies the number of the CPU
+\fB\-S\fR <CPU core 1> [[<CPU core 2>] ...]
.IP
-to manage and is in the range 0 to (n\-1) where 'n' is the number
-of CPUs in the system. Without this option the program creates
-copies of itself to manage every CPU in the system.
-.IP
-To have a CPU stay at the highest clock speed to maximize performance send
-the process controlling that CPU the SIGUSR1 signal.
-.IP
-To have a CPU stay at the lowest clock speed to maximize battery life send
-the process controlling that CPU the SIGUSR2 signal.
-.IP
-To resume having a CPU's clock speed dynamically scaled send the process
-controlling that CPU the SIGHUP signal.
+Manage only a single group of CPU cores. All of the specified
+cores will controlled as a single group (locked to the same speed)
+and are in the range 0 to n-1 where 'n' is the total number of CPU
+cores in the system. Note that when specifying multiple cores the
+list must be enclosed in quotes. Without this option the program
+creates copies of itself to manage every core of every CPU in the
+system and automatically determines core groups. If you are running
+on an old kernel and get an error message about not being able to
+open an "affected_cpus" file then you must run this program
+separately for each group of cores that must be controlled together
+(which probably means for each physical CPU) and use this option.
.SS "Notes:"
+To have a CPU core stay at the highest clock speed to maximize performance send
+the process controlling that CPU core the SIGUSR1 signal.
+.PP
+To have a CPU core stay at the lowest clock speed to maximize battery life send
+the process controlling that CPU core the SIGUSR2 signal.
+.PP
+To resume having a CPU core's clock speed dynamically scaled send the process
+controlling that CPU core the SIGHUP signal.
+.PP
Many modern Linux systems support in-kernel cpu frequency scaling. The
cpuspeed daemon only works in conjunction with the 'userspace' frequency
scaling governor. Other governors, such as 'ondemand' and 'conservative',
@@ -135,7 +146,7 @@
.na
.nf
Carl Thompson \- cet [at] carlthompson.net
- (copyright 2002 \- 2005)
+ (copyright 2002 \- 2008)
.TP
Modernized init script, configuration file and man page:
.na
More information about the fedora-extras-commits
mailing list