summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorJack Reed <jreed@redhat.com>2012-10-31 02:23:13 (GMT)
committerJack Reed <jreed@redhat.com>2012-10-31 02:23:13 (GMT)
commit4fee5984bacb51835b21d6ccf8af0d12ab163cb9 (patch)
tree72bdc3f583421a21808b3d4b4cd81fe6f2d3e78d
parentd36218d91d98182861ddb5ebad0b37e8cca2f08b (diff)
downloadpower-management-guide-4fee5984bacb51835b21d6ccf8af0d12ab163cb9.zip
power-management-guide-4fee5984bacb51835b21d6ccf8af0d12ab163cb9.tar.gz
power-management-guide-4fee5984bacb51835b21d6ccf8af0d12ab163cb9.tar.xz
Updating and expanding Chapter 3 with features of cpupower command
-rw-r--r--en-US/CPU_Idle_States.xml6
-rw-r--r--en-US/CPU_Monitors.xml30
-rw-r--r--en-US/CPU_Power_Saving.xml44
-rw-r--r--en-US/CPUfreq_Setup.xml16
-rw-r--r--en-US/Core_Frequency_Tuning.xml198
-rw-r--r--en-US/Power_Management_Guide.xml10
-rw-r--r--en-US/Tuned.xml83
7 files changed, 297 insertions, 90 deletions
diff --git a/en-US/CPU_Idle_States.xml b/en-US/CPU_Idle_States.xml
index 5f4f7d8..5a124d4 100644
--- a/en-US/CPU_Idle_States.xml
+++ b/en-US/CPU_Idle_States.xml
@@ -43,6 +43,12 @@
</listitem>
</varlistentry>
</variablelist>
+
+ <para>
+ To view available idle states and other statistics for the CPUidle driver, run the following command:
+ </para>
+<screen><command>cpupower idle-info</command></screen>
+
<para>
Recent Intel CPUs with the "Nehalem" microarchitecture feature a new C-State, C6, which can reduce the voltage supply of a CPU to zero, but typically reduces power consumption by between 80% and 90%. The kernel in Fedora&nbsp;&PRODVER; includes optimizations for this new C-State.
</para>
diff --git a/en-US/CPU_Monitors.xml b/en-US/CPU_Monitors.xml
new file mode 100644
index 0000000..b3cbd59
--- /dev/null
+++ b/en-US/CPU_Monitors.xml
@@ -0,0 +1,30 @@
+<?xml version='1.0'?>
+<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN" "http://www.oasis-open.org/docarticle/xml/4.5/docarticlex.dtd" [
+]>
+<section id="cpu_monitors">
+ <title>CPU Monitors</title>
+ <para>
+ <application>cpupower</application> features a selection of monitors that provide idle and sleep state statistics and frequency information and report on processor topology. Some monitors are processor-specific, while others are compatible with any processor. Refer to the <command>cpupower-monitor</command> man page for details on what each monitor measures and which systems they are compatible with.
+ </para>
+ <para>
+ Use the following options with the <command>cpupower monitor</command> command:
+ </para>
+
+
+ <itemizedlist>
+ <listitem>
+ <para><command>-l</command> &mdash; list all monitors available on your system.</para>
+ </listitem>
+
+
+ <listitem>
+ <para><command>-m <replaceable>&lt;monitor1&gt;</replaceable>, <replaceable>&lt;monitor2&gt;</replaceable></command> &mdash; display specific monitors. Their identifiers can be found by running <command>-l</command>.</para>
+ </listitem>
+
+ <listitem>
+ <para><command>command</command> &mdash; display the idle statistics and CPU demands of a specific command.</para>
+ </listitem>
+
+ </itemizedlist>
+
+</section>
diff --git a/en-US/CPU_Power_Saving.xml b/en-US/CPU_Power_Saving.xml
new file mode 100644
index 0000000..e08096b
--- /dev/null
+++ b/en-US/CPU_Power_Saving.xml
@@ -0,0 +1,44 @@
+<?xml version='1.0'?>
+<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN" "http://www.oasis-open.org/docarticle/xml/4.5/docarticlex.dtd" [
+]>
+<section id="cpu_power_saving">
+ <title>CPU Power Saving Policies</title>
+ <para>
+ <application>cpupower</application> provides ways to regulate your processor's power saving policies.
+ </para>
+ <para>
+ Use the following options with the <command>cpupower set</command> command:
+ </para>
+
+
+ <variablelist>
+ <varlistentry>
+ <term>--perf-bias <replaceable>&lt;0-15&gt;</replaceable></term>
+ <listitem>
+ <para>
+ Allows software on supported Intel processors to more actively contribute to determining the balance between optimum performance and saving power. This does not override other power saving policies. Assigned values range from 0 to 15, where 0 is optimum performance and 15 is optimum power efficiency.
+ </para>
+ <para>
+ By default, this option applies to all cores. To apply it only to individual cores, add the <command>--cpu <replaceable>&lt;cpulist&gt;</replaceable></command> option.
+ </para>
+ </listitem>
+ </varlistentry>
+ <varlistentry>
+ <term>--sched-mc &lt;0|1|2&gt;</term>
+ <listitem>
+ <para>
+ Restricts the use of power by system processes to the cores in one CPU package before other CPU packages are drawn from. 0 sets no restrictions, 1 initially employs only a single CPU package, and 2 does this in addition to favouring semi-idle CPU packages for handling task wakeups.
+ </para>
+ </listitem>
+ </varlistentry>
+ <varlistentry>
+ <term>--sched-smt &lt;0|1|2&gt;</term>
+ <listitem>
+ <para>
+ Restricts the use of power by system processes to the thread siblings of one CPU core before drawing on other cores. 0 sets no restrictions, 1 initially employs only a single CPU package, and 2 does this in addition to favouring semi-idle CPU packages for handling task wakeups.
+ </para>
+ </listitem>
+ </varlistentry>
+ </variablelist>
+
+</section>
diff --git a/en-US/CPUfreq_Setup.xml b/en-US/CPUfreq_Setup.xml
index 7dec229..5a65664 100644
--- a/en-US/CPUfreq_Setup.xml
+++ b/en-US/CPUfreq_Setup.xml
@@ -19,7 +19,7 @@
<para>Before selecting and configuring a CPUfreq governor, you need to add the appropriate CPUfreq driver first.</para>
- <remark>add: how do you determine what the appropriate cpufreq driver is? that is how to choose between p4-clockmod and acpi-cpufreq?</remark>
+ <remark>add: how do you determine what the appropriate cpufreq driver is? that is how to choose between p4-clockmd and acpi-cpufreq?</remark>
<remark>Answer: always use acpi-cpufreq!</remark>
<procedure><title>How to Add a CPUfreq Driver</title>
@@ -130,15 +130,15 @@
</step>
- <step><para>Once the CPUfreq driver is set up, you can view which governor the system is currently using with:</para>
+ <!-- <step><para>Once the CPUfreq driver is set up, you can view which governor the system is currently using with:</para>
<screen><command>cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor</command></screen>
- </step>
+ </step> -->
</procedure>
<para>You can also view which governors are available for use for a specific CPU using:</para>
-<screen><command>cat /sys/devices/system/cpu/<replaceable>[cpu ID]</replaceable>/cpufreq/scaling_available_governors</command></screen>
+<screen><command>cpupower frequency-info --governors</command></screen>
<para>Some CPUfreq governors may not be available for you to use. In this case, use <command>modprobe</command> to add the necessary kernel modules that enable the specific CPUfreq governor you wish to use. These kernel modules are available in <filename>/lib/modules/<replaceable>[kernel version]</replaceable>/kernel/drivers/cpufreq/</filename>.</para>
@@ -215,13 +215,7 @@
Once a governor is listed as available for your CPU, you can enable it using:
</para>
-<screen><command>echo <replaceable>[governor]</replaceable> &gt; /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor</command></screen>
-
- <para>
- where <command><replaceable>[governor]</replaceable></command> is the type of governor, not the full name listed in <xref linkend="governor_types"/>. For example, to specify <command>cpufreq_powersave</command>, the command is:
- </para>
-<screen><command>echo powersave &gt; /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor</command></screen>
-
+<screen><command>cpupower frequency-set --governor <replaceable>[governor]</replaceable></command></screen>
</step>
</procedure>
diff --git a/en-US/Core_Frequency_Tuning.xml b/en-US/Core_Frequency_Tuning.xml
index db28608..d8e03ef 100644
--- a/en-US/Core_Frequency_Tuning.xml
+++ b/en-US/Core_Frequency_Tuning.xml
@@ -36,174 +36,204 @@
<tertiary>tuning</tertiary>
</indexterm>
- <para>Once you've chosen an appropriate CPUfreq governor, you can further tune the speed of each CPU using the tunables found in <filename>/sys/devices/system/cpu/<replaceable>[cpu ID]</replaceable>/cpufreq/</filename>. These tunables are:</para>
+<para>Once you have chosen an appropriate CPUfreq governor, you can view CPU speed and policy information with the <command>cpupower frequency-info</command> command and further tune the speed of each CPU with options for <command>cpupower frequency-set</command>.</para>
+<para>
+ For <command>cpupower frequency-info</command>, the following options are available:
+</para>
<itemizedlist>
- <listitem><para><filename>cpuinfo_min_freq</filename> &mdash; Shows the CPU's available minimum operating frequency (in KHz).</para>
+ <listitem><para><command>--freq</command> &mdash; Shows the current speed of the CPU according to the CPUfreq core, in KHz.</para>
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
- <tertiary>cpuinfo_min_freq</tertiary>
+ <tertiary>current frequency</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
- <tertiary>cpuinfo_min_freq</tertiary>
+ <tertiary>current frequency</tertiary>
</indexterm>
<indexterm>
- <primary>cpuinfo_min_freq</primary>
+ <primary>current frequency</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
</indexterm>
</listitem>
- <listitem>
-
+ <listitem><para><command>--hwfreq</command> &mdash; Shows the current speed of the CPU according to the hardware, in KHz (only available as root).</para>
+
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
- <tertiary>cpuinfo_max_freq</tertiary>
+ <tertiary>current hardware frequency</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
- <tertiary>cpuinfo_max_freq</tertiary>
+ <tertiary>current hardware frequency</tertiary>
</indexterm>
<indexterm>
- <primary>cpuinfo_max_freq</primary>
+ <primary>current hardware frequency</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
</indexterm>
-
- <para><filename>cpuinfo_max_freq</filename> &mdash; Shows the CPU's available maximum operating frequency (in KHz).</para></listitem>
+
+ </listitem>
<listitem>
+
+ <para><command>--driver</command> &mdash; Shows what CPUfreq driver is used to set the frequency on this CPU.</para>
+
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
- <tertiary>scaling_driver</tertiary>
+ <tertiary>Current driver</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
- <tertiary>scaling_driver</tertiary>
+ <tertiary>Current driver</tertiary>
</indexterm>
<indexterm>
- <primary>scaling_driver</primary>
+ <primary>Current driver</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
-</indexterm>
- <para><filename>scaling_driver</filename> &mdash; Shows what CPUfreq driver is used to set the frequency on this CPU.</para></listitem>
+</indexterm>
+
+ </listitem>
<listitem>
+
+ <para><command>--governors</command> &mdash; Shows the CPUfreq governors available in this kernel. If you wish to use a CPUfreq governor that is not listed in this file, refer to <xref linkend="enabling_a_cpufreq_governor"/> in <xref linkend="cpufreq_setup"/> for instructions on how to do so.</para>
+
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
-<tertiary>scaling_available_governors</tertiary>
+<tertiary>available governors</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
-<tertiary>scaling_available_governors</tertiary>
+<tertiary>available governors</tertiary>
</indexterm>
<indexterm>
-<primary>scaling_available_governors</primary>
+<primary>available governors</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
</indexterm>
-
- <para><filename>scaling_available_governors</filename> &mdash; Shows the CPUfreq governors available in this kernel. If you wish to use a CPUfreq governor that is not listed in this file, refer to <xref linkend="enabling_a_cpufreq_governor"/> in <xref linkend="cpufreq_setup"/> for instructions on how to do so.</para></listitem>
+
+ </listitem>
- <listitem>
-<indexterm>
+
+ <listitem>
+ <para><command>--affected-cpus</command> &mdash; Lists CPUs that require frequency coordination software.</para>
+
+<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
-<tertiary>scaling_governor</tertiary>
+<tertiary>affected-cpus</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
-<tertiary>scaling_governor</tertiary>
+<tertiary>affected-cpus</tertiary>
</indexterm>
<indexterm>
-<primary>scaling_governor</primary>
+<primary>affected-cpus</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
</indexterm>
- <para><filename>scaling_governor</filename> &mdash; Shows what CPUfreq governor is currently in use. To use a different governor, simply use <command>echo <replaceable>[governor]</replaceable> > /sys/devices/system/cpu/<replaceable>[cpu ID]</replaceable>/cpufreq/scaling_governor</command> (refer to <xref linkend="enabling_a_cpufreq_governor"/> in <xref linkend="cpufreq_setup"/> for more information).</para></listitem>
+
+ </listitem>
+
+ <listitem>
- <listitem>
+ <para><command>--policy</command> &mdash; Shows the range of the current CPUfreq policy, in KHz, and the currently active governor.</para>
+
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
-<tertiary>cpuinfo_cur_freq</tertiary>
+<tertiary>current policy</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
-<tertiary>cpuinfo_cur_freq</tertiary>
+<tertiary>current policy</tertiary>
</indexterm>
<indexterm>
-<primary>cpuinfo_cur_freq</primary>
+<primary>current policy</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
-</indexterm>
- <para><filename>cpuinfo_cur_freq</filename> &mdash; Shows the current speed of the CPU (in KHz).</para></listitem>
+</indexterm>
+
+ </listitem>
+
+ <listitem>
+
+ <para><command>--hwlimits</command> &mdash; Lists available frequencies for the CPU, in KHz.</para>
- <listitem>
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
-<tertiary>scaling_available_frequencies</tertiary>
+<tertiary>available frequencies</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
-<tertiary>scaling_available_frequencies</tertiary>
+<tertiary>available frequencies</tertiary>
</indexterm>
<indexterm>
-<primary>scaling_available_frequencies</primary>
+<primary>available frequencies</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
</indexterm>
- <para><filename>scaling_available_frequencies</filename> &mdash; Lists available frequencies for the CPU, in KHz.</para></listitem>
+
+</listitem>
+
+ </itemizedlist>
- <listitem>
+<para>
+ For <command>cpupower frequency-set</command>, the following options are available:
+</para>
+ <itemizedlist>
+
+ <listitem>
+ <para><command>--min <replaceable>&lt;freq&gt;</replaceable></command> and <command>--max <replaceable>&lt;freq&gt;</replaceable></command> &mdash; Set the <firstterm>policy limits</firstterm> of the CPU, in KHz.</para>
+
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
-<tertiary>scaling_min_freq</tertiary>
+<tertiary>minimum frequency</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
-<tertiary>scaling_min_freq</tertiary>
+<tertiary>minimum frequency</tertiary>
</indexterm>
<indexterm>
-<primary>scaling_min_freq</primary>
+<primary>minimum frequency</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
-</indexterm>
- <para><filename>scaling_min_freq</filename> and <filename>scaling_max_freq</filename> &mdash; Sets the <firstterm>policy limits</firstterm> of the CPU, in KHz.</para>
+</indexterm>
<important>
<title>Important</title>
@@ -225,61 +255,107 @@
<tertiary>tuning</tertiary>
</indexterm>
- <para>When setting policy limits, you should set <filename>scaling_max_freq</filename> before <filename>scaling_min_freq</filename>.</para>
+ <para>When setting policy limits, you should set <filename>--max</filename> before <filename>--min</filename>.</para>
</important>
</listitem>
+
+ <listitem>
+
+ <para><command>--freq <replaceable>&lt;freq&gt;</replaceable></command> &mdash; Set a specific clock speed for the CPU, in KHz. You can only set a speed within the policy limits of the CPU (as per <filename>--min</filename> and <filename>--max</filename>).</para>
+
+<indexterm>
+<primary>CPUfreq</primary>
+<secondary>tuning</secondary>
+<tertiary>specific frequency</tertiary>
+</indexterm>
- <listitem>
+<indexterm>
+<primary>tuning</primary>
+<secondary>CPUfreq</secondary>
+<tertiary>specific frequency</tertiary>
+</indexterm>
+
+<indexterm>
+<primary>specific frequency</primary>
+<secondary>CPUfreq</secondary>
+<tertiary>tuning</tertiary>
+</indexterm>
+
+
+ </listitem>
+
+ <listitem>
+
+ <para><command>--governor <replaceable>&lt;gov&gt;</replaceable></command> &mdash; Set a new CPUfreq governor.</para>
+
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
-<tertiary>affected_cpus</tertiary>
+<tertiary>specific frequency</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
-<tertiary>affected_cpus</tertiary>
+<tertiary>specific frequency</tertiary>
</indexterm>
<indexterm>
-<primary>affected_cpus</primary>
+<primary>specific frequency</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
-</indexterm>
- <para><filename>affected_cpus</filename> &mdash; Lists CPUs that require frequency coordination software.</para></listitem>
-
- <listitem>
+</indexterm>
+
+ </listitem>
+
+
+ <!-- <listitem>
+
+ <para><filename>scaling_governor</filename> &mdash; Shows what CPUfreq governor is currently in use. To use a different governor, simply use <command>echo <replaceable>[governor]</replaceable> > /sys/devices/system/cpu/<replaceable>[cpu ID]</replaceable>/cpufreq/scaling_governor</command> (refer to <xref linkend="enabling_a_cpufreq_governor"/> in <xref linkend="cpufreq_setup"/> for more information).</para>
+
<indexterm>
<primary>CPUfreq</primary>
<secondary>tuning</secondary>
-<tertiary>scaling_setspeed</tertiary>
+<tertiary>scaling_governor</tertiary>
</indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>CPUfreq</secondary>
-<tertiary>scaling_setspeed</tertiary>
+<tertiary>scaling_governor</tertiary>
</indexterm>
<indexterm>
-<primary>scaling_setspeed</primary>
+<primary>scaling_governor</primary>
<secondary>CPUfreq</secondary>
<tertiary>tuning</tertiary>
-</indexterm>
- <para><filename>scaling_setspeed</filename> &mdash; Used to change the clock speed of the CPU, in KHz. You can only set a speed within the policy limits of the CPU (as per <filename>scaling_min_freq</filename> and <filename>scaling_max_freq</filename>).</para></listitem>
+</indexterm>
+
+ </listitem> -->
+
+
+
+
</itemizedlist>
+
+ <note>
+ <title>Alternative to <command>cpupower</command></title>
+ <para>
+ If you do not have the <package>kernel-tools</package> package installed, CPUfreq settings can be viewed in the tunables found in <filename>/sys/devices/system/cpu/<replaceable>[cpuid]</replaceable>/cpufreq/</filename>. Settings and values can be changed by writing to these tunables. For example, to set the minimum clock speed of cpu0 to 360 KHz, use:
+<screen><command>echo 360000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq</command></screen>
+ </para>
+ </note>
- <para>To view the current value of each tunable, use <command>cat <replaceable>[tunable]</replaceable></command>. For example, to view the current speed of cpu0 (in KHz), use:</para>
+ <!-- <para>To view the current value of each tunable, use <command>cat <replaceable>[tunable]</replaceable></command>. For example, to view the current speed of cpu0 (in KHz), use:</para>
<para><command>cat /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_cur_freq</command>.</para>
<para>To change the value of each tunable, use <command>echo <replaceable>[value]</replaceable> > /sys/devices/system/cpu/<replaceable>[cpu ID]</replaceable>/cpufreq/<replaceable>[tunable]</replaceable></command>. For example, to set the minimum clock speed of cpu0 to 360 KHz, use:</para>
- <para><command>echo 360000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq</command></para>
+ <para><command>echo 360000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq</command></para> -->
</section> \ No newline at end of file
diff --git a/en-US/Power_Management_Guide.xml b/en-US/Power_Management_Guide.xml
index 31ce9aa..68a13dc 100644
--- a/en-US/Power_Management_Guide.xml
+++ b/en-US/Power_Management_Guide.xml
@@ -34,9 +34,19 @@
<chapter id="Core_Infrastructure">
<title>Core Infrastructure and Mechanics</title>
+
+ <important>
+ <title>Package required for <command>cpupower</command></title>
+ <para>
+ To use the <command>cpupower</command> command featured in this chapter, ensure you have the <package>kernel-tools</package> package installed.
+ </para>
+ </important>
+
<xi:include href="CPU_Idle_States.xml" xmlns:xi="http://www.w3.org/2001/XInclude" />
<xi:include href="CPUfreq_Governors.xml" xmlns:xi="http://www.w3.org/2001/XInclude" />
<!-- <xi:include href="Video_Driver.xml" xmlns:xi="http://www.w3.org/2001/XInclude" /> -->
+ <xi:include href="CPU_Monitors.xml" xmlns:xi="http://www.w3.org/2001/XInclude" />
+ <xi:include href="CPU_Power_Saving.xml" xmlns:xi="http://www.w3.org/2001/XInclude" />
<xi:include href="Suspend_Resume.xml" xmlns:xi="http://www.w3.org/2001/XInclude" />
<xi:include href="Tickless_Kernel.xml" xmlns:xi="http://www.w3.org/2001/XInclude" />
<xi:include href="ASPM.xml" xmlns:xi="http://www.w3.org/2001/XInclude" />
diff --git a/en-US/Tuned.xml b/en-US/Tuned.xml
index 58665e6..2bbe92c 100644
--- a/en-US/Tuned.xml
+++ b/en-US/Tuned.xml
@@ -217,70 +217,117 @@ enabled=True</programlisting>
</para>
<screen><command>tuned-adm off</command></screen>
<para>
- When you first install <application>tuned</application>, the <literal>default</literal> profile will be active. Fedora&nbsp;&PRODVER; also includes the following predefined profiles:
+ When you first install <application>tuned</application>, the <literal>balanced</literal> profile will be active. The following predefined profiles are available after installing <package>tuned</package>:
</para>
<variablelist>
+
+ <varlistentry>
+ <term>balanced</term>
+ <listitem>
+ <para>
+ the default power-saving profile. It enables CPU and disk plugins of <application>tuned</application> and activates the <command>ondemand</command> governor.
+ </para>
+ </listitem>
+ </varlistentry>
+
<varlistentry>
- <term>default</term>
+ <term>powersave</term>
<listitem>
<para>
- the default power-saving profile. It has the lowest impact on power saving of the available profiles and only enables CPU and disk plugins of <application>tuned</application>.
+ a profile for maximum power saving performance. It enables USB autosuspend, WiFi power saving, and ALPM power saving for SATA host adapters (refer to <xref linkend="ALPM"/>). It also schedules multi-core power savings for systems with a low wakeup rate and activates the <command>ondemand</command> governor.
</para>
</listitem>
</varlistentry>
+
<varlistentry>
- <term>desktop-powersave</term>
+ <term>throughput-performance</term>
<listitem>
<para>
- a power-saving profile directed at desktop systems. Enables ALPM power saving for SATA host adapters (refer to <xref linkend="ALPM"/>) as well as the CPU, Ethernet, and disk plugins of <application>tuned</application>.
+ a server profile for typical throughput performance tuning. It disables <application>tuned</application> and <application>ktune</application> power saving mechanisms, enables <application>sysctl</application> settings that improve the throughput performance of your disk and network I/O, and switches to the <application>deadline scheduler</application>.
</para>
</listitem>
</varlistentry>
+
<varlistentry>
- <term>server-powersave</term>
+ <term>latency-performance</term>
<listitem>
<para>
- a power-saving profile directed at server systems. Enables ALPM powersaving for SATA host adapters and activates the CPU and disk plugins of <application>tuned</application>.
+ a server profile for typical latency performance tuning. it disables <application>tuned</application> and <application>ktune</application> power saving mechanisms and enables <application>sysctl</application> settings that improve the latency performance of your network I/O.
</para>
</listitem>
</varlistentry>
+
<varlistentry>
- <term>laptop-ac-powersave</term>
+ <term>virtual-guest</term>
<listitem>
<para>
- a medium-impact power-saving profile directed at laptops running on AC. Enables ALPM powersaving for SATA host adapters, Wi-Fi power saving, as well as the CPU, Ethernet, and disk plugins of <application>tuned</application>.
+ a profile designed for virtual guests based on the enterprise-storage profile that, among other tasks, decreases virtual memory swappiness.
</para>
</listitem>
</varlistentry>
+
<varlistentry>
- <term>laptop-battery-powersave</term>
+ <term>virtual-host</term>
<listitem>
<para>
- a high-impact power-saving profile directed at laptops running on battery. It activates all power saving mechanisms from the previous profiles, plus it enables the multi-core power-savings scheduler for low wakeup systems and makes sure that the ondemand governor is active and that AC97 audio power-saving is enabled. On compatible Asus EEEPCs, it enables SHE support in the CPU tuning plugin. You can use this profile to save the maximum amount of power on any kind of system, not only laptops on battery power. The trade-off for this profile is a noticeable impact on performance, specifically latency of disk and network I/O.
+ a profile designed for virtual hosts based on the enterprise-storage profile that, among other tasks, decreases virtual memory swappiness and is more aggressive in its writeback of dirty pages.
</para>
</listitem>
</varlistentry>
+
+ </variablelist>
+
+ <para>
+ Additional predefined profiles are available by installing the <package>tuned-profiles-compat</package> package:
+ </para>
+
+ <variablelist>
+
<varlistentry>
- <term>spindown-disk</term>
+ <term>default</term>
<listitem>
<para>
- a power-saving profile for machines with classic HDDs to maximize spindown time. It disables the <application>tuned</application> power savings mechanism, disables USB autosuspend, disables Bluetooth, enables Wi-Fi power saving, disables logs syncing, increases disk write-back time, and lowers disk swappiness. All partitions are remounted with <option>noatime</option>.
+ this has the lowest impact on power saving of the available profiles and only enables CPU and disk plugins of <application>tuned</application>.
</para>
</listitem>
</varlistentry>
- <varlistentry>
- <term>throughput-performance</term>
+ <varlistentry>
+ <term>desktop-powersave</term>
<listitem>
<para>
- a server profile for typical throughput performance tuning. It disables <application>tuned</application> and <application>ktune</application> power saving mechanisms, enables <application>sysctl</application> settings that improve the throughput performance of your disk and network I/O, and switches to the <application>deadline scheduler</application>.
+ a power-saving profile directed at desktop systems. Enables ALPM power saving for SATA host adapters (refer to <xref linkend="ALPM"/>) as well as the CPU, Ethernet, and disk plugins of <application>tuned</application>.
</para>
</listitem>
</varlistentry>
<varlistentry>
- <term>latency-performance</term>
+ <term>server-powersave</term>
<listitem>
<para>
- a server profile for typical latency performance tuning. it disables <application>tuned</application> and <application>ktune</application> power saving mechanisms and enables <application>sysctl</application> settings that improve the latency performance of your network I/O.
+ a power-saving profile directed at server systems. Enables ALPM powersaving for SATA host adapters and activates the CPU and disk plugins of <application>tuned</application>.
+ </para>
+ </listitem>
+ </varlistentry>
+ <varlistentry>
+ <term>laptop-ac-powersave</term>
+ <listitem>
+ <para>
+ a medium-impact power-saving profile directed at laptops running on AC. Enables ALPM powersaving for SATA host adapters, Wi-Fi power saving, as well as the CPU, Ethernet, and disk plugins of <application>tuned</application>.
+ </para>
+ </listitem>
+ </varlistentry>
+ <varlistentry>
+ <term>laptop-battery-powersave</term>
+ <listitem>
+ <para>
+ a high-impact power-saving profile directed at laptops running on battery. It activates all power saving mechanisms from the previous profiles, plus it enables the multi-core power-savings scheduler for low wakeup systems and makes sure that the ondemand governor is active and that AC97 audio power-saving is enabled. On compatible Asus EEEPCs, it enables SHE support in the CPU tuning plugin. You can use this profile to save the maximum amount of power on any kind of system, not only laptops on battery power. The trade-off for this profile is a noticeable impact on performance, specifically latency of disk and network I/O.
+ </para>
+ </listitem>
+ </varlistentry>
+ <varlistentry>
+ <term>spindown-disk</term>
+ <listitem>
+ <para>
+ a power-saving profile for machines with classic HDDs to maximize spindown time. It disables the <application>tuned</application> power savings mechanism, disables USB autosuspend, disables Bluetooth, enables Wi-Fi power saving, disables logs syncing, increases disk write-back time, and lowers disk swappiness. All partitions are remounted with <option>noatime</option>.
</para>
</listitem>
</varlistentry>