[system-administrators-guide] Applying improvements to OProfile

commit 29a16eb6e18cac53164a85cbd99f75bdd3aae2b6 Author: Stephen Wadeley <swadeley(a)redhat.com&gt; Date: Sat Dec 13 14:58:01 2014 +0100 Applying improvements to OProfile from upstream version, after proofreading and SME feedback en-US/OProfile.xml | 1082 +++++++++++++++++++++++++++++++++------------------- 1 files changed, 694 insertions(+), 388 deletions(-) --- diff --git a/en-US/OProfile.xml b/en-US/OProfile.xml index dd9f724..e5e2ffa 100644 --- a/en-US/OProfile.xml +++ b/en-US/OProfile.xml @@ -4,20 +4,18 @@ <chapter id="ch-OProfile"> <title>OProfile</title> - <indexterm - significance="normal"> + <indexterm> <primary>system analysis</primary> <secondary>OProfile</secondary> <see>OProfile</see> </indexterm> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> </indexterm> <para>OProfile is a low overhead, system-wide performance monitoring tool. It uses the performance monitoring hardware on the processor to retrieve information about the kernel and executables on the system, such as when memory is referenced, the number of L2 cache requests, and the number of hardware interrupts received. On a &MAJOROS; system, the <filename>oprofile</filename> package must be installed to use this tool.</para> - <para>Many processors include dedicated performance monitoring hardware. This hardware makes it possible to detect when certain events happen (such as the requested data not being in cache). The hardware normally takes the form of one or more <firstterm>counters</firstterm> that are incremented each time an event takes place. When the counter value, essentially rolls over, an interrupt is generated, making it possible to control the amount of detail (and therefore, overhead) produced by performance monitoring.</para> + <para>Many processors include dedicated performance monitoring hardware. This hardware makes it possible to detect when certain events happen (such as the requested data not being in cache). The hardware normally takes the form of one or more <firstterm>counters</firstterm> that are incremented each time an event takes place. When the counter value increments, an interrupt is generated, making it possible to control the amount of detail (and therefore, overhead) produced by performance monitoring.</para> <para>OProfile uses this hardware (or a timer-based substitute in cases where performance monitoring hardware is not present) to collect <firstterm>samples</firstterm> of performance-related data each time a counter generates an interrupt. These samples are periodically written out to disk; later, the data contained in these samples can then be used to generate reports on system-level and application-level performance.</para> - <para>OProfile is a useful tool, but be aware of some limitations when using it:</para> + <para>Be aware of the following limitations when using OProfile:</para> <itemizedlist> <listitem> <para> @@ -38,7 +36,7 @@ </listitem> <listitem> <para> - <emphasis>Hardware performance counters do not work on guest virtual machines</emphasis> — Because the hardware performance counters are not available on virtual systems, you need to use the <computeroutput>timer</computeroutput> mode. Run the command <command>opcontrol --deinit</command>, and then execute <command>modprobe oprofile timer=1</command> to enable the <computeroutput>timer</computeroutput> mode.</para> + <emphasis>Hardware performance counters do not work on guest virtual machines</emphasis> — Because the hardware performance counters are not available on virtual systems, you need to use the <computeroutput>timer</computeroutput> mode. Enter the command <command>opcontrol --deinit</command>, and then execute <command>modprobe oprofile timer=1</command> to enable the <computeroutput>timer</computeroutput> mode.</para> </listitem> <listitem> <para> @@ -48,14 +46,13 @@ <section id="s1-oprofile-overview-tools"> <title>Overview of Tools</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>overview of tools</secondary> </indexterm> <para> <xref - linkend="tb-oprofile-tools"/> provides a brief overview of the tools provided with the <filename>oprofile</filename> package.</para> + linkend="tb-oprofile-tools"/> provides a brief overview of the most commonly used tools provided with the <filename>oprofile</filename> package.</para> <table id="tb-oprofile-tools"> <title>OProfile Commands</title> @@ -116,6 +113,14 @@ </row> <row> <entry> + <command>operf</command> + </entry> + <entry> + <para>Recommended tool to be used in place of <command>opcontrol</command> for profiling. See <xref linkend="s1-using-operf"/> for details.</para> For differences between <command>operf</command> and <command>opcontrol</command> see <xref linkend="s2-operf_vs_opcontrol"/>. + </entry> + </row> + <row> + <entry> <command>opreport</command> </entry> <entry> @@ -134,59 +139,203 @@ </tbody> </tgroup> </table> + <section id="s2-operf_vs_opcontrol"> + <title>operf vs. opcontrol</title> + <para> + There are two mutually exclusive methods for collecting profiling data with OProfile. You can either use the newer and preferred <command>operf</command> or the <command>opcontrol</command> tool. + </para> + <bridgehead>operf</bridgehead> + <para> + This is the recommended mode for profiling. The <command>operf</command> tool uses the Linux Performance Events Subsystem, and therefore does not require the <emphasis>oprofile</emphasis> kernel driver. The <command>operf</command> tool allows you to target your profiling more precisely, as a single process or system-wide, and also allows OProfile to co-exist better with other tools using the performance monitoring hardware on your system. Unlike <command>opcontrol</command>, it can be used without the <systemitem class="username">root</systemitem> privileges. However, <command>operf</command> is also capable of system-wide operations with use of the <option>--system-wide</option> option, where root authority is required. + </para> + <para> + With <command>operf</command>, there is no initial setup needed. You can invoke <command>operf</command> with command-line options to specify your profiling settings. After that, you can run the OProfile post-processing tools described in <xref linkend="s1-oprofile-analyzing-data"/>. See <xref linkend="s1-using-operf"/> for further information. + </para> + <bridgehead>opcontrol</bridgehead> + <para> + This mode consists of the <command>opcontrol</command> shell script, the <systemitem class="daemon">oprofiled</systemitem> daemon, and several post-processing tools. The <command>opcontrol</command> command is used for configuring, starting, and stopping a profiling session. An OProfile kernel driver, usually built as a kernel module, is used for collecting samples, which are then recorded into sample files by <systemitem class="daemon">oprofiled</systemitem>. You can use legacy mode only if you have <systemitem class="username">root</systemitem> privileges. In certain cases, such as when you need to sample areas with disabled interrupt request (IRQ), this is a better alternative. + </para> + <para> + Before OProfile can be run in legacy mode, it must be configured as shown in <xref linkend="s1-oprofile-configuring"/>. These settings are then applied when starting OProfile (<xref linkend="s1-oprofile-starting"/>). + </para> + </section> + </section> + <section id="s1-using-operf"> + <title>Using operf</title> + <para> + <command>operf</command> is the recommended profiling mode that does not require initial setup before starting. All settings are specified as command-line options and there is no separate command to start the profiling process. To stop <command>operf</command>, press Ctrl+C. The typical <command>operf</command> command syntax looks as follows: + </para> + <synopsis><command>operf</command> <replaceable>options</replaceable> <replaceable>range</replaceable> <replaceable>command</replaceable> <replaceable>args</replaceable></synopsis> + <para> + Replace <replaceable>options</replaceable> with the desired command-line options to specify your profiling settings. Full set of options is described in <systemitem>operf(1)</systemitem> manual page. Replace <replaceable>range</replaceable> with one of the following: + </para> + <para> + <option>--system-wide</option> - this setting allows for global profiling, see <xref linkend="using_operf_system-wide"/> + </para> + <para> + <option>--pid=<replaceable>PID</replaceable></option> - this is to profile a running application, where <replaceable>PID</replaceable> is the process ID of the process you want to profile. + </para> + <para> + With <replaceable>command</replaceable> and <replaceable>args</replaceable>, you can define a specific command or application to be profiled, and also the input arguments that this command or application requires. Either <replaceable>command</replaceable>, <option>--pid</option> or <option>--system-wide</option> is required, but these cannot be used simultaneously. + </para> + <para> + When you invoke <command>operf</command> on a command line without setting the <replaceable>range</replaceable> option, data will be collected for the children processes. + </para> + <note id="using_operf_system-wide"> + <title>Using <command>operf</command> in System-wide Mode</title> + <para> + To run <command>operf</command> <option>--system-wide</option>, you need <systemitem class="username">root</systemitem> authority. When finished profiling, you can stop <command>operf</command> with <literal>Ctrl+C</literal>. + </para> + <para> + If you run <command>operf</command> <option>--system-wide</option> as a background process (with <literal>&amp;</literal>), stop it in a controlled manner in order to process the collected profile data. For this purpose, use: + </para> + <synopsis><command>kill -SIGINT operf-PID </command></synopsis> + <para> + When running <command>operf</command> <option>--system-wide</option>, it is recommended that your current working directory is <filename class="directory">/root</filename> or a subdirectory of <filename class="directory">/root</filename> so that sample data files are not stored in locations accessible by regular users. + </para> + </note> + <section + id="s2-operf-kernel"> + <title>Specifying the Kernel</title> + <para> + To monitor the kernel, execute the following command: + </para> + <synopsis><command>operf</command> <option>--vmlinux</option>=<replaceable>vmlinux_path</replaceable></synopsis> + <para> + With this option, you can specify a path to a vmlinux file that matches the running kernel. Kernel samples will be attributed to this binary, allowing post-processing tools to attribute samples to the appropriate kernel symbols. If this option is not specified, all kernel samples will be attributed to a pseudo binary named "no-vmlinux". + </para> + </section> + <section id="s2-operf-events"> + <title>Setting Events to Monitor</title> + <para> + Most processors contain counters, which are used by OProfile to monitor specific events. As shown in <xref linkend="tb-oprofile-processors"/>, the number of counters available depends on the processor. + </para> + <para> + The events for each counter can be configured via the command line or with a graphical interface. For more information on the graphical interface, see <xref linkend="s1-oprofile-gui"/>. If the counter cannot be set to a specific event, an error message is displayed. + </para> + <note> + <title>Older Processors and operf</title> + <para> + Some older processor models are not supported by the underlying Linux Performance Events Subsystem kernel and therefore are not supported by <command>operf</command>. If you receive this message: + <screen> +Your kernel's Performance Events Subsystem does not support your processor type + </screen> + when attempting to use <command>operf</command>, try profiling with <command>opcontrol</command> to see if your processor type may be supported by OProfile's legacy mode. + </para> + </note> + <note> + <title>Using operf on Virtual Systems</title> + <para> + Since hardware performance counters are not available on guest virtual machines, you have to enable <emphasis>timer</emphasis> mode to use <application>operf</application> on virtual systems. To do so, type as <systemitem class="username">root</systemitem>: + </para> + <synopsis><command>opcontrol</command> <option>--deinit</option></synopsis> + <synopsis><command>modprobe</command> <command>oprofile</command> <option>timer=1</option></synopsis> + </note> + + <para> + To set the event for each configurable counter via the command line, use: + </para> + <synopsis><command>operf</command> <option>--events</option>=<replaceable>event1</replaceable>,<replaceable>event2</replaceable>&hellip;</synopsis> + <para> + Here, pass a comma-separated list of event specifications for profiling. Each event specification is a colon-separated list of attributes in the following form: + </para> + <synopsis><replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable>:<replaceable>unit-mask</replaceable>:<replaceable>kernel</replaceable>:<replaceable>user</replaceable></synopsis> + <para> + <xref linkend="tab_event_specifications"/> summarizes these options. The last three values are optional, if you omit them, they will be set to their default values. Note that certain events do require a unit mask. + </para> + <table id="tab_event_specifications"> + <title>Event Specifications</title> + <tgroup cols='2' align='left'> + <thead> + <row> + <entry>Specification</entry> + <entry>Description</entry> + </row> + </thead> + <tbody> + <row> + <entry><replaceable>event-name</replaceable></entry> + <entry> The exact symbolic event name taken from <command>ophelp</command></entry> + </row> + <row> + <entry><replaceable>sample-rate</replaceable></entry> + <entry>The number of events to wait before sampling again. The smaller the count, the more frequent the samples. For events that do not happen frequently, a lower count may be needed to capture a statistically significant number of event instances. On the other hand, sampling too frequently can overload the system. By default, OProfile uses a time-based event set, which creates a sample every 100,000 clock cycles per processor. </entry> + </row> + <row> + <entry><replaceable>unit-mask</replaceable></entry> + <entry> + Unit masks, which further define the event, are listed in <command>ophelp</command>. + You can insert either a hexadecimal value, beginning with "0x", or a string that matches the first word of the unit mask description in <command>ophelp</command>. Definition by name is valid only for unit masks having "extra:" parameters, as shown by the output of <command>ophelp</command>. This type of unit mask cannot be defined with a hexadecimal value. Note that on certain architectures, there can be multiple unit masks with the same hexadecimal value. In that case they have to be specified by their names only. + </entry> + </row> + <row> + <entry><replaceable>kernel</replaceable></entry> + <entry>Specifies whether to profile kernel code (insert <literal>0</literal> or <literal>1</literal>(default))</entry> + </row> + <row> + <entry><replaceable>user</replaceable></entry> + <entry>Specifies whether to profile user-space code (insert <literal>0</literal> or <literal>1</literal> (default))</entry> + </row> + </tbody> + </tgroup> + </table> + <para> + The events available vary depending on the processor type. When no event specification is given, the default event for the running processor type will be used for profiling. See <xref linkend="tb-oprofile-default-events"/> for a list of these default events. To determine the events available for profiling, use the <command>ophelp</command> command. + </para> + <synopsis><command>ophelp</command></synopsis> + </section> + <section id="s2-operf-categorization"> + <title>Categorization of Samples</title> + <para> + The <option>--separate-thread</option> option categorizes samples by thread group ID (tgid) and thread ID (tid). This is useful for seeing per-thread samples in multi-threaded applications. When used in conjunction with the <option>--system-wide</option> option, <option>--separate-thread</option> is also useful for seeing per-process (i.e., per-thread group) samples for the case where multiple processes are executing the same program during a profiling run. + </para> + <para> + The <option>--separate-cpu</option> option categorizes samples by CPU. + </para> + </section> </section> <section id="s1-oprofile-configuring"> - <title>Configuring OProfile</title> - <indexterm - significance="normal"> + <title>Configuring OProfile Using Legacy Mode</title> + <indexterm> <primary>OProfile</primary> <secondary>configuring</secondary> </indexterm> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> - <secondary> - <command>opcontrol</command> - </secondary> + <secondary><command>opcontrol</command></secondary> </indexterm> - <indexterm - significance="normal"> - <primary> - <command>opcontrol</command> - </primary> + <indexterm> + <primary><command>opcontrol</command></primary> <see>OProfile</see> </indexterm> - <para>Before OProfile can be run, it must be configured. At a minimum, selecting to monitor the kernel (or selecting not to monitor the kernel) is required. The following sections describe how to use the <command>opcontrol</command> utility to configure OProfile. As the <command>opcontrol</command> commands are executed, the setup options are saved to the <filename>/root/.oprofile/daemonrc</filename> file.</para> + <para> + Before OProfile can be run in legacy mode, it must be configured. At a minimum, selecting to monitor the kernel (or selecting not to monitor the kernel) is required. The following sections describe how to use the <command>opcontrol</command> utility to configure OProfile. As the <command>opcontrol</command> commands are executed, the setup options are saved to the <filename>/root/.oprofile/daemonrc</filename> file. + </para> <section id="s2-oprofile-kernel"> <title>Specifying the Kernel</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>monitoring the kernel</secondary> </indexterm> <para>First, configure whether OProfile should monitor the kernel. This is the only configuration option that is required before starting OProfile. All others are optional.</para> - <para>To monitor the kernel, execute the following command as root:</para> - <indexterm - significance="normal"> + <para>To monitor the kernel, execute the following command as <systemitem class="username">root</systemitem>: + </para> + <indexterm> <primary>OProfile</primary> - <secondary> - <command>opcontrol</command> - </secondary> - <tertiary> - <option>--vmlinux=</option> - </tertiary> + <secondary><command>opcontrol</command></secondary> + <tertiary><option>--vmlinux=</option></tertiary> </indexterm> <!-- TBD6: /usr/lib/debug is an obsolete directory for Fedora 12 --> <screen>~]#&#160;<command>opcontrol --setup --vmlinux=/usr/lib/debug/lib/modules/`uname -r`/vmlinux</command></screen> - <note> + <important> <title>Install the debuginfo package</title> - <para>The <package>debuginfo</package> package for the kernel must be installed (which contains the uncompressed kernel) in order to monitor the kernel.</para> - </note> - <para>To configure OProfile not to monitor the kernel, execute the following command as root:</para> - <indexterm - significance="normal"> + <para>In order to monitor the kernel, the <package>debuginfo</package> package which contains the uncompressed kernel must be installed. + </para> + </important> + <para>To configure OProfile not to monitor the kernel, execute the following command as <systemitem class="username">root</systemitem>:</para> + <indexterm> <primary>OProfile</primary> <secondary> <command>opcontrol</command> @@ -198,14 +347,13 @@ <screen>~]#&#160;<command>opcontrol --setup --no-vmlinux</command></screen> <para>This command also loads the <computeroutput>oprofile</computeroutput> kernel module, if it is not already loaded, and creates the <filename>/dev/oprofile/</filename> directory, if it does not already exist. See <xref linkend="s1-oprofile-dev-oprofile"/> for details about this directory.</para> - <para>Setting whether samples should be collected within the kernel only changes what data is collected, not how or where the collected data is stored. To generate different sample files for the kernel and application libraries, refer to <xref + <para>Setting whether samples should be collected within the kernel only changes what data is collected, not how or where the collected data is stored. To generate different sample files for the kernel and application libraries, see <xref linkend="s2-oprofile-starting-separate"/>.</para> </section> <section id="s2-oprofile-events"> <title>Setting Events to Monitor</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>events</secondary> <tertiary>setting</tertiary> @@ -232,222 +380,297 @@ <thead> <row> <entry> - Processor - </entry> + Processor + </entry> <entry> <command>cpu_type</command> </entry> <entry> - Number of Counters - </entry> + Number of Counters + </entry> </row> </thead> <tbody> <row> <entry> - AMD64 - </entry> + AMD64 + </entry> <entry> - x86-64/hammer - </entry> + x86-64/hammer + </entry> <entry> - 4 - </entry> + 4 + </entry> </row> <row> <entry> - AMD Athlon - </entry> + AMD Family 10h + </entry> <entry> - i386/athlon - </entry> + x86-64/family10 + </entry> <entry> - 4 - </entry> + 4 + </entry> </row> <row> <entry> - AMD Family 10h - </entry> + AMD Family 11h + </entry> <entry> - x86-64/family10 - </entry> + x86-64/family11 + </entry> <entry> - 4 - </entry> + 4 + </entry> </row> <row> <entry> - AMD Family 11h - </entry> + AMD Family 12h + </entry> <entry> - x86-64/family11 - </entry> + x86-64/family12 + </entry> <entry> - 4 - </entry> + 4 + </entry> </row> <row> <entry> - AMD Family 12h - </entry> + AMD Family 14h + </entry> <entry> - x86-64/family12 - </entry> + x86-64/family14 + </entry> <entry> - 4 - </entry> + 4 + </entry> </row> <row> <entry> - AMD Family 14h - </entry> + AMD Family 15h + </entry> <entry> - x86-64/family14 - </entry> + x86-64/family15 + </entry> <entry> - 4 - </entry> + 6 + </entry> </row> <row> <entry> - AMD Family 15h - </entry> + Applied Micro X-Gene + </entry> <entry> - x86-64/family15 - </entry> + arm/armv8-xgene + </entry> <entry> - 6 - </entry> + 4 + </entry> </row> <row> <entry> - IBM eServer System i and IBM eServer System p - </entry> + ARM Cortex A53 + </entry> <entry> - timer - </entry> + arm/armv8-ca53 + </entry> <entry> - 1 - </entry> + 6 + </entry> </row> <row> <entry> - IBM POWER4 - </entry> + ARM Cortex A57 + </entry> <entry> - ppc64/power4 - </entry> + arm/armv8-ca57 + </entry> <entry> - 8 - </entry> + 6 + </entry> </row> <row> <entry> - IBM POWER5 - </entry> + IBM eServer System i and IBM eServer System p + </entry> <entry> - ppc64/power5 - </entry> + timer + </entry> <entry> - 6 - </entry> + 1 + </entry> </row> <row> <entry> - IBM PowerPC 970 - </entry> + IBM POWER4 + </entry> <entry> - ppc64/970 - </entry> + ppc64/power4 + </entry> <entry> - 8 - </entry> + 8 + </entry> </row> <row> <entry> - IBM S/390 and IBM System z - </entry> + IBM POWER5 + </entry> <entry> - timer - </entry> + ppc64/power5 + </entry> <entry> - 1 - </entry> + 6 + </entry> </row> <row> <entry> - Intel Core i7 - </entry> + IBM PowerPC 970 + </entry> <entry> - i386/core_i7 - </entry> + ppc64/970 + </entry> <entry> - 4 - </entry> + 8 + </entry> + </row> + <row> + <entry>IBM PowerPC 970MP</entry> + <entry> ppc64/970MP</entry> + <entry>8</entry> + </row> + <row> + <entry> IBM POWER5+</entry> + <entry>ppc64/power5+</entry> + <entry>6</entry> + </row> + <row> + <entry> IBM POWER5++</entry> + <entry>ppc64/power5++</entry> + <entry>6</entry> + </row> + <row> + <entry> IBM POWER56</entry> + <entry>ppc64/power6</entry> + <entry>6</entry> + </row> + <row> + <entry> IBM POWER7</entry> + <entry>ppc64/power7</entry> + <entry>6</entry> + </row> + <row> + <entry> IBM POWER8</entry> + <entry>ppc64/power7</entry> + <entry>8</entry> </row> <row> <entry> - Intel Nehalem microarchitecture - </entry> + IBM S/390 and IBM System z + </entry> <entry> - i386/nehalem - </entry> + timer + </entry> <entry> - 4 - </entry> + 1 + </entry> </row> <row> <entry> - Intel Pentium 4 (non-hyper-threaded) - </entry> + Intel Core i7 + </entry> <entry> - i386/p4 - </entry> + i386/core_i7 + </entry> <entry> - 8 - </entry> + 4 + </entry> </row> <row> <entry> - Intel Pentium 4 (hyper-threaded) - </entry> + Intel Nehalem microarchitecture + </entry> <entry> - i386/p4-ht - </entry> + i386/nehalem + </entry> <entry> - 4 - </entry> + 4 + </entry> </row> <row> <entry> - Intel Westmere microarchitecture - </entry> + Intel Westmere microarchitecture + </entry> <entry> - i386/westmere - </entry> + i386/westmere + </entry> <entry> - 4 - </entry> + 4 + </entry> + </row> + <row> + <entry>Intel Haswell microarchitecture (non-hyper-threaded)</entry> + <entry>i386/haswell</entry> + <entry>8</entry> + </row> + <row> + <entry>Intel Haswell microarchitecture (hyper-threaded)</entry> + <entry>i386/haswell-ht</entry> + <entry>4</entry> + </row> + <row> + <entry>Intel Ivy Bridge microarchitecture (non-hyper-threaded)</entry> + <entry>i386/ivybridge</entry> + <entry>8</entry> + </row> + <row> + <entry>Intel Ivy Bridge microarchitecture (hyper-threaded)</entry> + <entry>i386/ivybridge-ht</entry> + <entry>4</entry> + </row> + <row> + <entry>Intel Sandy Bridge microarchitecture (non-hyper-threaded)</entry> + <entry>i386/sandybridge</entry> + <entry>8</entry> + </row> + <row> + <entry>Intel Sandy Bridge microarchitecture</entry> + <entry>i386/sandybridge-ht</entry> + <entry>4</entry> + </row> + <row> + <entry>Intel Broadwell microarchitecture (non-hyper-threaded)</entry> + <entry>i386/broadwell</entry> + <entry>8</entry> + </row> + <row> + <entry>Intel Broadwell microarchitecture (hyper-threaded)</entry> + <entry>i386/broadwell-ht</entry> + <entry>4</entry> + </row> + <row> + <entry>Intel Silvermont microarchitecture</entry> + <entry>i386/silvermont</entry> + <entry>2</entry> </row> <row> <entry> - TIMER_INT - </entry> + TIMER_INT + </entry> <entry> - timer - </entry> + timer + </entry> <entry> - 1 - </entry> + 1 + </entry> </row> </tbody> </tgroup> </table> <para>Use <xref - linkend="tb-oprofile-processors"/> to verify that the correct processor type was detected and to determine the number of events that can be monitored simultaneously. <computeroutput>timer</computeroutput> is used as the processor type if the processor does not have supported performance monitoring hardware.</para> + linkend="tb-oprofile-processors"/> to determine the number of events that can be monitored simultaneously for your CPU type. If the processor does not have supported performance monitoring hardware, the <computeroutput>timer</computeroutput> is used as the processor type.</para> <para>If <computeroutput>timer</computeroutput> is used, events cannot be set for any processor because the hardware does not have support for hardware performance counters. Instead, the timer interrupt is used for profiling.</para> - <para>If <computeroutput>timer</computeroutput> is not used as the processor type, the events monitored can be changed, and counter 0 for the processor is set to a time-based event by default. If more than one counter exists on the processor, the counters other than counter 0 are not set to an event by default. The default events monitored are shown in <xref + <para>If <computeroutput>timer</computeroutput> is not used as the processor type, the events monitored can be changed, and counter 0 for the processor is set to a time-based event by default. If more than one counter exists on the processor, the counters other than 0 are not set to an event by default. The default events monitored are shown in <xref linkend="tb-oprofile-default-events"/>.</para> <table id="tb-oprofile-default-events"> @@ -469,141 +692,203 @@ <thead> <row> <entry> - Processor - </entry> + Processor + </entry> <entry> - Default Event for Counter - </entry> + Default Event for Counter + </entry> <entry> - Description - </entry> + Description + </entry> </row> </thead> <tbody> <row> <entry> - AMD Athlon and AMD64 - </entry> + AMD Athlon and AMD64 + </entry> <entry> - CPU_CLK_UNHALTED - </entry> + CPU_CLK_UNHALTED + </entry> <entry> - The processor's clock is not halted - </entry> + The processor's clock is not halted + </entry> </row> <row> <entry> - AMD Family 10h, AMD Family 11h, AMD Family 12h - </entry> + AMD Family 10h, AMD Family 11h, AMD Family 12h + </entry> <entry> - CPU_CLK_UNHALTED - </entry> + CPU_CLK_UNHALTED + </entry> <entry> - The processor's clock is not halted - </entry> + The processor's clock is not halted + </entry> </row> <row> <entry> - AMD Family 14h, AMD Family 15h - </entry> + AMD Family 14h, AMD Family 15h + </entry> <entry> - CPU_CLK_UNHALTED - </entry> + CPU_CLK_UNHALTED + </entry> <entry> - The processor's clock is not halted - </entry> + The processor's clock is not halted + </entry> </row> <row> <entry> - IBM POWER4 - </entry> + Applied Micro X-Gene + </entry> <entry> - CYCLES - </entry> + CPU_CYCLES + </entry> <entry> - Processor Cycles - </entry> + Processor Cycles + </entry> </row> <row> <entry> - IBM POWER5 - </entry> + ARM Cortex A53 + </entry> <entry> - CYCLES - </entry> + CPU_CYCLES + </entry> <entry> - Processor Cycles - </entry> + Processor Cycles + </entry> </row> <row> <entry> - IBM PowerPC 970 - </entry> + ARM Cortex A57 + </entry> <entry> - CYCLES - </entry> + CPU_CYCLES + </entry> <entry> - Processor Cycles - </entry> + Processor Cycles + </entry> </row> - - - - <row> <entry> - Intel Core i7 - </entry> + IBM POWER4 + </entry> <entry> - CPU_CLK_UNHALTED - </entry> + CYCLES + </entry> <entry> - The processor's clock is not halted - </entry> + Processor Cycles + </entry> + </row> + <row> + <entry> + IBM POWER5 + </entry> + <entry> + CYCLES + </entry> + <entry> + Processor Cycles + </entry> + </row> + <row> + <entry> + IBM POWER8 + </entry> + <entry> + CYCLES + </entry> + <entry> + Processor Cycles + </entry> + </row> + <row> + <entry> + IBM PowerPC 970 + </entry> + <entry> + CYCLES + </entry> + <entry> + Processor Cycles + </entry> + </row> + <row> + <entry> + Intel Core i7 + </entry> + <entry> + CPU_CLK_UNHALTED + </entry> + <entry> + The processor's clock is not halted + </entry> </row> <row> <entry> - Intel Nehalem microarchitecture - </entry> + Intel Nehalem microarchitecture + </entry> <entry> - CPU_CLK_UNHALTED - </entry> + CPU_CLK_UNHALTED + </entry> <entry> - The processor's clock is not halted - </entry> + The processor's clock is not halted + </entry> </row> <row> <entry> - Intel Pentium 4 (hyper-threaded and non-hyper-threaded) - </entry> + Intel Pentium 4 (hyper-threaded and non-hyper-threaded) + </entry> + <entry> + GLOBAL_POWER_EVENTS + </entry> + <entry> + The time during which the processor is not stopped + </entry> + </row> + <row> + <entry> + Intel Westmere microarchitecture + </entry> + <entry> + CPU_CLK_UNHALTED + </entry> + <entry> + The processor's clock is not halted + </entry> + </row> + <row> + <entry> + Intel Broadwell microarchitecture + </entry> <entry> - GLOBAL_POWER_EVENTS - </entry> + CPU_CLK_UNHALTED + </entry> <entry> - The time during which the processor is not stopped - </entry> + The processor's clock is not halted + </entry> </row> <row> <entry> - Intel Westmere microarchitecture - </entry> + Intel Silvermont microarchitecture + </entry> <entry> - CPU_CLK_UNHALTED - </entry> + CPU_CLK_UNHALTED + </entry> <entry> - The processor's clock is not halted - </entry> + The processor's clock is not halted + </entry> </row> <row> <entry> - TIMER_INT - </entry> + TIMER_INT + </entry> <entry> - (none) - </entry> + (none) + </entry> <entry> - Sample for each timer interrupt - </entry> + Sample for each timer interrupt + </entry> </row> </tbody> </tgroup> @@ -611,21 +896,30 @@ <para>The number of events that can be monitored at one time is determined by the number of counters for the processor. However, it is not a one-to-one correlation; on some processors, certain events must be mapped to specific counters. To determine the number of counters available, execute the following command:</para> <screen>~]#&#160;<command>ls -d /dev/oprofile/[0-9]*</command></screen> <para>The events available vary depending on the processor type. To determine the events available for profiling, execute the following command as root (the list is specific to the system's processor type):</para> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary> <command>ophelp</command> </secondary> </indexterm> - <indexterm - significance="normal"> - <primary> - <command>ophelp</command> - </primary> + <indexterm> + <primary><command>ophelp</command></primary> </indexterm> <screen>~]#&#160;<command>ophelp</command></screen> - <para>The events for each counter can be configured via the command line or with a graphical interface. For more information on the graphical interface, refer to <xref + <note> + <title>Make sure that OProfile is configured</title> + <para> + Unless OProfile is properly configured, <command>ophelp</command> fails with the following error message: + </para> + <screen>Unable to open cpu_type file for reading +Make sure you have done opcontrol --init +cpu_type 'unset' is not valid +you should upgrade oprofile or force the use of timer mode</screen> + <para> + To configure OProfile, follow the instructions in <xref linkend="s1-oprofile-configuring" />. + </para> + </note> + <para>The events for each counter can be configured via the command line or with a graphical interface. For more information on the graphical interface, see <xref linkend="s1-oprofile-gui"/>. If the counter cannot be set to a specific event, an error message is displayed.</para> <para>To set the event for each configurable counter via the command line, use <command>opcontrol</command>:</para> <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable></command></screen> @@ -633,39 +927,39 @@ <section id="s3-oprofile-events-sampling"> <title>Sampling Rate</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>events</secondary> <tertiary>sampling rate</tertiary> </indexterm> - <para>By default, a time-based event set is selected. It creates a sample every 100,000 clock cycles per processor. If the timer interrupt is used, the timer is set to whatever the jiffy rate is and is not user-settable. If the <computeroutput>cpu_type</computeroutput> is not <computeroutput>timer</computeroutput>, each event can have a <firstterm>sampling rate</firstterm> set for it. The sampling rate is the number of events between each sample snapshot.</para> + <para>By default, a time-based event set is selected. It creates a sample every 100,000 clock cycles per processor. If the timer interrupt is used, the timer is set to the respective rate and is not user-settable. If the <computeroutput>cpu_type</computeroutput> is not <computeroutput>timer</computeroutput>, each event can have a <firstterm>sampling rate</firstterm> set for it. The sampling rate is the number of events between each sample snapshot.</para> <para>When setting the event for the counter, a sample rate can also be specified:</para> <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable></command></screen> <para>Replace <replaceable>sample-rate</replaceable> with the number of events to wait before sampling again. The smaller the count, the more frequent the samples. For events that do not happen frequently, a lower count may be needed to capture the event instances.</para> <warning> <title>Sampling too frequently can overload the system</title> - <para>Be extremely careful when setting sampling rates. Sampling too frequently can overload the system, causing the system to appear as if it is frozen or causing the system to actually freeze.</para> + <para>Be extremely careful when setting sampling rates. Sampling too frequently can overload the system, causing the system to appear frozen or causing the system to actually freeze.</para> </warning> </section> <section id="s3-oprofile-events-unit-masks"> <title>Unit Masks</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>unit mask</secondary> </indexterm> <para>Some user performance monitoring events may also require unit masks to further define the event.</para> <para>Unit masks for each event are listed with the <command>ophelp</command> command. The values for each unit mask are listed in hexadecimal format. To specify more than one unit mask, the hexadecimal values must be combined using a bitwise <firstterm>or</firstterm> operation.</para> <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable>:<replaceable>unit-mask</replaceable></command></screen> + <para> + Note that on certain architectures, there can be multiple unit masks with the same hexadecimal value. In that case they have to be specified by their names only. + </para> </section> </section> <section id="s2-oprofile-starting-separate"> <title>Separating Kernel and User-space Profiles</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>configuring</secondary> <tertiary>separating profiles</tertiary> @@ -675,29 +969,33 @@ <para>Execute the following command to start profiling kernel mode for the counter again:</para> <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable>:<replaceable>unit-mask</replaceable>:1</command></screen> <para>To configure OProfile to ignore events in user mode for a specific counter, execute the following command:</para> - <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable>:<replaceable>unit-mask</replaceable>:<replaceable>kernel</replaceable>:0</command></screen> + <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable>:<replaceable>unit-mask</replaceable>:1:0</command></screen> <para>Execute the following command to start profiling user mode for the counter again:</para> - <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable>:<replaceable>unit-mask</replaceable>:<replaceable>kernel</replaceable>:1</command></screen> + <screen>~]#&#160;<command>opcontrol --event=<replaceable>event-name</replaceable>:<replaceable>sample-rate</replaceable>:<replaceable>unit-mask</replaceable>:1:1</command></screen> <para>When the OProfile daemon writes the profile data to sample files, it can separate the kernel and library profile data into separate sample files. To configure how the daemon writes to sample files, execute the following command as root:</para> <screen>~]#&#160;<command>opcontrol --separate=<replaceable>choice</replaceable></command></screen> <para> - <replaceable>choice</replaceable> can be one of the following:</para> + The <replaceable>choice</replaceable> argument can be one of the following:</para> <itemizedlist> <listitem> <para> - <computeroutput>none</computeroutput> — do not separate the profiles (default)</para> + <computeroutput>none</computeroutput> — Do not separate the profiles (default). + </para> </listitem> <listitem> <para> - <command>library</command> — generate per-application profiles for libraries</para> + <command>library</command> — Generate per-application profiles for libraries. + </para> </listitem> <listitem> <para> - <command>kernel</command> — generate per-application profiles for the kernel and kernel modules</para> + <command>kernel</command> — Generate per-application profiles for the kernel and kernel modules. + </para> </listitem> <listitem> <para> - <command>all</command> — generate per-application profiles for libraries and per-application profiles for the kernel and kernel modules</para> + <command>all</command> — Generate per-application profiles for libraries and per-application profiles for the kernel and kernel modules. + </para> </listitem> </itemizedlist> <para>If <option>--separate=library</option> is used, the sample file name includes the name of the executable as well as the name of the library.</para> @@ -709,15 +1007,13 @@ </section> <section id="s1-oprofile-starting"> - <title>Starting and Stopping OProfile</title> - <indexterm - significance="normal"> + <title>Starting and Stopping OProfile Using Legacy Mode</title> + <indexterm> <primary>OProfile</primary> <secondary>starting</secondary> </indexterm> <para>To start monitoring the system with OProfile, execute the following command as root:</para> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary> <command>opcontrol</command> @@ -730,62 +1026,65 @@ <para>Output similar to the following is displayed:</para> <screen>Using log file /var/lib/oprofile/oprofiled.log Daemon started. Profiler running.</screen> <para>The settings in <filename>/root/.oprofile/daemonrc</filename> are used.</para> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary> <command>oprofiled</command> </secondary> </indexterm> - <indexterm - significance="normal"> - <primary> - <command>oprofiled</command> - </primary> + <indexterm> + <primary><command>oprofiled</command></primary> <see>OProfile</see> </indexterm> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary> <command>oprofiled</command> </secondary> <tertiary>log file</tertiary> </indexterm> - <para>The OProfile daemon, <command>oprofiled</command>, is started; it periodically writes the sample data to the <filename>/var/lib/oprofile/samples/</filename> directory. The log file for the daemon is located at <filename>/var/lib/oprofile/oprofiled.log</filename>.</para> + <para>The OProfile daemon, <command>oprofiled</command>, is started; it periodically writes the sample data to the <filename class="directory">/var/lib/oprofile/samples/</filename> directory. The log file for the daemon is located at <filename>/var/lib/oprofile/oprofiled.log</filename>.</para> <important> <title>Disable the nmi_watchdog registers</title> <para>On a &MAJOROSVER; system, the <computeroutput>nmi_watchdog</computeroutput> registers with the <computeroutput>perf</computeroutput> subsystem. Due to this, the <computeroutput>perf</computeroutput> subsystem grabs control of the performance counter registers at boot time, blocking OProfile from working.</para> - <para>To resolve this, either boot with the <command>nmi_watchdog=0</command> kernel parameter set, or run the following command to disable <computeroutput>nmi_watchdog</computeroutput> at run time:</para> - <screen>~]#&#160;<command>echo 0 > /proc/sys/kernel/nmi_watchdog</command></screen> - <para>To re-enable <computeroutput>nmi_watchdog</computeroutput>, use the following command:</para> - <screen>~]#&#160;<command>echo 1 > /proc/sys/kernel/nmi_watchdog</command></screen> + <para>To resolve this, either boot with the <command>nmi_watchdog=0</command> kernel parameter set, or run the following command as <systemitem class="username">root</systemitem> to disable <computeroutput>nmi_watchdog</computeroutput> at run time:</para> + <screen>~]#&#160;<command>echo 0 > /proc/sys/kernel/nmi_watchdog</command></screen> + <para>To re-enable <computeroutput>nmi_watchdog</computeroutput>, use the following command as <systemitem class="username">root</systemitem>:</para> + <screen>~]#&#160;<command>echo 1 > /proc/sys/kernel/nmi_watchdog</command></screen> </important> <para>To stop the profiler, execute the following command as root:</para> <screen>~]#&#160;<command>opcontrol --shutdown</command></screen> </section> <section id="s1-oprofile-saving-data"> - <title>Saving Data</title> - <indexterm - significance="normal"> + <title>Saving Data in Legacy Mode</title> + <indexterm> <primary>OProfile</primary> <secondary>saving data</secondary> </indexterm> <para>Sometimes it is useful to save samples at a specific time. For example, when profiling an executable, it may be useful to gather different samples based on different input data sets. If the number of events to be monitored exceeds the number of counters available for the processor, multiple runs of OProfile can be used to collect data, saving the sample data to different files each time.</para> - <para>To save the current set of sample files, execute the following command, replacing <replaceable>name</replaceable> with a unique descriptive name for the current session.</para> + <para>To save the current set of sample files, execute the following command, replacing <replaceable>name</replaceable> with a unique descriptive name for the current session:</para> <screen>~]#&#160;<command>opcontrol --save=<replaceable>name</replaceable></command></screen> - <para>The directory <filename>/var/lib/oprofile/samples/<replaceable>name</replaceable>/</filename> is created and the current sample files are copied to it.</para> + <para>The command creates the directory <filename>/var/lib/oprofile/samples/<replaceable>name</replaceable>/</filename> and the current sample files are copied to it.</para> + <para> + To specify the session directory to hold the sample data, use the <option>--session-dir</option> option. If not specified, the data is saved in the <filename class="directory">oprofile_data/</filename> directory on the current path. + </para> </section> <section id="s1-oprofile-analyzing-data"> <title>Analyzing the Data</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>reading data</secondary> </indexterm> - <para>Periodically, the OProfile daemon, <command>oprofiled</command>, collects the samples and writes them to the <filename>/var/lib/oprofile/samples/</filename> directory. Before reading the data, make sure all data has been written to this directory by executing the following command as root:</para> + <para> + The same OProfile post-processing tools are used whether you collect your profile with <command>operf</command> or <command>opcontrol</command> in legacy mode. + </para> + <para> + By default, <command>operf</command> stores the profiling data in the <filename class="directory"><replaceable>current_dir</replaceable>/oprofile_data/</filename> directory. You can change to a different location with the <option>--session-dir</option> option. The usual post-profiling analysis tools such as <command>opreport</command> and <command>opannotate</command> can be used to generate profile reports. These tools search for samples in <filename class="directory"><replaceable>current_dir</replaceable>/oprofile_data/</filename> first. If this directory does not exist, the analysis tools use the standard session directory of <filename class="directory">/var/lib/oprofile/</filename>. Statistics, such as total samples received and lost samples, are written to the <filename><replaceable>session_dir</replaceable>/samples/operf.log</filename> file. + </para> + <para> + When using legacy mode, the OProfile daemon, <command>oprofiled</command>, periodically collects the samples and writes them to the <filename>/var/lib/oprofile/samples/</filename> directory. Before reading the data, make sure all data has been written to this directory by executing the following command as root:</para> <screen>~]#&#160;<command>opcontrol --dump</command></screen> <para>Each sample file name is based on the name of the executable. For example, the samples for the default event on a Pentium III processor for <command>/bin/bash</command> becomes:</para> <screen>\{root\}/bin/bash/\{dep\}/\{root\}/bin/bash/CPU_CLK_UNHALTED.100000</screen> @@ -805,31 +1104,26 @@ <para>Use these tools, along with the binaries profiled, to generate reports that can be further analyzed.</para> <warning> <title>Back up the executable and the sample files</title> - <para>The executable being profiled must be used with these tools to analyze the data. If it must change after the data is collected, back up the executable used to create the samples as well as the sample files. Please note that the sample file and the binary have to agree. Making a backup is not going to work if they do not match. <command>oparchive</command> can be used to address this problem.</para> + <para>The executable being profiled must be used with these tools to analyze the data. If it must change after the data is collected, back up the executable used to create the samples as well as the sample files. Note that the names of the sample file and the binary have to agree. You cannot make a backup if these names do not match. As an alternative, <command>oparchive</command> can be used to address this problem.</para> </warning> - <para>Samples for each executable are written to a single sample file. Samples from each dynamically linked library are also written to a single sample file. While OProfile is running, if the executable being monitored changes and a sample file for the executable exists, the existing sample file is automatically deleted. Thus, if the existing sample file is needed, it must be backed up, along with the executable used to create it before replacing the executable with a new version. The OProfile analysis tools use the executable file that created the samples during analysis. If the executable changes the analysis tools will be unable to analyze the associated samples. See <xref + <para>Samples for each executable are written to a single sample file. Samples from each dynamically linked library are also written to a single sample file. While OProfile is running, if the executable being monitored changes and a sample file for the executable exists, the existing sample file is automatically deleted. Thus, if the existing sample file is needed, it must be backed up, along with the executable used to create it before replacing the executable with a new version. The OProfile analysis tools use the executable file that created the samples during analysis. If the executable changes, the analysis tools will be unable to analyze the associated samples. See <xref linkend="s1-oprofile-saving-data"/> for details on how to back up the sample file.</para> <section id="s2-oprofile-reading-opreport"> - <title>Using <command>opreport</command> - </title> - <indexterm - significance="normal"> + <title>Using <command>opreport</command></title> + <indexterm> <primary>OProfile</primary> <secondary> <command>opreport</command> </secondary> </indexterm> - <indexterm - significance="normal"> - <primary> - <command>opreport</command> - </primary> + <indexterm> + <primary><command>opreport</command></primary> <see>OProfile</see> </indexterm> - <para>The <command>opreport</command> tool provides an overview of all the executables being profiled.</para> - <para>The following is part of a sample output:</para> - <screen>Profiling through timer interrupt + <para>The <command>opreport</command> tool provides an overview of all the executables being profiled. The following is part of a sample output from the <command>opreport</command> command:</para> + <screen>~]$&nbsp;<command>opreport</command> +Profiling through timer interrupt TIMER:0| samples| %| ------------------ @@ -855,39 +1149,42 @@ samples| %| 1 0.0038 libX11.so.6.2 1 0.0038 libgthread-2.0.so.0.400.7 1 0.0038 libwnck-1.so.4.9.0</screen> - <para>Each executable is listed on its own line. The first column is the number of samples recorded for the executable. The second column is the percentage of samples relative to the total number of samples. The third column is the name of the executable.</para> - <para>See the <command>opreport</command> man page for a list of available command line options, such as the <option>-r</option> option used to sort the output from the executable with the smallest number of samples to the one with the largest number of samples.</para> + <para> + Each executable is listed on its own line. The first column is the number of samples recorded for the executable. The second column is the percentage of samples relative to the total number of samples. The third column is the name of the executable. + </para> + <para> + See the <systemitem>opreport(1)</systemitem> manual page for a list of available command-line options, such as the <option>-r</option> option used to sort the output from the executable with the smallest number of samples to the one with the largest number of samples. You can also use the <option>-t</option> or <option>--threshold</option> option to trim the output of <command>opcontrol</command>. + </para> </section> <section id="s2-oprofile-reading-opreport-single"> <title>Using opreport on a Single Executable</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary> <command>opreport</command> </secondary> <tertiary>on a single executable</tertiary> </indexterm> - <indexterm - significance="normal"> - <primary> - <command>opreport</command> - </primary> + <indexterm> + <primary><command>opreport</command></primary> <see>OProfile</see> </indexterm> - <para>To retrieve more detailed profiled information about a specific executable, use <command>opreport</command>:</para> + <para>To retrieve more detailed profiled information about a specific executable, use the <command>opreport</command> command:</para> <screen>~]#&#160;<command>opreport <replaceable>mode</replaceable> <replaceable>executable</replaceable></command></screen> <para> - <replaceable>executable</replaceable> must be the full path to the executable to be analyzed. <replaceable>mode</replaceable> must be one of the following:</para> + Replace <replaceable>executable</replaceable> with the full path to the executable to be analyzed. <replaceable>mode</replaceable> stands for one of the following options:</para> <variablelist> <varlistentry> <term> <option>-l</option> </term> <listitem> - <para>List sample data by symbols. For example, the following is part of the output from running the command <command>opreport -l /lib/tls/libc-<replaceable>version</replaceable>.so</command>:</para> + <para>This option is used to list sample data by symbols. For example, running this command: + <synopsis>~]#&nbsp;<command>opreport -l /lib/tls/libc-<replaceable>version</replaceable>.so</command></synopsis> + </para> + <para> produces the following output:</para> <screen>samples % symbol name 12 21.4286 __gconv_transform_utf8_internal 5 8.9286 _int_malloc 4 7.1429 malloc @@ -924,7 +1221,12 @@ samples| %| </option> </term> <listitem> - <para>List sample data specific to a symbol name. For example, the following output is from the command <command>opreport -l -i __gconv_transform_utf8_internal /lib/tls/libc-<replaceable>version</replaceable>.so</command>:</para> + <para>List sample data specific to a symbol name. For example, running: + <synopsis>~]#&nbsp;<command>opreport -l -i __gconv_transform_utf8_internal /lib/tls/libc-<replaceable>version</replaceable>.so</command></synopsis> + </para> + <para> + returns the following output: + </para> <screen>samples % symbol name 12 100.000 __gconv_transform_utf8_internal</screen> <para>The first line is a summary for the symbol/executable combination.</para> @@ -936,7 +1238,12 @@ samples| %| <option>-d</option> </term> <listitem> - <para>List sample data by symbols with more detail than <option>-l</option>. For example, the following output is from the command <command>opreport -l -d __gconv_transform_utf8_internal /lib/tls/libc-<replaceable>version</replaceable>.so</command>:</para> + <para>This option lists sample data by symbols with more detail than the <option>-l</option> option. For example, with the following command: + <synopsis>~]#&nbsp;<command>opreport -d -i __gconv_transform_utf8_internal /lib/tls/libc-<replaceable>version</replaceable>.so</command></synopsis> + </para> + <para> + this output is returned: + </para> <screen>vma samples % symbol name 00a98640 12 100.000 __gconv_transform_utf8_internal 00a98640 1 8.3333 @@ -955,11 +1262,11 @@ samples| %| </varlistentry> <varlistentry> <term> - <option>-x</option> - <replaceable>symbol-name</replaceable> + <option>-e</option> + <replaceable>symbol-name</replaceable>&hellip; </term> <listitem> - <para>Exclude the comma-separated list of symbols from the output.</para> + <para>With this option, you can exclude some symbols from the output. Replace <replaceable>symbol-name</replaceable> with the comma-separated list of symbols you want to exclude.</para> </listitem> </varlistentry> <varlistentry> @@ -967,27 +1274,25 @@ samples| %| <option>session</option>:<replaceable>name</replaceable> </term> <listitem> - <para>Specify the full path to the session or a directory relative to the <filename>/var/lib/oprofile/samples/</filename> directory.</para> + <para>Here, you can specify the full path to the session, a directory relative to the <filename class="directory">/var/lib/oprofile/samples/</filename> directory, or if you are using <command>operf</command>, a directory relative to <filename class="directory">./oprofile_data/samples/</filename>.</para> </listitem> </varlistentry> </variablelist> </section> <section id="s2-oprofile-module-output"> - <title>Getting more detailed output on the modules</title> - <indexterm - significance="normal"> + <title>Getting More Detailed Output on the Modules</title> + <indexterm> <primary>OProfile</primary> <secondary> <command>opreport</command> </secondary> </indexterm> - <indexterm - significance="normal"> + <indexterm> <primary/> <see>OProfile</see> </indexterm> - <para>OProfile collects data on a system-wide basis for kernel- and user-space code running on the machine. However, once a module is loaded into the kernel, the information about the origin of the kernel module is lost. The module could have come from the <filename>initrd</filename> file on boot up, the directory with the various kernel modules, or a locally created kernel module. As a result, when OProfile records sample for a module, it just lists the samples for the modules for an executable in the root directory, but this is unlikely to be the place with the actual code for the module. You will need to take some steps to make sure that analysis tools get the executable.</para> + <para>OProfile collects data on a system-wide basis for kernel- and user-space code running on the machine. However, once a module is loaded into the kernel, the information about the origin of the kernel module is lost. The module could come from the <filename>initrd</filename> file on boot up, the directory with the various kernel modules, or a locally created kernel module. As a result, when OProfile records samples for a module, it just lists the samples for the modules for an executable in the root directory, but this is unlikely to be the place with the actual code for the module. You will need to take some steps to make sure that analysis tools get the proper executable.</para> <para>To get a more detailed view of the actions of the module, you will need to either have the module "unstripped" (that is installed from a custom build) or have the <package>debuginfo</package> package installed for the kernel.</para> <para>Find out which kernel is running with the <command>uname -a</command> command, obtain the appropriate <package>debuginfo</package> package and install it on the machine.</para> <para>Then proceed with clearing out the samples from previous runs with the following command:</para> @@ -1015,49 +1320,40 @@ samples % symbol name </section> <section id="s2-oprofile-reading-opannotate"> - <title>Using <command>opannotate</command> - </title> - <indexterm - significance="normal"> + <title>Using <command>opannotate</command></title> + <indexterm> <primary>OProfile</primary> <secondary> <command>opannotate</command> </secondary> </indexterm> - <indexterm - significance="normal"> - <primary> - <command>opannotate</command> - </primary> + <indexterm> + <primary><command>opannotate</command></primary> <see>OProfile</see> </indexterm> - <para>The <command>opannotate</command> tool tries to match the samples for particular instructions to the corresponding lines in the source code. The resulting files generated should have the samples for the lines at the left. It also puts in a comment at the beginning of each function listing the total samples for the function.</para> - <para>For this utility to work, the appropriate <package>debuginfo</package> package for the executable must be installed on the system. By default, &MAJOROS; <package>debuginfo</package> packages are not installed together with their corresponding packages, which contain the executable, so that you have to obtain and install the <package>debuginfo</package> packages separately.</para> + <para>The <command>opannotate</command> tool tries to match the samples for particular instructions to the corresponding lines in the source code. The resulting generated files should have the samples for the lines at the left. It also puts in a comment at the beginning of each function listing the total samples for the function.</para> + <para>For this utility to work, the appropriate <package>debuginfo</package> package for the executable must be installed on the system. On &MAJOROS;, the <package>debuginfo</package> packages are not automatically installed with the corresponding packages that contain the executable. You have to obtain and install them separately.</para> <para>The general syntax for <command>opannotate</command> is as follows:</para> <screen>~]#&#160;<command>opannotate --search-dirs <replaceable>src-dir</replaceable> --source <replaceable>executable</replaceable></command></screen> - <para>The directory containing the source code and the executable to be analyzed must be specified. See the <command>opannotate</command> man page for a list of additional command line options.</para> + <para>These command-line options are mandatory. Replace <replaceable>src-dir</replaceable> with a path to the directory containing the source code and specify the executable to be analyzed. See the <systemitem>opannotate(1)</systemitem> manual page for a list of additional command line options.</para> </section> </section> <section id="s1-oprofile-dev-oprofile"> - <title>Understanding <filename>/dev/oprofile/</filename> - </title> - <indexterm - significance="normal"> + <title>Understanding the /dev/oprofile/ directory +</title> + <indexterm> <primary>OProfile</primary> - <secondary> - <filename>/dev/oprofile/</filename> - </secondary> + <secondary><filename class="directory">/dev/oprofile/</filename></secondary> </indexterm> - <indexterm - significance="normal"> - <primary> - <filename>/dev/oprofile/</filename> - </primary> + <indexterm> + <primary><filename class="directory">/dev/oprofile/</filename></primary> </indexterm> - <para>The <filename>/dev/oprofile/</filename> directory contains the file system for OProfile. Use the <command>cat</command> command to display the values of the virtual files in this file system. For example, the following command displays the type of processor OProfile detected:</para> - <screen>~]#&#160;<command>cat /dev/oprofile/cpu_type</command></screen> - <para>A directory exists in <filename>/dev/oprofile/</filename> for each counter. For example, if there are 2 counters, the directories <filename>/dev/oprofile/0/</filename> and <filename>dev/oprofile/1/</filename> exist.</para> + <para> + When using OProfile in legacy mode, the <filename class="directory">/dev/oprofile/</filename> directory is used to store the file system for OProfile. On the other hand, <command>operf</command> does not require <filename class="directory">/dev/oprofile/</filename>. Use the <command>cat</command> command to display the values of the virtual files in this file system. For example, the following command displays the type of processor OProfile detected: + </para> + <screen>~]#&#160;<command>cat /dev/oprofile/cpu_type</command></screen> + <para>A directory exists in <filename class="directory">/dev/oprofile/</filename> for each counter. For example, if there are 2 counters, the directories <filename class="directory">/dev/oprofile/0/</filename> and <filename class="directory">/dev/oprofile/1/</filename> exist.</para> <para>Each directory for a counter contains the following files:</para> <itemizedlist> <listitem> @@ -1099,7 +1395,7 @@ samples % symbol name <itemizedlist> <listitem> <para> - <emphasis>Determine which applications and services are used the most on a system</emphasis> — <command>opreport</command> can be used to determine how much processor time an application or service uses. If the system is used for multiple services but is under performing, the services consuming the most processor time can be moved to dedicated systems.</para> + <emphasis>Determine which applications and services are used the most on a system</emphasis> — <command>opreport</command> can be used to determine how much processor time an application or service uses. If the system is used for multiple services but is underperforming, the services consuming the most processor time can be moved to dedicated systems.</para> </listitem> <listitem> <para> @@ -1110,35 +1406,44 @@ samples % symbol name <section id="s1-oprofile-java-support"> <title>OProfile Support for Java</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>Java</secondary> </indexterm> - <para>OProfile allows you to profile dynamically compiled code (also known as "just-in-time" or JIT code) of the Java Virtual Machine (JVM). OProfile in &MAJOROSVER; includes build-in support for the JVM Tools Interface (JVMTI) agent library, which supports Java 1.5 and higher.</para> + <para>OProfile allows you to profile dynamically compiled code (also known as "just-in-time" or JIT code) of the Java Virtual Machine (JVM). OProfile in &MAJOROSVER; includes built-in support for the JVM Tools Interface (JVMTI) agent library, which supports Java 1.5 and higher.</para> <section id="s1-oprofile-java-profiling"> <title>Profiling Java Code</title> <para>To profile JIT code from the Java Virtual Machine with the JVMTI agent, add the following to the JVM startup parameters:</para> - <screen><option>-agentlib:jvmti_oprofile</option></screen> + <screen><option>-agentlib:</option><replaceable>jvmti_oprofile</replaceable></screen> + <para> + Where <replaceable>jvmti_oprofile</replaceable> is a path to the OProfile agent. For 64-bit JVM, the path looks as follows: + </para> + <screen><option>-agentlib:/usr/lib64/oprofile/libjvmti_oprofile.so</option></screen> + <para> + Currently, you can add one command-line option: <option>--debug</option>, which enables debugging mode. + </para> <note> <title>Install the oprofile-jit package</title> - <para>The <package>oprofile-jit</package> package must be installed on the system in order to profile JIT code with OProfile.</para> + <para>The <package>oprofile-jit</package> package must be installed on the system in order to profile JIT code with OProfile. With this package, you gain the capability to show method-level information.</para> </note> - <para>To learn more about Java support in OProfile, refer to the OProfile Manual, which is linked from <xref - linkend="s1-oprofile-additional-resources"/>.</para> + <para> + Depending on the JVM that you are using, you may have to install the <emphasis>debuginfo</emphasis> package for the JVM. For OpenJDK, this package is required, there is no debuginfo package for Oracle JDK. To keep the debug information packages synchronized with their respective non-debug packages, you also need to install the <emphasis>yum-plugin-auto-update-debug-info</emphasis> plug-in. This plug-in searches the debug information repository for corresponding updates. + </para> + <para> + After successful setup, you can apply the standard profiling and analyzing tools described in previous sections + </para> + <para>To learn more about Java support in OProfile, see the OProfile Manual, which is linked from <xref + linkend="s1-oprofile_additional_resources"/>.</para> </section> </section> <section id="s1-oprofile-gui"> <title>Graphical Interface</title> - <indexterm - significance="normal"> - <primary> - <command>oprof_start</command> - </primary> + <indexterm> + <primary><command>oprof_start</command></primary> </indexterm> - <para>Some OProfile preferences can be set with a graphical interface. To start it, execute the <command>oprof_start</command> command as root at a shell prompt. To use the graphical interface, you will need to have the <filename>oprofile-gui</filename> package installed.</para> + <para>Some OProfile preferences can be set with a graphical interface. Make sure you have the <filename>oprofile-gui</filename> package that provides the OProfile GUI installed on your system. To start the interface, execute the <command>oprof_start</command> command as root at a shell prompt.</para> <para>After changing any of the options, save them by clicking the <guibutton>Save and quit</guibutton> button. The preferences are written to <filename>/root/.oprofile/daemonrc</filename>, and the application exits. Exiting the application does not stop OProfile from sampling.</para> <para>On the <guilabel>Setup</guilabel> tab, to set events for the processor counters as discussed in <xref linkend="s2-oprofile-events"/>, select the counter from the pulldown menu and select the event from the list. A brief description of the event appears in the text box below the list. Only events available for the specific counter and the specific architecture are displayed. The interface also displays whether the profiler is running and some brief statistics about it.</para> @@ -1160,13 +1465,13 @@ samples % symbol name </mediaobject> </figure> <para>On the right side of the tab, select the <guilabel>Profile kernel</guilabel> option to count events in kernel mode for the currently selected event, as discussed in <xref - linkend="s2-oprofile-starting-separate"/>. If this option is unselected, no samples are collected for the kernel.</para> + linkend="s2-oprofile-starting-separate"/>. If this option is not selected, no samples are collected for the kernel.</para> <para>Select the <guilabel>Profile user binaries</guilabel> option to count events in user mode for the currently selected event, as discussed in <xref - linkend="s2-oprofile-starting-separate"/>. If this option is unselected, no samples are collected for user applications.</para> + linkend="s2-oprofile-starting-separate"/>. If this option is not selected, no samples are collected for user applications.</para> <para>Use the <guilabel>Count</guilabel> text field to set the sampling rate for the currently selected event as discussed in <xref linkend="s3-oprofile-events-sampling"/>.</para> <para>If any unit masks are available for the currently selected event, as discussed in <xref - linkend="s3-oprofile-events-unit-masks"/>, they are displayed in the <guilabel>Unit Masks</guilabel> area on the right side of the <guilabel>Setup</guilabel> tab. Select the checkbox beside the unit mask to enable it for the event.</para> + linkend="s3-oprofile-events-unit-masks"/>, they are displayed in the <guilabel>Unit Masks</guilabel> area on the right side of the <guilabel>Setup</guilabel> tab. Select the check box beside the unit mask to enable it for the event.</para> <para>On the <guilabel>Configuration</guilabel> tab, to profile the kernel, enter the name and location of the <filename>vmlinux</filename> file for the kernel to monitor in the <guilabel>Kernel image file</guilabel> text field. To configure OProfile not to monitor the kernel, select <guilabel>No kernel image</guilabel>.</para> <figure float="0" @@ -1184,7 +1489,7 @@ samples % symbol name </textobject> </mediaobject> </figure> - <para>If the <guilabel>Verbose</guilabel> option is selected, the <command>oprofiled</command> daemon log includes more information.</para> + <para>If the <guilabel>Verbose</guilabel> option is selected, the <command>oprofiled</command> daemon log includes more detailed information.</para> <para>If <guilabel>Per-application profiles</guilabel> is selected, OProfile generates per-application profiles for libraries. This is equivalent to the <command>opcontrol --separate=library</command> command. If <guilabel>Per-application profiles, including kernel</guilabel> is selected, OProfile generates per-application profiles for the kernel and kernel modules as discussed in <xref linkend="s2-oprofile-starting-separate"/>. This is equivalent to the <command>opcontrol --separate=kernel</command> command.</para> <para>To force data to be written to samples files as discussed in <xref @@ -1194,29 +1499,25 @@ samples % symbol name <section id="s1-oprofile-and-systemtap"> <title>OProfile and SystemTap</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>SystemTap</secondary> </indexterm> - <para>SystemTap is a tracing and probing tool that allows users to study and monitor the activities of the operating system in fine detail. It provides information similar to the output of tools like <command>netstat</command>, <command>ps</command>, <command>top</command>, and <command>iostat</command>; however, SystemTap is designed to provide more filtering and analysis options for collected information.</para> + <para>SystemTap is a tracing and probing tool that allows users to study and monitor the activities of the operating system in fine detail. It provides information similar to the output of tools like <command>netstat</command>, <command>ps</command>, <command>top</command>, and <command>iostat</command>; however, SystemTap is designed to provide more filtering and analysis options for the collected information.</para> <para>While using OProfile is suggested in cases of collecting data on where and why the processor spends time in a particular area of code, it is less usable when finding out why the processor stays idle.</para> - <para>You might want to use SystemTap when instrumenting specific places in code. Because SystemTap allows you to run the code instrumentation without having to stop and restart the instrumentation, it is particularly useful for instrumenting the kernel and daemons.</para> - <para>For more information on SystemTap, refer to <xref - linkend="s2-oprofile-useful-websites"/> for the relevant SystemTap documentation.</para> + <para>You might want to use SystemTap when instrumenting specific places in code. Because SystemTap allows you to run the code instrumentation without having to stop and restart the instrumented code, it is particularly useful for instrumenting the kernel and daemons.</para> + <para>For more information on SystemTap, see <xref + linkend="br-oprofile_online_documentation"/> for the relevant SystemTap documentation.</para> </section> <section - id="s1-oprofile-additional-resources"> + id="s1-oprofile_additional_resources"> <title>Additional Resources</title> - <indexterm - significance="normal"> + <indexterm> <primary>OProfile</primary> <secondary>additional resources</secondary> </indexterm> - <para>This chapter only highlights OProfile and how to configure and use it. To learn more, refer to the following resources.</para> - <section - id="s1-oprofile-installed-docs"> - <title>Installed Docs</title> + <para>To learn more about OProfile and how to configure it, see the following resources.</para> + <bridgehead id="br-oprofile_installed_documentation">Installed Documentation</bridgehead> <itemizedlist> <listitem> <para> @@ -1225,26 +1526,31 @@ samples % symbol name </listitem> <listitem> <para> - <command>oprofile</command> man page — Discusses <command>opcontrol</command>, <command>opreport</command>, <command>opannotate</command>, and <command>ophelp</command> + <filename>oprofile(1)</filename> manual page — Discusses <command>opcontrol</command>, <command>opreport</command>, <command>opannotate</command>, and <command>ophelp</command> </para> </listitem> + <listitem> + <para> + <filename>operf(1)</filename> manual page + </para> + </listitem> </itemizedlist> - </section> - <section - id="s2-oprofile-useful-websites"> - <title>Useful Websites</title> + <bridgehead id="br-oprofile_online_documentation">Online Documentation</bridgehead> <itemizedlist> <listitem> <para> <ulink - url="http://oprofile.sourceforge.net/">http://oprofile.sourc... — Contains the latest documentation, mailing lists, IRC channels, and more.</para> + url="http://oprofile.sourceforge.net/">http://oprofile.sourc... — Contains the latest upstream documentation, mailing lists, IRC channels, and more.</para> </listitem> + </itemizedlist> + <bridgehead id="br-oprofile_see_also">See Also</bridgehead> + <itemizedlist> <listitem> <para> <ulink - url="http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/ht... Beginners Guide</ulink> — Provides basic instructions on how to use SystemTap to monitor different subsystems of &MAJOROS; in finer detail.</para> + url="https://access.redhat.com/documentation/en-US/Red_Hat_Enterpris... Beginners Guide</ulink> — Provides basic instructions on how to use SystemTap to monitor different subsystems of &MAJOROS; in finer detail.</para> + </listitem> </itemizedlist> - </section> </section> </chapter>