Profilers - Revision history

Guil at 12:24, 18 January 2012

2012-01-18T12:24:48Z

FransMeulenbroeks at 07:34, 18 November 2008

2008-11-18T07:34:22Z

FransMeulenbroeks at 07:34, 18 November 2008

2008-11-18T07:34:05Z

GustavoBarbieri: /* Valgrind Cachegrind/Callgrind */

2008-11-17T22:53:20Z

‎Valgrind Cachegrind/Callgrind

GustavoBarbieri: /* User Interfaces */

2008-11-17T22:45:59Z

‎User Interfaces

GustavoBarbieri at 22:45, 17 November 2008

2008-11-17T22:45:26Z

New page

Linux systems count with a wide variety of profilers each of with their pros and cons, there is no magic bullet, it's recommended to use more than one tool when analyzing your application, I recommend you to use at least OProfile and Valgrind.

== OProfile ==

[http://oprofile.sourceforge.net/about OProfile] is a non-obtrusive system-wide profiler for Linux, it can use system's performance counters to give you insights on where to optimize your code.

Helper script for simple runs (remember to run <code>opcontrol --setup ...</code> once before this script):

#!/bin/sh
sudo opcontrol --init
sudo opcontrol --start
sudo opcontrol --dump
sudo opcontrol --reset
$@
sudo opcontrol --stop
opreport --symbols > oprof.txt

see also [http://oprofile.sourceforge.net/doc/oprofile-gui.html oprof_start]

=== User Interfaces ===

==== oprof_start ====

Qt3 GUI for OProfile, ships in official distribution, see http://oprofile.sourceforge.net/doc/oprofile-gui.html

==== KCachegrind ====

One can use reports that are usable by kcachegrind with (for more instruction see http://docs.kde.org/kde3/en/kdesdk/kcachegrind/using-kcachegrind.html):

opreport -gdf | op2callgrind

==== OProfileUI ====

Another useful application is [http://labs.o-hand.com/oprofileui/ OProfileUI], it provides a GTK+2 user interface to run and show statistics, including basic call graph.

=== Documentation ===

In depth explanation can be found at:
* [http://www-128.ibm.com/developerworks/linux/library/l-oprof.html IBM DeveloperWorks]
* [http://oprofile.sourceforge.net/doc/index.html OProfile Manual]
* [http://oprofile.sourceforge.net/faq/ OProfile FAQ]

=== Caveats ===

* JFFS2 does not provide OProfile requirements, running OProfile to output data into JFFS2 partition will fail, instead use a ramfs or some external card with ext2/3:
mkdir /var/lib/oprofile
# ram
mount -t ramfs none /var/lib/oprofile
# ext3
mount -t ext3 /dev/sda1 /var/lib/oprofile
* settings event counter too low or too high may hang your test machine. Check the recommended value for each event on specific platforms.

== GProf ==

Unlike OProfile, GProf requires applications to be compiled with special flag <code>-pg</code> so compiler will introduce marks in generated binary to measure runs. After run the <code>gmon.out</code> file will be created with measured data, this file can be viewed with tools like gprof or kprof.

=== User Interfaces ===

==== KProf ====

Outdated, but exists: http://kprof.sourceforge.net/

=== Documentation ===

* [http://sourceware.org/binutils/docs/gprof/index.html GProf Manual]

=== Caveats ===

* Imprecise
* Huge overhead

== Valgrind Cachegrind/Callgrind ==

[http://valgrind.org/info/about.html Valgrind] is tool suite for debugging and profiling. It's most famous for it's '''memcheck''' tool to check for memory leaks, invalid access, double frees and more, but also ships with interesting callgrind and cachegrind.

'''cachegrind''' will accurately simulate L1, L2 and D1 CPU caches so you can figure out what in your code is trashing your memory access.

'''callgrind''' is an extension of cachegrind that will collect call graphs.

=== User Interfaces ===

The most famous user interface is [http://kcachegrind.sourceforge.net/cgi-bin/show.cgi/KcacheGrindIndex KCachegrind]

=== Documentation ===

* [http://valgrind.org/docs/manual/manual.html Valgrind Manual]
* [http://valgrind.org/docs/manual/FAQ.html FAQ]
* http://docs.kde.org/kde3/en/kdesdk/kcachegrind/using-kcachegrind.html

=== Caveats ===

* Since these are simulators, your application will run very slow, from 20 to 100 times or more.

← Older revision		Revision as of 12:24, 18 January 2012
Line 70:		Line 70:
	* Imprecise		* Imprecise
	* Huge overhead		* Huge overhead
		+
		+	== ARM Streamline ==
		+
		+	[http://www.arm.com/streamline ARM Streamline] is a commercial sample-based system performance analyzer which brings together performance counters from the core(s) and OS, time- and event-based profiling, context switch tracepoints, and instrumented messages (like printf) to provide developers with a system-to-instruction drill-down ability.
		+
		+	Streamline is a component of the ARM DS-5 suite, which has a [http://www.arm.com/products/tools/software-tools/ds-5/ds-5-downloads.php 30-day evaluation] version available for download.

← Older revision		Revision as of 07:34, 18 November 2008
Line 96:		Line 96:
	* Since these are simulators, your application will run very slow, from 20 to 100 times or more.		* Since these are simulators, your application will run very slow, from 20 to 100 times or more.

−	[[Category:Tools]	+	[[Category:Tools]]

← Older revision		Revision as of 07:34, 18 November 2008
Line 95:		Line 95:

	* Since these are simulators, your application will run very slow, from 20 to 100 times or more.		* Since these are simulators, your application will run very slow, from 20 to 100 times or more.
		+
		+	[[Category:Tools]

@@ Line 87: / Line 87: @@
 === Documentation ===
 * [http://valgrind.org/docs/manual/manual.html Valgrind Manual]
 * [http://valgrind.org/docs/manual/FAQ.html FAQ]

← Older revision		Revision as of 22:45, 17 November 2008
Line 82:		Line 82:
	=== User Interfaces ===		=== User Interfaces ===

		+	==== KCachegrind ====
	The most famous user interface is [http://kcachegrind.sourceforge.net/cgi-bin/show.cgi/KcacheGrindIndex KCachegrind]		The most famous user interface is [http://kcachegrind.sourceforge.net/cgi-bin/show.cgi/KcacheGrindIndex KCachegrind]