Kernel Instrumentation - Revision history

Cschalle: Add category

2011-10-27T22:37:32Z

Add category

2011-10-27T22:20:09Z

2010-07-22T11:04:56Z

‎Boot Tracer

2010-02-22T17:40:22Z

‎ftrace

2009-04-07T02:22:48Z

2009-01-08T09:00:16Z

‎Boot Tracer

2009-01-08T08:58:08Z

Add boot tracer kernel feature

2008-11-14T14:05:06Z

Add SystemTap

2008-11-14T13:57:09Z

Add ftrace

2008-11-14T13:42:26Z

‎Printk Times

← Older revision		Revision as of 22:37, 27 October 2011
Line 69:		Line 69:
	**instrumentation systems tend to pollute the cache lines for the processor		**instrumentation systems tend to pollute the cache lines for the processor
	*There doesn't seem to be a single API to support in-kernel timing instrumentation which is supported on lots of different architectures. This is the main reason for CELF's current project to define an [[Instrumentation API]]		*There doesn't seem to be a single API to support in-kernel timing instrumentation which is supported on lots of different architectures. This is the main reason for CELF's current project to define an [[Instrumentation API]]
		+
		+	[[Category:Kernel]]

@@ Line 59: / Line 59: @@
 === SystemTap / Kprobes ===
-[http://sourceware.org/systemtap/ SystemTap] is a sophisticated kernel instrumentation tool that can be scripted with it's own language to gather information about a running kernel. It uses the Kprobes infrastructure to implement it's tracing.
+[http://sourceware.org/systemtap/ SystemTap] is a sophisticated kernel instrumentation tool that can be scripted with its own language to gather information about a running kernel. It uses the Kprobes infrastructure to implement it's tracing.
 == Notes ==

@@ Line 19: / Line 19: @@
 === Boot Tracer ===
-Starting from 2.6.28 the kernel has this new feature to optimize the boot time. It records the timings of the initcalls. Its aim is to be parsed by the scripts/bootgraph.pl tool to produce graphics about boot inefficiencies, giving a visual representation of the delays during initcalls. Users need to enable CONFIG_BOOT_TRACER, boot with the "initcall_debug" and "printk.time=1" parameters, and run "dmesg | perl scripts/bootgraph.pl > output.svg" to generate the final data.
+Starting from 2.6.28 the kernel has this new feature to optimize the boot time. It records the timings of the initcalls. Its aim is to be parsed by the scripts/bootgraph.pl tool to produce graphics about boot inefficiencies, giving a visual representation of the delays during initcalls. Users need to boot with the "initcall_debug" and "printk.time=1" parameters, and run "dmesg | perl scripts/bootgraph.pl > output.svg" to generate the final data.
 === Kernel Function Instrumentation (KFI) ===

@@ Line 53: / Line 53: @@
 Ftrace is a simple function tracer which initially came from the -rt patches but was mainlined in 2.6.27. Compiler profiling features are used to insert an instrumentation call (with gcc ''-pg'' option) that can be overwritten with a NOP sequence to ensure overhead is minimal with tracing disabled (this is enabled through CONFIG_DYNAMIC_FTRACE). There are a number of tracers in the kernel that use ftrace to trace high level events such as irq enabling/disabling preemption enabling/disabling, scheduler events and branch profiling.
-The interface to access ftrace can be found in ''/debugfs/tracing'', and is very extensively documented in [http://lxr.free-electrons.com/source/Documentation/ftrace.txt Documentation/ftrace.txt].
+The interface to access ftrace can be found in ''/debugfs/tracing'', and is very extensively documented in [http://lxr.free-electrons.com/source/Documentation/trace/ftrace.txt Documentation/trace/ftrace.txt].
 Steven Rostedt, ftrace's main developer, gave a conference on ftrace at the Ottawa Linux Symposium 2008. The [http://free-electrons.com/pub/video/2008/ols/ols2008-steven-rostedt-ftrace.ogg video] and [http://people.redhat.com/srostedt/ftrace-tutorial.odp slides].

@@ Line 51: / Line 51: @@
 === ftrace ===
-Ftrace is a simple function tracer which initially came from the -rt patches but was mainlined in 2.6.27. Compiler profiling features are used to insert an instrumentation call that can be overwritten with a NOP sequence to ensure overhead is minimal with tracing disabled. There are a number of tracers in the kernel that use ftrace to trace high level events such as irq enabling/disabling preemption enabling/disabling, scheduler events and branch profiling.
+Ftrace is a simple function tracer which initially came from the -rt patches but was mainlined in 2.6.27. Compiler profiling features are used to insert an instrumentation call (with gcc ''-pg'' option) that can be overwritten with a NOP sequence to ensure overhead is minimal with tracing disabled (this is enabled through CONFIG_DYNAMIC_FTRACE). There are a number of tracers in the kernel that use ftrace to trace high level events such as irq enabling/disabling preemption enabling/disabling, scheduler events and branch profiling.
-The interface to access ftrace can be found in /debugfs/tracing, and is documented in Documentation/ftrace.txt.
+The interface to access ftrace can be found in ''/debugfs/tracing'', and is very extensively documented in [http://lxr.free-electrons.com/source/Documentation/ftrace.txt Documentation/ftrace.txt].
 === SystemTap / Kprobes ===

← Older revision		Revision as of 08:58, 8 January 2009
Line 16:		Line 16:

	See [[Printk Times]]		See [[Printk Times]]
		+
		+	=== Boot Tracer ===
		+
		+	Starting from 2.6.28 the kernel has this new feature to optimaze the boot time. It records the timings of the initcalls. Its aim is to be parsed by the scripts/bootgraph.pl tool to produce graphics about boot inefficiencies, giving a visual representation of the delays during initcalls. Users need to enable CONFIG_BOOT_TRACER, boot with the "initcall_debug" and "printk.time=1" parameters, and run "dmesg \| perl scripts/bootgraph.pl > output.svg" to generate the final data.

	=== Kernel Function Instrumentation (KFI) ===		=== Kernel Function Instrumentation (KFI) ===

← Older revision		Revision as of 13:57, 14 November 2008
Line 44:		Line 44:

	See http://www.ussg.iu.edu/hypermail/linux/kernel/0201.3/0888.html		See http://www.ussg.iu.edu/hypermail/linux/kernel/0201.3/0888.html
		+
		+	=== ftrace ===
		+
		+	Ftrace is a simple function tracer which initially came from the -rt patches but was mainlined in 2.6.27. Compiler profiling features are used to insert an instrumentation call that can be overwritten with a NOP sequence to ensure overhead is minimal with tracing disabled. There are a number of tracers in the kernel that use ftrace to trace high level events such as irq enabling/disabling preemption enabling/disabling, scheduler events and branch profiling.
		+
		+	The interface to access ftrace can be found in /debugfs/tracing, and is documented in Documentation/ftrace.txt.

	== Notes ==		== Notes ==

@@ Line 13: / Line 13: @@
 === Printk Times ===
-Produces printk's with extra time data on them.
+Produces printk's with extra time data on them. As of kernel 2.6.11 this is part of the mainline kernel enabled by CONFIG_PRINTK_TIME. Previous versions can add it via a very simple patch. It works for bootup time measurements, or other places where you can just jam in a printk or two.
-−
-This is a very simple patch, which works for bootup time measurements, or other places where you can just jam in a printk or two.
 See [[Printk Times]]