tracepoints: format documentation

			    Mathieu Desnoyers

			    Mathieu Desnoyers

This document introduces Linux Kernel Tracepoints and their use. It provides

This document introduces Linux Kernel Tracepoints and their use. It

examples of how to insert tracepoints in the kernel and connect probe functions

provides examples of how to insert tracepoints in the kernel and

to them and provides some examples of probe functions.

connect probe functions to them and provides some examples of probe

functions.

* Purpose of tracepoints

* Purpose of tracepoints

A tracepoint placed in code provides a hook to call a function (probe) that you

A tracepoint placed in code provides a hook to call a function (probe)

can provide at runtime. A tracepoint can be "on" (a probe is connected to it) or

that you can provide at runtime. A tracepoint can be "on" (a probe is

"off" (no probe is attached). When a tracepoint is "off" it has no effect,

connected to it) or "off" (no probe is attached). When a tracepoint is

except for adding a tiny time penalty (checking a condition for a branch) and

"off" it has no effect, except for adding a tiny time penalty

space penalty (adding a few bytes for the function call at the end of the

(checking a condition for a branch) and space penalty (adding a few

instrumented function and adds a data structure in a separate section).  When a

bytes for the function call at the end of the instrumented function

tracepoint is "on", the function you provide is called each time the tracepoint

and adds a data structure in a separate section).  When a tracepoint

is executed, in the execution context of the caller. When the function provided

is "on", the function you provide is called each time the tracepoint

ends its execution, it returns to the caller (continuing from the tracepoint

is executed, in the execution context of the caller. When the function

site).

provided ends its execution, it returns to the caller (continuing from

the tracepoint site).

You can put tracepoints at important locations in the code. They are

You can put tracepoints at important locations in the code. They are

lightweight hooks that can pass an arbitrary number of parameters,

lightweight hooks that can pass an arbitrary number of parameters,

which prototypes are described in a tracepoint declaration placed in a header

which prototypes are described in a tracepoint declaration placed in a

file.

header file.

They can be used for tracing and performance accounting.

They can be used for tracing and performance accounting.

- subsys_eventname is an identifier unique to your event

- subsys_eventname is an identifier unique to your event

    - subsys is the name of your subsystem.

    - subsys is the name of your subsystem.

    - eventname is the name of the event to trace.

    - eventname is the name of the event to trace.

- TPPTOTO(int firstarg, struct task_struct *p) is the prototype of the function

  called by this tracepoint.

- TPARGS(firstarg, p) are the parameters names, same as found in the prototype.

Connecting a function (probe) to a tracepoint is done by providing a probe

- TPPTOTO(int firstarg, struct task_struct *p) is the prototype of the

(function to call) for the specific tracepoint through

  function called by this tracepoint.

- TPARGS(firstarg, p) are the parameters names, same as found in the

  prototype.

Connecting a function (probe) to a tracepoint is done by providing a

probe (function to call) for the specific tracepoint through

register_trace_subsys_eventname().  Removing a probe is done through

register_trace_subsys_eventname().  Removing a probe is done through

unregister_trace_subsys_eventname(); it will remove the probe.

unregister_trace_subsys_eventname(); it will remove the probe.

tracepoint_synchronize_unregister() must be called before the end of the module exit

function to make sure there is no caller left using the probe. This, and the

tracepoint_synchronize_unregister() must be called before the end of

fact that preemption is disabled around the probe call, make sure that probe

the module exit function to make sure there is no caller left using

removal and module unload are safe. See the "Probe example" section below for a

the probe. This, and the fact that preemption is disabled around the

sample probe module.

probe call, make sure that probe removal and module unload are safe.

See the "Probe example" section below for a sample probe module.

The tracepoint mechanism supports inserting multiple instances of the same

tracepoint, but a single definition must be made of a given tracepoint name over

The tracepoint mechanism supports inserting multiple instances of the

all the kernel to make sure no type conflict will occur. Name mangling of the

same tracepoint, but a single definition must be made of a given

tracepoints is done using the prototypes to make sure typing is correct.

tracepoint name over all the kernel to make sure no type conflict will

Verification of probe type correctness is done at the registration site by the

occur. Name mangling of the tracepoints is done using the prototypes

compiler. Tracepoints can be put in inline functions, inlined static functions,

to make sure typing is correct. Verification of probe type correctness

and unrolled loops as well as regular functions.

is done at the registration site by the compiler. Tracepoints can be

put in inline functions, inlined static functions, and unrolled loops

The naming scheme "subsys_event" is suggested here as a convention intended

as well as regular functions.

to limit collisions. Tracepoint names are global to the kernel: they are

considered as being the same whether they are in the core kernel image or in

The naming scheme "subsys_event" is suggested here as a convention

modules.

intended to limit collisions. Tracepoint names are global to the

kernel: they are considered as being the same whether they are in the

core kernel image or in modules.

If the tracepoint has to be used in kernel modules, an

If the tracepoint has to be used in kernel modules, an

EXPORT_TRACEPOINT_SYMBOL_GPL() or EXPORT_TRACEPOINT_SYMBOL() can be used to

EXPORT_TRACEPOINT_SYMBOL_GPL() or EXPORT_TRACEPOINT_SYMBOL() can be

export the defined tracepoints.

used to export the defined tracepoints.

* Probe / tracepoint example

* Probe / tracepoint example