[lttng-modules.git] / README

LTTng 2.0 modules

Mathieu Desnoyers
July 19, 2011

LTTng 2.0 kernel modules build against a vanilla or distribution kernel, without
need for additional patches. Other features:

- Produces CTF (Common Trace Format) natively,
  (http://www.efficios.com/ctf)
- Function tracer, perf counters and kprobes support,
- Integrated interface for both kernel and userspace tracing,
- Have the ability to attach "context" information to events in the
  trace (e.g. any perf counter, pid, ppid, tid, comm name, etc). So
  basically, all the perf "required fields" like "preempt count" and
  "bkl count" are all optional, specified on a per-tracing-session basis
  (except for timestamp and event id, which are mandatory).

To build and install, you will need to have your kernel headers available (or
access to your full kernel source tree), and use:

make
make install

If you need to specify the target directory to the kernel you want to build
against, use:

KERNELDIR=path_to_kernel_dir make
KERNELDIR=path_to_kernel_dir make install

Use lttng-tools to control the tracer. LTTng tools should automatically load the
kernel modules when needed.

So far, it has been tested on vanilla kernels 2.6.38 and 2.6.39 (on x86 at the
moment). It should work fine with newer kernels and other architectures, but
expect build issues with kernels older than 2.6.36. The clock source currently
used is the standard gettimeofday (slower, less scalable and less precise than
the LTTng 0.x clocks). Support for LTTng 0.x clocks will be added back soon into
LTTng 2.0.
Commit	Line	Data
98541fcf MD	1	LTTng 2.0 modules
	2
	3	Mathieu Desnoyers
	4	July 19, 2011
	5
	6	LTTng 2.0 kernel modules build against a vanilla or distribution kernel, without
	7	need for additional patches. Other features:
	8
	9	- Produces CTF (Common Trace Format) natively,
	10	(http://www.efficios.com/ctf)
	11	- Function tracer, perf counters and kprobes support,
	12	- Integrated interface for both kernel and userspace tracing,
	13	- Have the ability to attach "context" information to events in the
	14	trace (e.g. any perf counter, pid, ppid, tid, comm name, etc). So
	15	basically, all the perf "required fields" like "preempt count" and
	16	"bkl count" are all optional, specified on a per-tracing-session basis
	17	(except for timestamp and event id, which are mandatory).
	18
	19	To build and install, you will need to have your kernel headers available (or
	20	access to your full kernel source tree), and use:
	21
	22	make
	23	make install
	24
	25	If you need to specify the target directory to the kernel you want to build
	26	against, use:
	27
	28	KERNELDIR=path_to_kernel_dir make
	29	KERNELDIR=path_to_kernel_dir make install
	30
	31	Use lttng-tools to control the tracer. LTTng tools should automatically load the
	32	kernel modules when needed.
	33
	34	So far, it has been tested on vanilla kernels 2.6.38 and 2.6.39 (on x86 at the
	35	moment). It should work fine with newer kernels and other architectures, but
	36	expect build issues with kernels older than 2.6.36. The clock source currently
	37	used is the standard gettimeofday (slower, less scalable and less precise than
	38	the LTTng 0.x clocks). Support for LTTng 0.x clocks will be added back soon into
	39	LTTng 2.0.