1=======================
2Kernel Probes (Kprobes)
3=======================
4
5:Author: Jim Keniston <jkenisto@us.ibm.com>
6:Author: Prasanna S Panchamukhi <prasanna.panchamukhi@gmail.com>
7:Author: Masami Hiramatsu <mhiramat@redhat.com>
8
9.. CONTENTS
10
11  1. Concepts: Kprobes, and Return Probes
12  2. Architectures Supported
13  3. Configuring Kprobes
14  4. API Reference
15  5. Kprobes Features and Limitations
16  6. Probe Overhead
17  7. TODO
18  8. Kprobes Example
19  9. Kretprobes Example
20  10. Deprecated Features
21  Appendix A: The kprobes debugfs interface
22  Appendix B: The kprobes sysctl interface
23  Appendix C: References
24
25Concepts: Kprobes and Return Probes
26=========================================
27
28Kprobes enables you to dynamically break into any kernel routine and
29collect debugging and performance information non-disruptively. You
30can trap at almost any kernel code address [1]_, specifying a handler
31routine to be invoked when the breakpoint is hit.
32
33.. [1] some parts of the kernel code can not be trapped, see
34       :ref:`kprobes_blacklist`)
35
36There are currently two types of probes: kprobes, and kretprobes
37(also called return probes).  A kprobe can be inserted on virtually
38any instruction in the kernel.  A return probe fires when a specified
39function returns.
40
41In the typical case, Kprobes-based instrumentation is packaged as
42a kernel module.  The module's init function installs ("registers")
43one or more probes, and the exit function unregisters them.  A
44registration function such as register_kprobe() specifies where
45the probe is to be inserted and what handler is to be called when
46the probe is hit.
47
48There are also ``register_/unregister_*probes()`` functions for batch
49registration/unregistration of a group of ``*probes``. These functions
50can speed up unregistration process when you have to unregister
51a lot of probes at once.
52
53The next four subsections explain how the different types of
54probes work and how jump optimization works.  They explain certain
55things that you'll need to know in order to make the best use of
56Kprobes -- e.g., the difference between a pre_handler and
57a post_handler, and how to use the maxactive and nmissed fields of
58a kretprobe.  But if you're in a hurry to start using Kprobes, you
59can skip ahead to :ref:`kprobes_archs_supported`.
60
61How Does a Kprobe Work?
62-----------------------
63
64When a kprobe is registered, Kprobes makes a copy of the probed
65instruction and replaces the first byte(s) of the probed instruction
66with a breakpoint instruction (e.g., int3 on i386 and x86_64).
67
68When a CPU hits the breakpoint instruction, a trap occurs, the CPU's
69registers are saved, and control passes to Kprobes via the
70notifier_call_chain mechanism.  Kprobes executes the "pre_handler"
71associated with the kprobe, passing the handler the addresses of the
72kprobe struct and the saved registers.
73
74Next, Kprobes single-steps its copy of the probed instruction.
75(It would be simpler to single-step the actual instruction in place,
76but then Kprobes would have to temporarily remove the breakpoint
77instruction.  This would open a small time window when another CPU
78could sail right past the probepoint.)
79
80After the instruction is single-stepped, Kprobes executes the
81"post_handler," if any, that is associated with the kprobe.
82Execution then continues with the instruction following the probepoint.
83
84Changing Execution Path
85-----------------------
86
87Since kprobes can probe into a running kernel code, it can change the
88register set, including instruction pointer. This operation requires
89maximum care, such as keeping the stack frame, recovering the execution
90path etc. Since it operates on a running kernel and needs deep knowledge
91of computer architecture and concurrent computing, you can easily shoot
92your foot.
93
94If you change the instruction pointer (and set up other related
95registers) in pre_handler, you must return !0 so that kprobes stops
96single stepping and just returns to the given address.
97This also means post_handler should not be called anymore.
98
99Note that this operation may be harder on some architectures which use
100TOC (Table of Contents) for function call, since you have to setup a new
101TOC for your function in your module, and recover the old one after
102returning from it.
103
104Return Probes
105-------------
106
107How Does a Return Probe Work?
108^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
109
110When you call register_kretprobe(), Kprobes establishes a kprobe at
111the entry to the function.  When the probed function is called and this
112probe is hit, Kprobes saves a copy of the return address, and replaces
113the return address with the address of a "trampoline."  The trampoline
114is an arbitrary piece of code -- typically just a nop instruction.
115At boot time, Kprobes registers a kprobe at the trampoline.
116
117When the probed function executes its return instruction, control
118passes to the trampoline and that probe is hit.  Kprobes' trampoline
119handler calls the user-specified return handler associated with the
120kretprobe, then sets the saved instruction pointer to the saved return
121address, and that's where execution resumes upon return from the trap.
122
123While the probed function is executing, its return address is
124stored in an object of type kretprobe_instance.  Before calling
125register_kretprobe(), the user sets the maxactive field of the
126kretprobe struct to specify how many instances of the specified
127function can be probed simultaneously.  register_kretprobe()
128pre-allocates the indicated number of kretprobe_instance objects.
129
130For example, if the function is non-recursive and is called with a
131spinlock held, maxactive = 1 should be enough.  If the function is
132non-recursive and can never relinquish the CPU (e.g., via a semaphore
133or preemption), NR_CPUS should be enough.  If maxactive <= 0, it is
134set to a default value: max(10, 2*NR_CPUS).
135
136It's not a disaster if you set maxactive too low; you'll just miss
137some probes.  In the kretprobe struct, the nmissed field is set to
138zero when the return probe is registered, and is incremented every
139time the probed function is entered but there is no kretprobe_instance
140object available for establishing the return probe.
141
142Kretprobe entry-handler
143^^^^^^^^^^^^^^^^^^^^^^^
144
145Kretprobes also provides an optional user-specified handler which runs
146on function entry. This handler is specified by setting the entry_handler
147field of the kretprobe struct. Whenever the kprobe placed by kretprobe at the
148function entry is hit, the user-defined entry_handler, if any, is invoked.
149If the entry_handler returns 0 (success) then a corresponding return handler
150is guaranteed to be called upon function return. If the entry_handler
151returns a non-zero error then Kprobes leaves the return address as is, and
152the kretprobe has no further effect for that particular function instance.
153
154Multiple entry and return handler invocations are matched using the unique
155kretprobe_instance object associated with them. Additionally, a user
156may also specify per return-instance private data to be part of each
157kretprobe_instance object. This is especially useful when sharing private
158data between corresponding user entry and return handlers. The size of each
159private data object can be specified at kretprobe registration time by
160setting the data_size field of the kretprobe struct. This data can be
161accessed through the data field of each kretprobe_instance object.
162
163In case probed function is entered but there is no kretprobe_instance
164object available, then in addition to incrementing the nmissed count,
165the user entry_handler invocation is also skipped.
166
167.. _kprobes_jump_optimization:
168
169How Does Jump Optimization Work?
170--------------------------------
171
172If your kernel is built with CONFIG_OPTPROBES=y (currently this flag
173is automatically set 'y' on x86/x86-64, non-preemptive kernel) and
174the "debug.kprobes_optimization" kernel parameter is set to 1 (see
175sysctl(8)), Kprobes tries to reduce probe-hit overhead by using a jump
176instruction instead of a breakpoint instruction at each probepoint.
177
178Init a Kprobe
179^^^^^^^^^^^^^
180
181When a probe is registered, before attempting this optimization,
182Kprobes inserts an ordinary, breakpoint-based kprobe at the specified
183address. So, even if it's not possible to optimize this particular
184probepoint, there'll be a probe there.
185
186Safety Check
187^^^^^^^^^^^^
188
189Before optimizing a probe, Kprobes performs the following safety checks:
190
191- Kprobes verifies that the region that will be replaced by the jump
192  instruction (the "optimized region") lies entirely within one function.
193  (A jump instruction is multiple bytes, and so may overlay multiple
194  instructions.)
195
196- Kprobes analyzes the entire function and verifies that there is no
197  jump into the optimized region.  Specifically:
198
199  - the function contains no indirect jump;
200  - the function contains no instruction that causes an exception (since
201    the fixup code triggered by the exception could jump back into the
202    optimized region -- Kprobes checks the exception tables to verify this);
203  - there is no near jump to the optimized region (other than to the first
204    byte).
205
206- For each instruction in the optimized region, Kprobes verifies that
207  the instruction can be executed out of line.
208
209Preparing Detour Buffer
210^^^^^^^^^^^^^^^^^^^^^^^
211
212Next, Kprobes prepares a "detour" buffer, which contains the following
213instruction sequence:
214
215- code to push the CPU's registers (emulating a breakpoint trap)
216- a call to the trampoline code which calls user's probe handlers.
217- code to restore registers
218- the instructions from the optimized region
219- a jump back to the original execution path.
220
221Pre-optimization
222^^^^^^^^^^^^^^^^
223
224After preparing the detour buffer, Kprobes verifies that none of the
225following situations exist:
226
227- The probe has a post_handler.
228- Other instructions in the optimized region are probed.
229- The probe is disabled.
230
231In any of the above cases, Kprobes won't start optimizing the probe.
232Since these are temporary situations, Kprobes tries to start
233optimizing it again if the situation is changed.
234
235If the kprobe can be optimized, Kprobes enqueues the kprobe to an
236optimizing list, and kicks the kprobe-optimizer workqueue to optimize
237it.  If the to-be-optimized probepoint is hit before being optimized,
238Kprobes returns control to the original instruction path by setting
239the CPU's instruction pointer to the copied code in the detour buffer
240-- thus at least avoiding the single-step.
241
242Optimization
243^^^^^^^^^^^^
244
245The Kprobe-optimizer doesn't insert the jump instruction immediately;
246rather, it calls synchronize_rcu() for safety first, because it's
247possible for a CPU to be interrupted in the middle of executing the
248optimized region [3]_.  As you know, synchronize_rcu() can ensure
249that all interruptions that were active when synchronize_rcu()
250was called are done, but only if CONFIG_PREEMPT=n.  So, this version
251of kprobe optimization supports only kernels with CONFIG_PREEMPT=n [4]_.
252
253After that, the Kprobe-optimizer calls stop_machine() to replace
254the optimized region with a jump instruction to the detour buffer,
255using text_poke_smp().
256
257Unoptimization
258^^^^^^^^^^^^^^
259
260When an optimized kprobe is unregistered, disabled, or blocked by
261another kprobe, it will be unoptimized.  If this happens before
262the optimization is complete, the kprobe is just dequeued from the
263optimized list.  If the optimization has been done, the jump is
264replaced with the original code (except for an int3 breakpoint in
265the first byte) by using text_poke_smp().
266
267.. [3] Please imagine that the 2nd instruction is interrupted and then
268   the optimizer replaces the 2nd instruction with the jump *address*
269   while the interrupt handler is running. When the interrupt
270   returns to original address, there is no valid instruction,
271   and it causes an unexpected result.
272
273.. [4] This optimization-safety checking may be replaced with the
274   stop-machine method that ksplice uses for supporting a CONFIG_PREEMPT=y
275   kernel.
276
277NOTE for geeks:
278The jump optimization changes the kprobe's pre_handler behavior.
279Without optimization, the pre_handler can change the kernel's execution
280path by changing regs->ip and returning 1.  However, when the probe
281is optimized, that modification is ignored.  Thus, if you want to
282tweak the kernel's execution path, you need to suppress optimization,
283using one of the following techniques:
284
285- Specify an empty function for the kprobe's post_handler.
286
287or
288
289- Execute 'sysctl -w debug.kprobes_optimization=n'
290
291.. _kprobes_blacklist:
292
293Blacklist
294---------
295
296Kprobes can probe most of the kernel except itself. This means
297that there are some functions where kprobes cannot probe. Probing
298(trapping) such functions can cause a recursive trap (e.g. double
299fault) or the nested probe handler may never be called.
300Kprobes manages such functions as a blacklist.
301If you want to add a function into the blacklist, you just need
302to (1) include linux/kprobes.h and (2) use NOKPROBE_SYMBOL() macro
303to specify a blacklisted function.
304Kprobes checks the given probe address against the blacklist and
305rejects registering it, if the given address is in the blacklist.
306
307.. _kprobes_archs_supported:
308
309Architectures Supported
310=======================
311
312Kprobes and return probes are implemented on the following
313architectures:
314
315- i386 (Supports jump optimization)
316- x86_64 (AMD-64, EM64T) (Supports jump optimization)
317- ppc64
318- ia64 (Does not support probes on instruction slot1.)
319- sparc64 (Return probes not yet implemented.)
320- arm
321- ppc
322- mips
323- s390
324- parisc
325
326Configuring Kprobes
327===================
328
329When configuring the kernel using make menuconfig/xconfig/oldconfig,
330ensure that CONFIG_KPROBES is set to "y", look for "Kprobes" under
331"General architecture-dependent options".
332
333So that you can load and unload Kprobes-based instrumentation modules,
334make sure "Loadable module support" (CONFIG_MODULES) and "Module
335unloading" (CONFIG_MODULE_UNLOAD) are set to "y".
336
337Also make sure that CONFIG_KALLSYMS and perhaps even CONFIG_KALLSYMS_ALL
338are set to "y", since kallsyms_lookup_name() is used by the in-kernel
339kprobe address resolution code.
340
341If you need to insert a probe in the middle of a function, you may find
342it useful to "Compile the kernel with debug info" (CONFIG_DEBUG_INFO),
343so you can use "objdump -d -l vmlinux" to see the source-to-object
344code mapping.
345
346API Reference
347=============
348
349The Kprobes API includes a "register" function and an "unregister"
350function for each type of probe. The API also includes "register_*probes"
351and "unregister_*probes" functions for (un)registering arrays of probes.
352Here are terse, mini-man-page specifications for these functions and
353the associated probe handlers that you'll write. See the files in the
354samples/kprobes/ sub-directory for examples.
355
356register_kprobe
357---------------
358
359::
360
361	#include <linux/kprobes.h>
362	int register_kprobe(struct kprobe *kp);
363
364Sets a breakpoint at the address kp->addr.  When the breakpoint is hit, Kprobes
365calls kp->pre_handler.  After the probed instruction is single-stepped, Kprobe
366calls kp->post_handler.  Any or all handlers can be NULL. If kp->flags is set
367KPROBE_FLAG_DISABLED, that kp will be registered but disabled, so, its handlers
368aren't hit until calling enable_kprobe(kp).
369
370.. note::
371
372   1. With the introduction of the "symbol_name" field to struct kprobe,
373      the probepoint address resolution will now be taken care of by the kernel.
374      The following will now work::
375
376	kp.symbol_name = "symbol_name";
377
378      (64-bit powerpc intricacies such as function descriptors are handled
379      transparently)
380
381   2. Use the "offset" field of struct kprobe if the offset into the symbol
382      to install a probepoint is known. This field is used to calculate the
383      probepoint.
384
385   3. Specify either the kprobe "symbol_name" OR the "addr". If both are
386      specified, kprobe registration will fail with -EINVAL.
387
388   4. With CISC architectures (such as i386 and x86_64), the kprobes code
389      does not validate if the kprobe.addr is at an instruction boundary.
390      Use "offset" with caution.
391
392register_kprobe() returns 0 on success, or a negative errno otherwise.
393
394User's pre-handler (kp->pre_handler)::
395
396	#include <linux/kprobes.h>
397	#include <linux/ptrace.h>
398	int pre_handler(struct kprobe *p, struct pt_regs *regs);
399
400Called with p pointing to the kprobe associated with the breakpoint,
401and regs pointing to the struct containing the registers saved when
402the breakpoint was hit.  Return 0 here unless you're a Kprobes geek.
403
404User's post-handler (kp->post_handler)::
405
406	#include <linux/kprobes.h>
407	#include <linux/ptrace.h>
408	void post_handler(struct kprobe *p, struct pt_regs *regs,
409			  unsigned long flags);
410
411p and regs are as described for the pre_handler.  flags always seems
412to be zero.
413
414register_kretprobe
415------------------
416
417::
418
419	#include <linux/kprobes.h>
420	int register_kretprobe(struct kretprobe *rp);
421
422Establishes a return probe for the function whose address is
423rp->kp.addr.  When that function returns, Kprobes calls rp->handler.
424You must set rp->maxactive appropriately before you call
425register_kretprobe(); see "How Does a Return Probe Work?" for details.
426
427register_kretprobe() returns 0 on success, or a negative errno
428otherwise.
429
430User's return-probe handler (rp->handler)::
431
432	#include <linux/kprobes.h>
433	#include <linux/ptrace.h>
434	int kretprobe_handler(struct kretprobe_instance *ri,
435			      struct pt_regs *regs);
436
437regs is as described for kprobe.pre_handler.  ri points to the
438kretprobe_instance object, of which the following fields may be
439of interest:
440
441- ret_addr: the return address
442- rp: points to the corresponding kretprobe object
443- task: points to the corresponding task struct
444- data: points to per return-instance private data; see "Kretprobe
445	entry-handler" for details.
446
447The regs_return_value(regs) macro provides a simple abstraction to
448extract the return value from the appropriate register as defined by
449the architecture's ABI.
450
451The handler's return value is currently ignored.
452
453unregister_*probe
454------------------
455
456::
457
458	#include <linux/kprobes.h>
459	void unregister_kprobe(struct kprobe *kp);
460	void unregister_kretprobe(struct kretprobe *rp);
461
462Removes the specified probe.  The unregister function can be called
463at any time after the probe has been registered.
464
465.. note::
466
467   If the functions find an incorrect probe (ex. an unregistered probe),
468   they clear the addr field of the probe.
469
470register_*probes
471----------------
472
473::
474
475	#include <linux/kprobes.h>
476	int register_kprobes(struct kprobe **kps, int num);
477	int register_kretprobes(struct kretprobe **rps, int num);
478
479Registers each of the num probes in the specified array.  If any
480error occurs during registration, all probes in the array, up to
481the bad probe, are safely unregistered before the register_*probes
482function returns.
483
484- kps/rps: an array of pointers to ``*probe`` data structures
485- num: the number of the array entries.
486
487.. note::
488
489   You have to allocate(or define) an array of pointers and set all
490   of the array entries before using these functions.
491
492unregister_*probes
493------------------
494
495::
496
497	#include <linux/kprobes.h>
498	void unregister_kprobes(struct kprobe **kps, int num);
499	void unregister_kretprobes(struct kretprobe **rps, int num);
500
501Removes each of the num probes in the specified array at once.
502
503.. note::
504
505   If the functions find some incorrect probes (ex. unregistered
506   probes) in the specified array, they clear the addr field of those
507   incorrect probes. However, other probes in the array are
508   unregistered correctly.
509
510disable_*probe
511--------------
512
513::
514
515	#include <linux/kprobes.h>
516	int disable_kprobe(struct kprobe *kp);
517	int disable_kretprobe(struct kretprobe *rp);
518
519Temporarily disables the specified ``*probe``. You can enable it again by using
520enable_*probe(). You must specify the probe which has been registered.
521
522enable_*probe
523-------------
524
525::
526
527	#include <linux/kprobes.h>
528	int enable_kprobe(struct kprobe *kp);
529	int enable_kretprobe(struct kretprobe *rp);
530
531Enables ``*probe`` which has been disabled by disable_*probe(). You must specify
532the probe which has been registered.
533
534Kprobes Features and Limitations
535================================
536
537Kprobes allows multiple probes at the same address. Also,
538a probepoint for which there is a post_handler cannot be optimized.
539So if you install a kprobe with a post_handler, at an optimized
540probepoint, the probepoint will be unoptimized automatically.
541
542In general, you can install a probe anywhere in the kernel.
543In particular, you can probe interrupt handlers.  Known exceptions
544are discussed in this section.
545
546The register_*probe functions will return -EINVAL if you attempt
547to install a probe in the code that implements Kprobes (mostly
548kernel/kprobes.c and ``arch/*/kernel/kprobes.c``, but also functions such
549as do_page_fault and notifier_call_chain).
550
551If you install a probe in an inline-able function, Kprobes makes
552no attempt to chase down all inline instances of the function and
553install probes there.  gcc may inline a function without being asked,
554so keep this in mind if you're not seeing the probe hits you expect.
555
556A probe handler can modify the environment of the probed function
557-- e.g., by modifying kernel data structures, or by modifying the
558contents of the pt_regs struct (which are restored to the registers
559upon return from the breakpoint).  So Kprobes can be used, for example,
560to install a bug fix or to inject faults for testing.  Kprobes, of
561course, has no way to distinguish the deliberately injected faults
562from the accidental ones.  Don't drink and probe.
563
564Kprobes makes no attempt to prevent probe handlers from stepping on
565each other -- e.g., probing printk() and then calling printk() from a
566probe handler.  If a probe handler hits a probe, that second probe's
567handlers won't be run in that instance, and the kprobe.nmissed member
568of the second probe will be incremented.
569
570As of Linux v2.6.15-rc1, multiple handlers (or multiple instances of
571the same handler) may run concurrently on different CPUs.
572
573Kprobes does not use mutexes or allocate memory except during
574registration and unregistration.
575
576Probe handlers are run with preemption disabled or interrupt disabled,
577which depends on the architecture and optimization state.  (e.g.,
578kretprobe handlers and optimized kprobe handlers run without interrupt
579disabled on x86/x86-64).  In any case, your handler should not yield
580the CPU (e.g., by attempting to acquire a semaphore, or waiting I/O).
581
582Since a return probe is implemented by replacing the return
583address with the trampoline's address, stack backtraces and calls
584to __builtin_return_address() will typically yield the trampoline's
585address instead of the real return address for kretprobed functions.
586(As far as we can tell, __builtin_return_address() is used only
587for instrumentation and error reporting.)
588
589If the number of times a function is called does not match the number
590of times it returns, registering a return probe on that function may
591produce undesirable results. In such a case, a line:
592kretprobe BUG!: Processing kretprobe d000000000041aa8 @ c00000000004f48c
593gets printed. With this information, one will be able to correlate the
594exact instance of the kretprobe that caused the problem. We have the
595do_exit() case covered. do_execve() and do_fork() are not an issue.
596We're unaware of other specific cases where this could be a problem.
597
598If, upon entry to or exit from a function, the CPU is running on
599a stack other than that of the current task, registering a return
600probe on that function may produce undesirable results.  For this
601reason, Kprobes doesn't support return probes (or kprobes)
602on the x86_64 version of __switch_to(); the registration functions
603return -EINVAL.
604
605On x86/x86-64, since the Jump Optimization of Kprobes modifies
606instructions widely, there are some limitations to optimization. To
607explain it, we introduce some terminology. Imagine a 3-instruction
608sequence consisting of a two 2-byte instructions and one 3-byte
609instruction.
610
611::
612
613		IA
614		|
615	[-2][-1][0][1][2][3][4][5][6][7]
616		[ins1][ins2][  ins3 ]
617		[<-     DCR       ->]
618		[<- JTPR ->]
619
620	ins1: 1st Instruction
621	ins2: 2nd Instruction
622	ins3: 3rd Instruction
623	IA:  Insertion Address
624	JTPR: Jump Target Prohibition Region
625	DCR: Detoured Code Region
626
627The instructions in DCR are copied to the out-of-line buffer
628of the kprobe, because the bytes in DCR are replaced by
629a 5-byte jump instruction. So there are several limitations.
630
631a) The instructions in DCR must be relocatable.
632b) The instructions in DCR must not include a call instruction.
633c) JTPR must not be targeted by any jump or call instruction.
634d) DCR must not straddle the border between functions.
635
636Anyway, these limitations are checked by the in-kernel instruction
637decoder, so you don't need to worry about that.
638
639Probe Overhead
640==============
641
642On a typical CPU in use in 2005, a kprobe hit takes 0.5 to 1.0
643microseconds to process.  Specifically, a benchmark that hits the same
644probepoint repeatedly, firing a simple handler each time, reports 1-2
645million hits per second, depending on the architecture.  A return-probe
646hit typically takes 50-75% longer than a kprobe hit.
647When you have a return probe set on a function, adding a kprobe at
648the entry to that function adds essentially no overhead.
649
650Here are sample overhead figures (in usec) for different architectures::
651
652  k = kprobe; r = return probe; kr = kprobe + return probe
653  on same function
654
655  i386: Intel Pentium M, 1495 MHz, 2957.31 bogomips
656  k = 0.57 usec; r = 0.92; kr = 0.99
657
658  x86_64: AMD Opteron 246, 1994 MHz, 3971.48 bogomips
659  k = 0.49 usec; r = 0.80; kr = 0.82
660
661  ppc64: POWER5 (gr), 1656 MHz (SMT disabled, 1 virtual CPU per physical CPU)
662  k = 0.77 usec; r = 1.26; kr = 1.45
663
664Optimized Probe Overhead
665------------------------
666
667Typically, an optimized kprobe hit takes 0.07 to 0.1 microseconds to
668process. Here are sample overhead figures (in usec) for x86 architectures::
669
670  k = unoptimized kprobe, b = boosted (single-step skipped), o = optimized kprobe,
671  r = unoptimized kretprobe, rb = boosted kretprobe, ro = optimized kretprobe.
672
673  i386: Intel(R) Xeon(R) E5410, 2.33GHz, 4656.90 bogomips
674  k = 0.80 usec; b = 0.33; o = 0.05; r = 1.10; rb = 0.61; ro = 0.33
675
676  x86-64: Intel(R) Xeon(R) E5410, 2.33GHz, 4656.90 bogomips
677  k = 0.99 usec; b = 0.43; o = 0.06; r = 1.24; rb = 0.68; ro = 0.30
678
679TODO
680====
681
682a. SystemTap (http://sourceware.org/systemtap): Provides a simplified
683   programming interface for probe-based instrumentation.  Try it out.
684b. Kernel return probes for sparc64.
685c. Support for other architectures.
686d. User-space probes.
687e. Watchpoint probes (which fire on data references).
688
689Kprobes Example
690===============
691
692See samples/kprobes/kprobe_example.c
693
694Kretprobes Example
695==================
696
697See samples/kprobes/kretprobe_example.c
698
699Deprecated Features
700===================
701
702Jprobes is now a deprecated feature. People who are depending on it should
703migrate to other tracing features or use older kernels. Please consider to
704migrate your tool to one of the following options:
705
706- Use trace-event to trace target function with arguments.
707
708  trace-event is a low-overhead (and almost no visible overhead if it
709  is off) statically defined event interface. You can define new events
710  and trace it via ftrace or any other tracing tools.
711
712  See the following urls:
713
714    - https://lwn.net/Articles/379903/
715    - https://lwn.net/Articles/381064/
716    - https://lwn.net/Articles/383362/
717
718- Use ftrace dynamic events (kprobe event) with perf-probe.
719
720  If you build your kernel with debug info (CONFIG_DEBUG_INFO=y), you can
721  find which register/stack is assigned to which local variable or arguments
722  by using perf-probe and set up new event to trace it.
723
724  See following documents:
725
726  - Documentation/trace/kprobetrace.rst
727  - Documentation/trace/events.rst
728  - tools/perf/Documentation/perf-probe.txt
729
730
731The kprobes debugfs interface
732=============================
733
734
735With recent kernels (> 2.6.20) the list of registered kprobes is visible
736under the /sys/kernel/debug/kprobes/ directory (assuming debugfs is mounted at //sys/kernel/debug).
737
738/sys/kernel/debug/kprobes/list: Lists all registered probes on the system::
739
740	c015d71a  k  vfs_read+0x0
741	c03dedc5  r  tcp_v4_rcv+0x0
742
743The first column provides the kernel address where the probe is inserted.
744The second column identifies the type of probe (k - kprobe and r - kretprobe)
745while the third column specifies the symbol+offset of the probe.
746If the probed function belongs to a module, the module name is also
747specified. Following columns show probe status. If the probe is on
748a virtual address that is no longer valid (module init sections, module
749virtual addresses that correspond to modules that've been unloaded),
750such probes are marked with [GONE]. If the probe is temporarily disabled,
751such probes are marked with [DISABLED]. If the probe is optimized, it is
752marked with [OPTIMIZED]. If the probe is ftrace-based, it is marked with
753[FTRACE].
754
755/sys/kernel/debug/kprobes/enabled: Turn kprobes ON/OFF forcibly.
756
757Provides a knob to globally and forcibly turn registered kprobes ON or OFF.
758By default, all kprobes are enabled. By echoing "0" to this file, all
759registered probes will be disarmed, till such time a "1" is echoed to this
760file. Note that this knob just disarms and arms all kprobes and doesn't
761change each probe's disabling state. This means that disabled kprobes (marked
762[DISABLED]) will be not enabled if you turn ON all kprobes by this knob.
763
764
765The kprobes sysctl interface
766============================
767
768/proc/sys/debug/kprobes-optimization: Turn kprobes optimization ON/OFF.
769
770When CONFIG_OPTPROBES=y, this sysctl interface appears and it provides
771a knob to globally and forcibly turn jump optimization (see section
772:ref:`kprobes_jump_optimization`) ON or OFF. By default, jump optimization
773is allowed (ON). If you echo "0" to this file or set
774"debug.kprobes_optimization" to 0 via sysctl, all optimized probes will be
775unoptimized, and any new probes registered after that will not be optimized.
776
777Note that this knob *changes* the optimized state. This means that optimized
778probes (marked [OPTIMIZED]) will be unoptimized ([OPTIMIZED] tag will be
779removed). If the knob is turned on, they will be optimized again.
780
781References
782==========
783
784For additional information on Kprobes, refer to the following URLs:
785
786- https://lwn.net/Articles/132196/
787- https://www.kernel.org/doc/ols/2006/ols2006v2-pages-109-124.pdf
788
789