xref: /linux/Documentation/trace/kprobes.rst (revision 3df692169e8486fc3dd91fcd5ea81c27a0bac033)
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- sparc64 (Return probes not yet implemented.)
319- arm
320- ppc
321- mips
322- s390
323- parisc
324
325Configuring Kprobes
326===================
327
328When configuring the kernel using make menuconfig/xconfig/oldconfig,
329ensure that CONFIG_KPROBES is set to "y", look for "Kprobes" under
330"General architecture-dependent options".
331
332So that you can load and unload Kprobes-based instrumentation modules,
333make sure "Loadable module support" (CONFIG_MODULES) and "Module
334unloading" (CONFIG_MODULE_UNLOAD) are set to "y".
335
336Also make sure that CONFIG_KALLSYMS and perhaps even CONFIG_KALLSYMS_ALL
337are set to "y", since kallsyms_lookup_name() is used by the in-kernel
338kprobe address resolution code.
339
340If you need to insert a probe in the middle of a function, you may find
341it useful to "Compile the kernel with debug info" (CONFIG_DEBUG_INFO),
342so you can use "objdump -d -l vmlinux" to see the source-to-object
343code mapping.
344
345API Reference
346=============
347
348The Kprobes API includes a "register" function and an "unregister"
349function for each type of probe. The API also includes "register_*probes"
350and "unregister_*probes" functions for (un)registering arrays of probes.
351Here are terse, mini-man-page specifications for these functions and
352the associated probe handlers that you'll write. See the files in the
353samples/kprobes/ sub-directory for examples.
354
355register_kprobe
356---------------
357
358::
359
360	#include <linux/kprobes.h>
361	int register_kprobe(struct kprobe *kp);
362
363Sets a breakpoint at the address kp->addr.  When the breakpoint is hit, Kprobes
364calls kp->pre_handler.  After the probed instruction is single-stepped, Kprobe
365calls kp->post_handler.  Any or all handlers can be NULL. If kp->flags is set
366KPROBE_FLAG_DISABLED, that kp will be registered but disabled, so, its handlers
367aren't hit until calling enable_kprobe(kp).
368
369.. note::
370
371   1. With the introduction of the "symbol_name" field to struct kprobe,
372      the probepoint address resolution will now be taken care of by the kernel.
373      The following will now work::
374
375	kp.symbol_name = "symbol_name";
376
377      (64-bit powerpc intricacies such as function descriptors are handled
378      transparently)
379
380   2. Use the "offset" field of struct kprobe if the offset into the symbol
381      to install a probepoint is known. This field is used to calculate the
382      probepoint.
383
384   3. Specify either the kprobe "symbol_name" OR the "addr". If both are
385      specified, kprobe registration will fail with -EINVAL.
386
387   4. With CISC architectures (such as i386 and x86_64), the kprobes code
388      does not validate if the kprobe.addr is at an instruction boundary.
389      Use "offset" with caution.
390
391register_kprobe() returns 0 on success, or a negative errno otherwise.
392
393User's pre-handler (kp->pre_handler)::
394
395	#include <linux/kprobes.h>
396	#include <linux/ptrace.h>
397	int pre_handler(struct kprobe *p, struct pt_regs *regs);
398
399Called with p pointing to the kprobe associated with the breakpoint,
400and regs pointing to the struct containing the registers saved when
401the breakpoint was hit.  Return 0 here unless you're a Kprobes geek.
402
403User's post-handler (kp->post_handler)::
404
405	#include <linux/kprobes.h>
406	#include <linux/ptrace.h>
407	void post_handler(struct kprobe *p, struct pt_regs *regs,
408			  unsigned long flags);
409
410p and regs are as described for the pre_handler.  flags always seems
411to be zero.
412
413register_kretprobe
414------------------
415
416::
417
418	#include <linux/kprobes.h>
419	int register_kretprobe(struct kretprobe *rp);
420
421Establishes a return probe for the function whose address is
422rp->kp.addr.  When that function returns, Kprobes calls rp->handler.
423You must set rp->maxactive appropriately before you call
424register_kretprobe(); see "How Does a Return Probe Work?" for details.
425
426register_kretprobe() returns 0 on success, or a negative errno
427otherwise.
428
429User's return-probe handler (rp->handler)::
430
431	#include <linux/kprobes.h>
432	#include <linux/ptrace.h>
433	int kretprobe_handler(struct kretprobe_instance *ri,
434			      struct pt_regs *regs);
435
436regs is as described for kprobe.pre_handler.  ri points to the
437kretprobe_instance object, of which the following fields may be
438of interest:
439
440- ret_addr: the return address
441- rp: points to the corresponding kretprobe object
442- task: points to the corresponding task struct
443- data: points to per return-instance private data; see "Kretprobe
444	entry-handler" for details.
445
446The regs_return_value(regs) macro provides a simple abstraction to
447extract the return value from the appropriate register as defined by
448the architecture's ABI.
449
450The handler's return value is currently ignored.
451
452unregister_*probe
453------------------
454
455::
456
457	#include <linux/kprobes.h>
458	void unregister_kprobe(struct kprobe *kp);
459	void unregister_kretprobe(struct kretprobe *rp);
460
461Removes the specified probe.  The unregister function can be called
462at any time after the probe has been registered.
463
464.. note::
465
466   If the functions find an incorrect probe (ex. an unregistered probe),
467   they clear the addr field of the probe.
468
469register_*probes
470----------------
471
472::
473
474	#include <linux/kprobes.h>
475	int register_kprobes(struct kprobe **kps, int num);
476	int register_kretprobes(struct kretprobe **rps, int num);
477
478Registers each of the num probes in the specified array.  If any
479error occurs during registration, all probes in the array, up to
480the bad probe, are safely unregistered before the register_*probes
481function returns.
482
483- kps/rps: an array of pointers to ``*probe`` data structures
484- num: the number of the array entries.
485
486.. note::
487
488   You have to allocate(or define) an array of pointers and set all
489   of the array entries before using these functions.
490
491unregister_*probes
492------------------
493
494::
495
496	#include <linux/kprobes.h>
497	void unregister_kprobes(struct kprobe **kps, int num);
498	void unregister_kretprobes(struct kretprobe **rps, int num);
499
500Removes each of the num probes in the specified array at once.
501
502.. note::
503
504   If the functions find some incorrect probes (ex. unregistered
505   probes) in the specified array, they clear the addr field of those
506   incorrect probes. However, other probes in the array are
507   unregistered correctly.
508
509disable_*probe
510--------------
511
512::
513
514	#include <linux/kprobes.h>
515	int disable_kprobe(struct kprobe *kp);
516	int disable_kretprobe(struct kretprobe *rp);
517
518Temporarily disables the specified ``*probe``. You can enable it again by using
519enable_*probe(). You must specify the probe which has been registered.
520
521enable_*probe
522-------------
523
524::
525
526	#include <linux/kprobes.h>
527	int enable_kprobe(struct kprobe *kp);
528	int enable_kretprobe(struct kretprobe *rp);
529
530Enables ``*probe`` which has been disabled by disable_*probe(). You must specify
531the probe which has been registered.
532
533Kprobes Features and Limitations
534================================
535
536Kprobes allows multiple probes at the same address. Also,
537a probepoint for which there is a post_handler cannot be optimized.
538So if you install a kprobe with a post_handler, at an optimized
539probepoint, the probepoint will be unoptimized automatically.
540
541In general, you can install a probe anywhere in the kernel.
542In particular, you can probe interrupt handlers.  Known exceptions
543are discussed in this section.
544
545The register_*probe functions will return -EINVAL if you attempt
546to install a probe in the code that implements Kprobes (mostly
547kernel/kprobes.c and ``arch/*/kernel/kprobes.c``, but also functions such
548as do_page_fault and notifier_call_chain).
549
550If you install a probe in an inline-able function, Kprobes makes
551no attempt to chase down all inline instances of the function and
552install probes there.  gcc may inline a function without being asked,
553so keep this in mind if you're not seeing the probe hits you expect.
554
555A probe handler can modify the environment of the probed function
556-- e.g., by modifying kernel data structures, or by modifying the
557contents of the pt_regs struct (which are restored to the registers
558upon return from the breakpoint).  So Kprobes can be used, for example,
559to install a bug fix or to inject faults for testing.  Kprobes, of
560course, has no way to distinguish the deliberately injected faults
561from the accidental ones.  Don't drink and probe.
562
563Kprobes makes no attempt to prevent probe handlers from stepping on
564each other -- e.g., probing printk() and then calling printk() from a
565probe handler.  If a probe handler hits a probe, that second probe's
566handlers won't be run in that instance, and the kprobe.nmissed member
567of the second probe will be incremented.
568
569As of Linux v2.6.15-rc1, multiple handlers (or multiple instances of
570the same handler) may run concurrently on different CPUs.
571
572Kprobes does not use mutexes or allocate memory except during
573registration and unregistration.
574
575Probe handlers are run with preemption disabled or interrupt disabled,
576which depends on the architecture and optimization state.  (e.g.,
577kretprobe handlers and optimized kprobe handlers run without interrupt
578disabled on x86/x86-64).  In any case, your handler should not yield
579the CPU (e.g., by attempting to acquire a semaphore, or waiting I/O).
580
581Since a return probe is implemented by replacing the return
582address with the trampoline's address, stack backtraces and calls
583to __builtin_return_address() will typically yield the trampoline's
584address instead of the real return address for kretprobed functions.
585(As far as we can tell, __builtin_return_address() is used only
586for instrumentation and error reporting.)
587
588If the number of times a function is called does not match the number
589of times it returns, registering a return probe on that function may
590produce undesirable results. In such a case, a line:
591kretprobe BUG!: Processing kretprobe d000000000041aa8 @ c00000000004f48c
592gets printed. With this information, one will be able to correlate the
593exact instance of the kretprobe that caused the problem. We have the
594do_exit() case covered. do_execve() and do_fork() are not an issue.
595We're unaware of other specific cases where this could be a problem.
596
597If, upon entry to or exit from a function, the CPU is running on
598a stack other than that of the current task, registering a return
599probe on that function may produce undesirable results.  For this
600reason, Kprobes doesn't support return probes (or kprobes)
601on the x86_64 version of __switch_to(); the registration functions
602return -EINVAL.
603
604On x86/x86-64, since the Jump Optimization of Kprobes modifies
605instructions widely, there are some limitations to optimization. To
606explain it, we introduce some terminology. Imagine a 3-instruction
607sequence consisting of a two 2-byte instructions and one 3-byte
608instruction.
609
610::
611
612		IA
613		|
614	[-2][-1][0][1][2][3][4][5][6][7]
615		[ins1][ins2][  ins3 ]
616		[<-     DCR       ->]
617		[<- JTPR ->]
618
619	ins1: 1st Instruction
620	ins2: 2nd Instruction
621	ins3: 3rd Instruction
622	IA:  Insertion Address
623	JTPR: Jump Target Prohibition Region
624	DCR: Detoured Code Region
625
626The instructions in DCR are copied to the out-of-line buffer
627of the kprobe, because the bytes in DCR are replaced by
628a 5-byte jump instruction. So there are several limitations.
629
630a) The instructions in DCR must be relocatable.
631b) The instructions in DCR must not include a call instruction.
632c) JTPR must not be targeted by any jump or call instruction.
633d) DCR must not straddle the border between functions.
634
635Anyway, these limitations are checked by the in-kernel instruction
636decoder, so you don't need to worry about that.
637
638Probe Overhead
639==============
640
641On a typical CPU in use in 2005, a kprobe hit takes 0.5 to 1.0
642microseconds to process.  Specifically, a benchmark that hits the same
643probepoint repeatedly, firing a simple handler each time, reports 1-2
644million hits per second, depending on the architecture.  A return-probe
645hit typically takes 50-75% longer than a kprobe hit.
646When you have a return probe set on a function, adding a kprobe at
647the entry to that function adds essentially no overhead.
648
649Here are sample overhead figures (in usec) for different architectures::
650
651  k = kprobe; r = return probe; kr = kprobe + return probe
652  on same function
653
654  i386: Intel Pentium M, 1495 MHz, 2957.31 bogomips
655  k = 0.57 usec; r = 0.92; kr = 0.99
656
657  x86_64: AMD Opteron 246, 1994 MHz, 3971.48 bogomips
658  k = 0.49 usec; r = 0.80; kr = 0.82
659
660  ppc64: POWER5 (gr), 1656 MHz (SMT disabled, 1 virtual CPU per physical CPU)
661  k = 0.77 usec; r = 1.26; kr = 1.45
662
663Optimized Probe Overhead
664------------------------
665
666Typically, an optimized kprobe hit takes 0.07 to 0.1 microseconds to
667process. Here are sample overhead figures (in usec) for x86 architectures::
668
669  k = unoptimized kprobe, b = boosted (single-step skipped), o = optimized kprobe,
670  r = unoptimized kretprobe, rb = boosted kretprobe, ro = optimized kretprobe.
671
672  i386: Intel(R) Xeon(R) E5410, 2.33GHz, 4656.90 bogomips
673  k = 0.80 usec; b = 0.33; o = 0.05; r = 1.10; rb = 0.61; ro = 0.33
674
675  x86-64: Intel(R) Xeon(R) E5410, 2.33GHz, 4656.90 bogomips
676  k = 0.99 usec; b = 0.43; o = 0.06; r = 1.24; rb = 0.68; ro = 0.30
677
678TODO
679====
680
681a. SystemTap (http://sourceware.org/systemtap): Provides a simplified
682   programming interface for probe-based instrumentation.  Try it out.
683b. Kernel return probes for sparc64.
684c. Support for other architectures.
685d. User-space probes.
686e. Watchpoint probes (which fire on data references).
687
688Kprobes Example
689===============
690
691See samples/kprobes/kprobe_example.c
692
693Kretprobes Example
694==================
695
696See samples/kprobes/kretprobe_example.c
697
698Deprecated Features
699===================
700
701Jprobes is now a deprecated feature. People who are depending on it should
702migrate to other tracing features or use older kernels. Please consider to
703migrate your tool to one of the following options:
704
705- Use trace-event to trace target function with arguments.
706
707  trace-event is a low-overhead (and almost no visible overhead if it
708  is off) statically defined event interface. You can define new events
709  and trace it via ftrace or any other tracing tools.
710
711  See the following urls:
712
713    - https://lwn.net/Articles/379903/
714    - https://lwn.net/Articles/381064/
715    - https://lwn.net/Articles/383362/
716
717- Use ftrace dynamic events (kprobe event) with perf-probe.
718
719  If you build your kernel with debug info (CONFIG_DEBUG_INFO=y), you can
720  find which register/stack is assigned to which local variable or arguments
721  by using perf-probe and set up new event to trace it.
722
723  See following documents:
724
725  - Documentation/trace/kprobetrace.rst
726  - Documentation/trace/events.rst
727  - tools/perf/Documentation/perf-probe.txt
728
729
730The kprobes debugfs interface
731=============================
732
733
734With recent kernels (> 2.6.20) the list of registered kprobes is visible
735under the /sys/kernel/debug/kprobes/ directory (assuming debugfs is mounted at //sys/kernel/debug).
736
737/sys/kernel/debug/kprobes/list: Lists all registered probes on the system::
738
739	c015d71a  k  vfs_read+0x0
740	c03dedc5  r  tcp_v4_rcv+0x0
741
742The first column provides the kernel address where the probe is inserted.
743The second column identifies the type of probe (k - kprobe and r - kretprobe)
744while the third column specifies the symbol+offset of the probe.
745If the probed function belongs to a module, the module name is also
746specified. Following columns show probe status. If the probe is on
747a virtual address that is no longer valid (module init sections, module
748virtual addresses that correspond to modules that've been unloaded),
749such probes are marked with [GONE]. If the probe is temporarily disabled,
750such probes are marked with [DISABLED]. If the probe is optimized, it is
751marked with [OPTIMIZED]. If the probe is ftrace-based, it is marked with
752[FTRACE].
753
754/sys/kernel/debug/kprobes/enabled: Turn kprobes ON/OFF forcibly.
755
756Provides a knob to globally and forcibly turn registered kprobes ON or OFF.
757By default, all kprobes are enabled. By echoing "0" to this file, all
758registered probes will be disarmed, till such time a "1" is echoed to this
759file. Note that this knob just disarms and arms all kprobes and doesn't
760change each probe's disabling state. This means that disabled kprobes (marked
761[DISABLED]) will be not enabled if you turn ON all kprobes by this knob.
762
763
764The kprobes sysctl interface
765============================
766
767/proc/sys/debug/kprobes-optimization: Turn kprobes optimization ON/OFF.
768
769When CONFIG_OPTPROBES=y, this sysctl interface appears and it provides
770a knob to globally and forcibly turn jump optimization (see section
771:ref:`kprobes_jump_optimization`) ON or OFF. By default, jump optimization
772is allowed (ON). If you echo "0" to this file or set
773"debug.kprobes_optimization" to 0 via sysctl, all optimized probes will be
774unoptimized, and any new probes registered after that will not be optimized.
775
776Note that this knob *changes* the optimized state. This means that optimized
777probes (marked [OPTIMIZED]) will be unoptimized ([OPTIMIZED] tag will be
778removed). If the knob is turned on, they will be optimized again.
779
780References
781==========
782
783For additional information on Kprobes, refer to the following URLs:
784
785- https://lwn.net/Articles/132196/
786- https://www.kernel.org/doc/ols/2006/ols2006v2-pages-109-124.pdf
787
788