Based on kernel version 3.16. Page generated on 2014-08-06 21:40 EST.
1 Title : Kernel Probes (Kprobes) 2 Authors : Jim Keniston <email@example.com> 3 : Prasanna S Panchamukhi <firstname.lastname@example.org> 4 : Masami Hiramatsu <email@example.com> 5 6 CONTENTS 7 8 1. Concepts: Kprobes, Jprobes, Return Probes 9 2. Architectures Supported 10 3. Configuring Kprobes 11 4. API Reference 12 5. Kprobes Features and Limitations 13 6. Probe Overhead 14 7. TODO 15 8. Kprobes Example 16 9. Jprobes Example 17 10. Kretprobes Example 18 Appendix A: The kprobes debugfs interface 19 Appendix B: The kprobes sysctl interface 20 21 1. Concepts: Kprobes, Jprobes, Return Probes 22 23 Kprobes enables you to dynamically break into any kernel routine and 24 collect debugging and performance information non-disruptively. You 25 can trap at almost any kernel code address(*), specifying a handler 26 routine to be invoked when the breakpoint is hit. 27 (*: some parts of the kernel code can not be trapped, see 1.5 Blacklist) 28 29 There are currently three types of probes: kprobes, jprobes, and 30 kretprobes (also called return probes). A kprobe can be inserted 31 on virtually any instruction in the kernel. A jprobe is inserted at 32 the entry to a kernel function, and provides convenient access to the 33 function's arguments. A return probe fires when a specified function 34 returns. 35 36 In the typical case, Kprobes-based instrumentation is packaged as 37 a kernel module. The module's init function installs ("registers") 38 one or more probes, and the exit function unregisters them. A 39 registration function such as register_kprobe() specifies where 40 the probe is to be inserted and what handler is to be called when 41 the probe is hit. 42 43 There are also register_/unregister_*probes() functions for batch 44 registration/unregistration of a group of *probes. These functions 45 can speed up unregistration process when you have to unregister 46 a lot of probes at once. 47 48 The next four subsections explain how the different types of 49 probes work and how jump optimization works. They explain certain 50 things that you'll need to know in order to make the best use of 51 Kprobes -- e.g., the difference between a pre_handler and 52 a post_handler, and how to use the maxactive and nmissed fields of 53 a kretprobe. But if you're in a hurry to start using Kprobes, you 54 can skip ahead to section 2. 55 56 1.1 How Does a Kprobe Work? 57 58 When a kprobe is registered, Kprobes makes a copy of the probed 59 instruction and replaces the first byte(s) of the probed instruction 60 with a breakpoint instruction (e.g., int3 on i386 and x86_64). 61 62 When a CPU hits the breakpoint instruction, a trap occurs, the CPU's 63 registers are saved, and control passes to Kprobes via the 64 notifier_call_chain mechanism. Kprobes executes the "pre_handler" 65 associated with the kprobe, passing the handler the addresses of the 66 kprobe struct and the saved registers. 67 68 Next, Kprobes single-steps its copy of the probed instruction. 69 (It would be simpler to single-step the actual instruction in place, 70 but then Kprobes would have to temporarily remove the breakpoint 71 instruction. This would open a small time window when another CPU 72 could sail right past the probepoint.) 73 74 After the instruction is single-stepped, Kprobes executes the 75 "post_handler," if any, that is associated with the kprobe. 76 Execution then continues with the instruction following the probepoint. 77 78 1.2 How Does a Jprobe Work? 79 80 A jprobe is implemented using a kprobe that is placed on a function's 81 entry point. It employs a simple mirroring principle to allow 82 seamless access to the probed function's arguments. The jprobe 83 handler routine should have the same signature (arg list and return 84 type) as the function being probed, and must always end by calling 85 the Kprobes function jprobe_return(). 86 87 Here's how it works. When the probe is hit, Kprobes makes a copy of 88 the saved registers and a generous portion of the stack (see below). 89 Kprobes then points the saved instruction pointer at the jprobe's 90 handler routine, and returns from the trap. As a result, control 91 passes to the handler, which is presented with the same register and 92 stack contents as the probed function. When it is done, the handler 93 calls jprobe_return(), which traps again to restore the original stack 94 contents and processor state and switch to the probed function. 95 96 By convention, the callee owns its arguments, so gcc may produce code 97 that unexpectedly modifies that portion of the stack. This is why 98 Kprobes saves a copy of the stack and restores it after the jprobe 99 handler has run. Up to MAX_STACK_SIZE bytes are copied -- e.g., 100 64 bytes on i386. 101 102 Note that the probed function's args may be passed on the stack 103 or in registers. The jprobe will work in either case, so long as the 104 handler's prototype matches that of the probed function. 105 106 1.3 Return Probes 107 108 1.3.1 How Does a Return Probe Work? 109 110 When you call register_kretprobe(), Kprobes establishes a kprobe at 111 the entry to the function. When the probed function is called and this 112 probe is hit, Kprobes saves a copy of the return address, and replaces 113 the return address with the address of a "trampoline." The trampoline 114 is an arbitrary piece of code -- typically just a nop instruction. 115 At boot time, Kprobes registers a kprobe at the trampoline. 116 117 When the probed function executes its return instruction, control 118 passes to the trampoline and that probe is hit. Kprobes' trampoline 119 handler calls the user-specified return handler associated with the 120 kretprobe, then sets the saved instruction pointer to the saved return 121 address, and that's where execution resumes upon return from the trap. 122 123 While the probed function is executing, its return address is 124 stored in an object of type kretprobe_instance. Before calling 125 register_kretprobe(), the user sets the maxactive field of the 126 kretprobe struct to specify how many instances of the specified 127 function can be probed simultaneously. register_kretprobe() 128 pre-allocates the indicated number of kretprobe_instance objects. 129 130 For example, if the function is non-recursive and is called with a 131 spinlock held, maxactive = 1 should be enough. If the function is 132 non-recursive and can never relinquish the CPU (e.g., via a semaphore 133 or preemption), NR_CPUS should be enough. If maxactive <= 0, it is 134 set to a default value. If CONFIG_PREEMPT is enabled, the default 135 is max(10, 2*NR_CPUS). Otherwise, the default is NR_CPUS. 136 137 It's not a disaster if you set maxactive too low; you'll just miss 138 some probes. In the kretprobe struct, the nmissed field is set to 139 zero when the return probe is registered, and is incremented every 140 time the probed function is entered but there is no kretprobe_instance 141 object available for establishing the return probe. 142 143 1.3.2 Kretprobe entry-handler 144 145 Kretprobes also provides an optional user-specified handler which runs 146 on function entry. This handler is specified by setting the entry_handler 147 field of the kretprobe struct. Whenever the kprobe placed by kretprobe at the 148 function entry is hit, the user-defined entry_handler, if any, is invoked. 149 If the entry_handler returns 0 (success) then a corresponding return handler 150 is guaranteed to be called upon function return. If the entry_handler 151 returns a non-zero error then Kprobes leaves the return address as is, and 152 the kretprobe has no further effect for that particular function instance. 153 154 Multiple entry and return handler invocations are matched using the unique 155 kretprobe_instance object associated with them. Additionally, a user 156 may also specify per return-instance private data to be part of each 157 kretprobe_instance object. This is especially useful when sharing private 158 data between corresponding user entry and return handlers. The size of each 159 private data object can be specified at kretprobe registration time by 160 setting the data_size field of the kretprobe struct. This data can be 161 accessed through the data field of each kretprobe_instance object. 162 163 In case probed function is entered but there is no kretprobe_instance 164 object available, then in addition to incrementing the nmissed count, 165 the user entry_handler invocation is also skipped. 166 167 1.4 How Does Jump Optimization Work? 168 169 If your kernel is built with CONFIG_OPTPROBES=y (currently this flag 170 is automatically set 'y' on x86/x86-64, non-preemptive kernel) and 171 the "debug.kprobes_optimization" kernel parameter is set to 1 (see 172 sysctl(8)), Kprobes tries to reduce probe-hit overhead by using a jump 173 instruction instead of a breakpoint instruction at each probepoint. 174 175 1.4.1 Init a Kprobe 176 177 When a probe is registered, before attempting this optimization, 178 Kprobes inserts an ordinary, breakpoint-based kprobe at the specified 179 address. So, even if it's not possible to optimize this particular 180 probepoint, there'll be a probe there. 181 182 1.4.2 Safety Check 183 184 Before optimizing a probe, Kprobes performs the following safety checks: 185 186 - Kprobes verifies that the region that will be replaced by the jump 187 instruction (the "optimized region") lies entirely within one function. 188 (A jump instruction is multiple bytes, and so may overlay multiple 189 instructions.) 190 191 - Kprobes analyzes the entire function and verifies that there is no 192 jump into the optimized region. Specifically: 193 - the function contains no indirect jump; 194 - the function contains no instruction that causes an exception (since 195 the fixup code triggered by the exception could jump back into the 196 optimized region -- Kprobes checks the exception tables to verify this); 197 and 198 - there is no near jump to the optimized region (other than to the first 199 byte). 200 201 - For each instruction in the optimized region, Kprobes verifies that 202 the instruction can be executed out of line. 203 204 1.4.3 Preparing Detour Buffer 205 206 Next, Kprobes prepares a "detour" buffer, which contains the following 207 instruction sequence: 208 - code to push the CPU's registers (emulating a breakpoint trap) 209 - a call to the trampoline code which calls user's probe handlers. 210 - code to restore registers 211 - the instructions from the optimized region 212 - a jump back to the original execution path. 213 214 1.4.4 Pre-optimization 215 216 After preparing the detour buffer, Kprobes verifies that none of the 217 following situations exist: 218 - The probe has either a break_handler (i.e., it's a jprobe) or a 219 post_handler. 220 - Other instructions in the optimized region are probed. 221 - The probe is disabled. 222 In any of the above cases, Kprobes won't start optimizing the probe. 223 Since these are temporary situations, Kprobes tries to start 224 optimizing it again if the situation is changed. 225 226 If the kprobe can be optimized, Kprobes enqueues the kprobe to an 227 optimizing list, and kicks the kprobe-optimizer workqueue to optimize 228 it. If the to-be-optimized probepoint is hit before being optimized, 229 Kprobes returns control to the original instruction path by setting 230 the CPU's instruction pointer to the copied code in the detour buffer 231 -- thus at least avoiding the single-step. 232 233 1.4.5 Optimization 234 235 The Kprobe-optimizer doesn't insert the jump instruction immediately; 236 rather, it calls synchronize_sched() for safety first, because it's 237 possible for a CPU to be interrupted in the middle of executing the 238 optimized region(*). As you know, synchronize_sched() can ensure 239 that all interruptions that were active when synchronize_sched() 240 was called are done, but only if CONFIG_PREEMPT=n. So, this version 241 of kprobe optimization supports only kernels with CONFIG_PREEMPT=n.(**) 242 243 After that, the Kprobe-optimizer calls stop_machine() to replace 244 the optimized region with a jump instruction to the detour buffer, 245 using text_poke_smp(). 246 247 1.4.6 Unoptimization 248 249 When an optimized kprobe is unregistered, disabled, or blocked by 250 another kprobe, it will be unoptimized. If this happens before 251 the optimization is complete, the kprobe is just dequeued from the 252 optimized list. If the optimization has been done, the jump is 253 replaced with the original code (except for an int3 breakpoint in 254 the first byte) by using text_poke_smp(). 255 256 (*)Please imagine that the 2nd instruction is interrupted and then 257 the optimizer replaces the 2nd instruction with the jump *address* 258 while the interrupt handler is running. When the interrupt 259 returns to original address, there is no valid instruction, 260 and it causes an unexpected result. 261 262 (**)This optimization-safety checking may be replaced with the 263 stop-machine method that ksplice uses for supporting a CONFIG_PREEMPT=y 264 kernel. 265 266 NOTE for geeks: 267 The jump optimization changes the kprobe's pre_handler behavior. 268 Without optimization, the pre_handler can change the kernel's execution 269 path by changing regs->ip and returning 1. However, when the probe 270 is optimized, that modification is ignored. Thus, if you want to 271 tweak the kernel's execution path, you need to suppress optimization, 272 using one of the following techniques: 273 - Specify an empty function for the kprobe's post_handler or break_handler. 274 or 275 - Execute 'sysctl -w debug.kprobes_optimization=n' 276 277 1.5 Blacklist 278 279 Kprobes can probe most of the kernel except itself. This means 280 that there are some functions where kprobes cannot probe. Probing 281 (trapping) such functions can cause a recursive trap (e.g. double 282 fault) or the nested probe handler may never be called. 283 Kprobes manages such functions as a blacklist. 284 If you want to add a function into the blacklist, you just need 285 to (1) include linux/kprobes.h and (2) use NOKPROBE_SYMBOL() macro 286 to specify a blacklisted function. 287 Kprobes checks the given probe address against the blacklist and 288 rejects registering it, if the given address is in the blacklist. 289 290 2. Architectures Supported 291 292 Kprobes, jprobes, and return probes are implemented on the following 293 architectures: 294 295 - i386 (Supports jump optimization) 296 - x86_64 (AMD-64, EM64T) (Supports jump optimization) 297 - ppc64 298 - ia64 (Does not support probes on instruction slot1.) 299 - sparc64 (Return probes not yet implemented.) 300 - arm 301 - ppc 302 - mips 303 304 3. Configuring Kprobes 305 306 When configuring the kernel using make menuconfig/xconfig/oldconfig, 307 ensure that CONFIG_KPROBES is set to "y". Under "Instrumentation 308 Support", look for "Kprobes". 309 310 So that you can load and unload Kprobes-based instrumentation modules, 311 make sure "Loadable module support" (CONFIG_MODULES) and "Module 312 unloading" (CONFIG_MODULE_UNLOAD) are set to "y". 313 314 Also make sure that CONFIG_KALLSYMS and perhaps even CONFIG_KALLSYMS_ALL 315 are set to "y", since kallsyms_lookup_name() is used by the in-kernel 316 kprobe address resolution code. 317 318 If you need to insert a probe in the middle of a function, you may find 319 it useful to "Compile the kernel with debug info" (CONFIG_DEBUG_INFO), 320 so you can use "objdump -d -l vmlinux" to see the source-to-object 321 code mapping. 322 323 4. API Reference 324 325 The Kprobes API includes a "register" function and an "unregister" 326 function for each type of probe. The API also includes "register_*probes" 327 and "unregister_*probes" functions for (un)registering arrays of probes. 328 Here are terse, mini-man-page specifications for these functions and 329 the associated probe handlers that you'll write. See the files in the 330 samples/kprobes/ sub-directory for examples. 331 332 4.1 register_kprobe 333 334 #include <linux/kprobes.h> 335 int register_kprobe(struct kprobe *kp); 336 337 Sets a breakpoint at the address kp->addr. When the breakpoint is 338 hit, Kprobes calls kp->pre_handler. After the probed instruction 339 is single-stepped, Kprobe calls kp->post_handler. If a fault 340 occurs during execution of kp->pre_handler or kp->post_handler, 341 or during single-stepping of the probed instruction, Kprobes calls 342 kp->fault_handler. Any or all handlers can be NULL. If kp->flags 343 is set KPROBE_FLAG_DISABLED, that kp will be registered but disabled, 344 so, its handlers aren't hit until calling enable_kprobe(kp). 345 346 NOTE: 347 1. With the introduction of the "symbol_name" field to struct kprobe, 348 the probepoint address resolution will now be taken care of by the kernel. 349 The following will now work: 350 351 kp.symbol_name = "symbol_name"; 352 353 (64-bit powerpc intricacies such as function descriptors are handled 354 transparently) 355 356 2. Use the "offset" field of struct kprobe if the offset into the symbol 357 to install a probepoint is known. This field is used to calculate the 358 probepoint. 359 360 3. Specify either the kprobe "symbol_name" OR the "addr". If both are 361 specified, kprobe registration will fail with -EINVAL. 362 363 4. With CISC architectures (such as i386 and x86_64), the kprobes code 364 does not validate if the kprobe.addr is at an instruction boundary. 365 Use "offset" with caution. 366 367 register_kprobe() returns 0 on success, or a negative errno otherwise. 368 369 User's pre-handler (kp->pre_handler): 370 #include <linux/kprobes.h> 371 #include <linux/ptrace.h> 372 int pre_handler(struct kprobe *p, struct pt_regs *regs); 373 374 Called with p pointing to the kprobe associated with the breakpoint, 375 and regs pointing to the struct containing the registers saved when 376 the breakpoint was hit. Return 0 here unless you're a Kprobes geek. 377 378 User's post-handler (kp->post_handler): 379 #include <linux/kprobes.h> 380 #include <linux/ptrace.h> 381 void post_handler(struct kprobe *p, struct pt_regs *regs, 382 unsigned long flags); 383 384 p and regs are as described for the pre_handler. flags always seems 385 to be zero. 386 387 User's fault-handler (kp->fault_handler): 388 #include <linux/kprobes.h> 389 #include <linux/ptrace.h> 390 int fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr); 391 392 p and regs are as described for the pre_handler. trapnr is the 393 architecture-specific trap number associated with the fault (e.g., 394 on i386, 13 for a general protection fault or 14 for a page fault). 395 Returns 1 if it successfully handled the exception. 396 397 4.2 register_jprobe 398 399 #include <linux/kprobes.h> 400 int register_jprobe(struct jprobe *jp) 401 402 Sets a breakpoint at the address jp->kp.addr, which must be the address 403 of the first instruction of a function. When the breakpoint is hit, 404 Kprobes runs the handler whose address is jp->entry. 405 406 The handler should have the same arg list and return type as the probed 407 function; and just before it returns, it must call jprobe_return(). 408 (The handler never actually returns, since jprobe_return() returns 409 control to Kprobes.) If the probed function is declared asmlinkage 410 or anything else that affects how args are passed, the handler's 411 declaration must match. 412 413 register_jprobe() returns 0 on success, or a negative errno otherwise. 414 415 4.3 register_kretprobe 416 417 #include <linux/kprobes.h> 418 int register_kretprobe(struct kretprobe *rp); 419 420 Establishes a return probe for the function whose address is 421 rp->kp.addr. When that function returns, Kprobes calls rp->handler. 422 You must set rp->maxactive appropriately before you call 423 register_kretprobe(); see "How Does a Return Probe Work?" for details. 424 425 register_kretprobe() returns 0 on success, or a negative errno 426 otherwise. 427 428 User's return-probe handler (rp->handler): 429 #include <linux/kprobes.h> 430 #include <linux/ptrace.h> 431 int kretprobe_handler(struct kretprobe_instance *ri, struct pt_regs *regs); 432 433 regs is as described for kprobe.pre_handler. ri points to the 434 kretprobe_instance object, of which the following fields may be 435 of interest: 436 - ret_addr: the return address 437 - rp: points to the corresponding kretprobe object 438 - task: points to the corresponding task struct 439 - data: points to per return-instance private data; see "Kretprobe 440 entry-handler" for details. 441 442 The regs_return_value(regs) macro provides a simple abstraction to 443 extract the return value from the appropriate register as defined by 444 the architecture's ABI. 445 446 The handler's return value is currently ignored. 447 448 4.4 unregister_*probe 449 450 #include <linux/kprobes.h> 451 void unregister_kprobe(struct kprobe *kp); 452 void unregister_jprobe(struct jprobe *jp); 453 void unregister_kretprobe(struct kretprobe *rp); 454 455 Removes the specified probe. The unregister function can be called 456 at any time after the probe has been registered. 457 458 NOTE: 459 If the functions find an incorrect probe (ex. an unregistered probe), 460 they clear the addr field of the probe. 461 462 4.5 register_*probes 463 464 #include <linux/kprobes.h> 465 int register_kprobes(struct kprobe **kps, int num); 466 int register_kretprobes(struct kretprobe **rps, int num); 467 int register_jprobes(struct jprobe **jps, int num); 468 469 Registers each of the num probes in the specified array. If any 470 error occurs during registration, all probes in the array, up to 471 the bad probe, are safely unregistered before the register_*probes 472 function returns. 473 - kps/rps/jps: an array of pointers to *probe data structures 474 - num: the number of the array entries. 475 476 NOTE: 477 You have to allocate(or define) an array of pointers and set all 478 of the array entries before using these functions. 479 480 4.6 unregister_*probes 481 482 #include <linux/kprobes.h> 483 void unregister_kprobes(struct kprobe **kps, int num); 484 void unregister_kretprobes(struct kretprobe **rps, int num); 485 void unregister_jprobes(struct jprobe **jps, int num); 486 487 Removes each of the num probes in the specified array at once. 488 489 NOTE: 490 If the functions find some incorrect probes (ex. unregistered 491 probes) in the specified array, they clear the addr field of those 492 incorrect probes. However, other probes in the array are 493 unregistered correctly. 494 495 4.7 disable_*probe 496 497 #include <linux/kprobes.h> 498 int disable_kprobe(struct kprobe *kp); 499 int disable_kretprobe(struct kretprobe *rp); 500 int disable_jprobe(struct jprobe *jp); 501 502 Temporarily disables the specified *probe. You can enable it again by using 503 enable_*probe(). You must specify the probe which has been registered. 504 505 4.8 enable_*probe 506 507 #include <linux/kprobes.h> 508 int enable_kprobe(struct kprobe *kp); 509 int enable_kretprobe(struct kretprobe *rp); 510 int enable_jprobe(struct jprobe *jp); 511 512 Enables *probe which has been disabled by disable_*probe(). You must specify 513 the probe which has been registered. 514 515 5. Kprobes Features and Limitations 516 517 Kprobes allows multiple probes at the same address. Currently, 518 however, there cannot be multiple jprobes on the same function at 519 the same time. Also, a probepoint for which there is a jprobe or 520 a post_handler cannot be optimized. So if you install a jprobe, 521 or a kprobe with a post_handler, at an optimized probepoint, the 522 probepoint will be unoptimized automatically. 523 524 In general, you can install a probe anywhere in the kernel. 525 In particular, you can probe interrupt handlers. Known exceptions 526 are discussed in this section. 527 528 The register_*probe functions will return -EINVAL if you attempt 529 to install a probe in the code that implements Kprobes (mostly 530 kernel/kprobes.c and arch/*/kernel/kprobes.c, but also functions such 531 as do_page_fault and notifier_call_chain). 532 533 If you install a probe in an inline-able function, Kprobes makes 534 no attempt to chase down all inline instances of the function and 535 install probes there. gcc may inline a function without being asked, 536 so keep this in mind if you're not seeing the probe hits you expect. 537 538 A probe handler can modify the environment of the probed function 539 -- e.g., by modifying kernel data structures, or by modifying the 540 contents of the pt_regs struct (which are restored to the registers 541 upon return from the breakpoint). So Kprobes can be used, for example, 542 to install a bug fix or to inject faults for testing. Kprobes, of 543 course, has no way to distinguish the deliberately injected faults 544 from the accidental ones. Don't drink and probe. 545 546 Kprobes makes no attempt to prevent probe handlers from stepping on 547 each other -- e.g., probing printk() and then calling printk() from a 548 probe handler. If a probe handler hits a probe, that second probe's 549 handlers won't be run in that instance, and the kprobe.nmissed member 550 of the second probe will be incremented. 551 552 As of Linux v2.6.15-rc1, multiple handlers (or multiple instances of 553 the same handler) may run concurrently on different CPUs. 554 555 Kprobes does not use mutexes or allocate memory except during 556 registration and unregistration. 557 558 Probe handlers are run with preemption disabled. Depending on the 559 architecture and optimization state, handlers may also run with 560 interrupts disabled (e.g., kretprobe handlers and optimized kprobe 561 handlers run without interrupt disabled on x86/x86-64). In any case, 562 your handler should not yield the CPU (e.g., by attempting to acquire 563 a semaphore). 564 565 Since a return probe is implemented by replacing the return 566 address with the trampoline's address, stack backtraces and calls 567 to __builtin_return_address() will typically yield the trampoline's 568 address instead of the real return address for kretprobed functions. 569 (As far as we can tell, __builtin_return_address() is used only 570 for instrumentation and error reporting.) 571 572 If the number of times a function is called does not match the number 573 of times it returns, registering a return probe on that function may 574 produce undesirable results. In such a case, a line: 575 kretprobe BUG!: Processing kretprobe d000000000041aa8 @ c00000000004f48c 576 gets printed. With this information, one will be able to correlate the 577 exact instance of the kretprobe that caused the problem. We have the 578 do_exit() case covered. do_execve() and do_fork() are not an issue. 579 We're unaware of other specific cases where this could be a problem. 580 581 If, upon entry to or exit from a function, the CPU is running on 582 a stack other than that of the current task, registering a return 583 probe on that function may produce undesirable results. For this 584 reason, Kprobes doesn't support return probes (or kprobes or jprobes) 585 on the x86_64 version of __switch_to(); the registration functions 586 return -EINVAL. 587 588 On x86/x86-64, since the Jump Optimization of Kprobes modifies 589 instructions widely, there are some limitations to optimization. To 590 explain it, we introduce some terminology. Imagine a 3-instruction 591 sequence consisting of a two 2-byte instructions and one 3-byte 592 instruction. 593 594 IA 595 | 596 [-2][-1] 597 [ins1][ins2][ ins3 ] 598 [<- DCR ->] 599 [<- JTPR ->] 600 601 ins1: 1st Instruction 602 ins2: 2nd Instruction 603 ins3: 3rd Instruction 604 IA: Insertion Address 605 JTPR: Jump Target Prohibition Region 606 DCR: Detoured Code Region 607 608 The instructions in DCR are copied to the out-of-line buffer 609 of the kprobe, because the bytes in DCR are replaced by 610 a 5-byte jump instruction. So there are several limitations. 611 612 a) The instructions in DCR must be relocatable. 613 b) The instructions in DCR must not include a call instruction. 614 c) JTPR must not be targeted by any jump or call instruction. 615 d) DCR must not straddle the border between functions. 616 617 Anyway, these limitations are checked by the in-kernel instruction 618 decoder, so you don't need to worry about that. 619 620 6. Probe Overhead 621 622 On a typical CPU in use in 2005, a kprobe hit takes 0.5 to 1.0 623 microseconds to process. Specifically, a benchmark that hits the same 624 probepoint repeatedly, firing a simple handler each time, reports 1-2 625 million hits per second, depending on the architecture. A jprobe or 626 return-probe hit typically takes 50-75% longer than a kprobe hit. 627 When you have a return probe set on a function, adding a kprobe at 628 the entry to that function adds essentially no overhead. 629 630 Here are sample overhead figures (in usec) for different architectures. 631 k = kprobe; j = jprobe; r = return probe; kr = kprobe + return probe 632 on same function; jr = jprobe + return probe on same function 633 634 i386: Intel Pentium M, 1495 MHz, 2957.31 bogomips 635 k = 0.57 usec; j = 1.00; r = 0.92; kr = 0.99; jr = 1.40 636 637 x86_64: AMD Opteron 246, 1994 MHz, 3971.48 bogomips 638 k = 0.49 usec; j = 0.76; r = 0.80; kr = 0.82; jr = 1.07 639 640 ppc64: POWER5 (gr), 1656 MHz (SMT disabled, 1 virtual CPU per physical CPU) 641 k = 0.77 usec; j = 1.31; r = 1.26; kr = 1.45; jr = 1.99 642 643 6.1 Optimized Probe Overhead 644 645 Typically, an optimized kprobe hit takes 0.07 to 0.1 microseconds to 646 process. Here are sample overhead figures (in usec) for x86 architectures. 647 k = unoptimized kprobe, b = boosted (single-step skipped), o = optimized kprobe, 648 r = unoptimized kretprobe, rb = boosted kretprobe, ro = optimized kretprobe. 649 650 i386: Intel(R) Xeon(R) E5410, 2.33GHz, 4656.90 bogomips 651 k = 0.80 usec; b = 0.33; o = 0.05; r = 1.10; rb = 0.61; ro = 0.33 652 653 x86-64: Intel(R) Xeon(R) E5410, 2.33GHz, 4656.90 bogomips 654 k = 0.99 usec; b = 0.43; o = 0.06; r = 1.24; rb = 0.68; ro = 0.30 655 656 7. TODO 657 658 a. SystemTap (http://sourceware.org/systemtap): Provides a simplified 659 programming interface for probe-based instrumentation. Try it out. 660 b. Kernel return probes for sparc64. 661 c. Support for other architectures. 662 d. User-space probes. 663 e. Watchpoint probes (which fire on data references). 664 665 8. Kprobes Example 666 667 See samples/kprobes/kprobe_example.c 668 669 9. Jprobes Example 670 671 See samples/kprobes/jprobe_example.c 672 673 10. Kretprobes Example 674 675 See samples/kprobes/kretprobe_example.c 676 677 For additional information on Kprobes, refer to the following URLs: 678 http://www-106.ibm.com/developerworks/library/l-kprobes.html?ca=dgr-lnxw42Kprobe 679 http://www.redhat.com/magazine/005mar05/features/kprobes/ 680 http://www-users.cs.umn.edu/~boutcher/kprobes/ 681 http://www.linuxsymposium.org/2006/linuxsymposium_procv2.pdf (pages 101-115) 682 683 684 Appendix A: The kprobes debugfs interface 685 686 With recent kernels (> 2.6.20) the list of registered kprobes is visible 687 under the /sys/kernel/debug/kprobes/ directory (assuming debugfs is mounted at //sys/kernel/debug). 688 689 /sys/kernel/debug/kprobes/list: Lists all registered probes on the system 690 691 c015d71a k vfs_read+0x0 692 c011a316 j do_fork+0x0 693 c03dedc5 r tcp_v4_rcv+0x0 694 695 The first column provides the kernel address where the probe is inserted. 696 The second column identifies the type of probe (k - kprobe, r - kretprobe 697 and j - jprobe), while the third column specifies the symbol+offset of 698 the probe. If the probed function belongs to a module, the module name 699 is also specified. Following columns show probe status. If the probe is on 700 a virtual address that is no longer valid (module init sections, module 701 virtual addresses that correspond to modules that've been unloaded), 702 such probes are marked with [GONE]. If the probe is temporarily disabled, 703 such probes are marked with [DISABLED]. If the probe is optimized, it is 704 marked with [OPTIMIZED]. 705 706 /sys/kernel/debug/kprobes/enabled: Turn kprobes ON/OFF forcibly. 707 708 Provides a knob to globally and forcibly turn registered kprobes ON or OFF. 709 By default, all kprobes are enabled. By echoing "0" to this file, all 710 registered probes will be disarmed, till such time a "1" is echoed to this 711 file. Note that this knob just disarms and arms all kprobes and doesn't 712 change each probe's disabling state. This means that disabled kprobes (marked 713 [DISABLED]) will be not enabled if you turn ON all kprobes by this knob. 714 715 716 Appendix B: The kprobes sysctl interface 717 718 /proc/sys/debug/kprobes-optimization: Turn kprobes optimization ON/OFF. 719 720 When CONFIG_OPTPROBES=y, this sysctl interface appears and it provides 721 a knob to globally and forcibly turn jump optimization (see section 722 1.4) ON or OFF. By default, jump optimization is allowed (ON). 723 If you echo "0" to this file or set "debug.kprobes_optimization" to 724 0 via sysctl, all optimized probes will be unoptimized, and any new 725 probes registered after that will not be optimized. Note that this 726 knob *changes* the optimized state. This means that optimized probes 727 (marked [OPTIMIZED]) will be unoptimized ([OPTIMIZED] tag will be 728 removed). If the knob is turned on, they will be optimized again.