1Title : Kernel Probes (Kprobes) 2Authors : Jim Keniston <jkenisto@us.ibm.com> 3 : Prasanna S Panchamukhi <prasanna@in.ibm.com> 4 5CONTENTS 6 71. Concepts: Kprobes, Jprobes, Return Probes 82. Architectures Supported 93. Configuring Kprobes 104. API Reference 115. Kprobes Features and Limitations 126. Probe Overhead 137. TODO 148. Kprobes Example 159. Jprobes Example 1610. Kretprobes Example 17Appendix A: The kprobes debugfs interface 18 191. Concepts: Kprobes, Jprobes, Return Probes 20 21Kprobes enables you to dynamically break into any kernel routine and 22collect debugging and performance information non-disruptively. You 23can trap at almost any kernel code address, specifying a handler 24routine to be invoked when the breakpoint is hit. 25 26There are currently three types of probes: kprobes, jprobes, and 27kretprobes (also called return probes). A kprobe can be inserted 28on virtually any instruction in the kernel. A jprobe is inserted at 29the entry to a kernel function, and provides convenient access to the 30function's arguments. A return probe fires when a specified function 31returns. 32 33In the typical case, Kprobes-based instrumentation is packaged as 34a kernel module. The module's init function installs ("registers") 35one or more probes, and the exit function unregisters them. A 36registration function such as register_kprobe() specifies where 37the probe is to be inserted and what handler is to be called when 38the probe is hit. 39 40There are also register_/unregister_*probes() functions for batch 41registration/unregistration of a group of *probes. These functions 42can speed up unregistration process when you have to unregister 43a lot of probes at once. 44 45The next three subsections explain how the different types of 46probes work. They explain certain things that you'll need to 47know in order to make the best use of Kprobes -- e.g., the 48difference between a pre_handler and a post_handler, and how 49to use the maxactive and nmissed fields of a kretprobe. But 50if you're in a hurry to start using Kprobes, you can skip ahead 51to section 2. 52 531.1 How Does a Kprobe Work? 54 55When a kprobe is registered, Kprobes makes a copy of the probed 56instruction and replaces the first byte(s) of the probed instruction 57with a breakpoint instruction (e.g., int3 on i386 and x86_64). 58 59When a CPU hits the breakpoint instruction, a trap occurs, the CPU's 60registers are saved, and control passes to Kprobes via the 61notifier_call_chain mechanism. Kprobes executes the "pre_handler" 62associated with the kprobe, passing the handler the addresses of the 63kprobe struct and the saved registers. 64 65Next, Kprobes single-steps its copy of the probed instruction. 66(It would be simpler to single-step the actual instruction in place, 67but then Kprobes would have to temporarily remove the breakpoint 68instruction. This would open a small time window when another CPU 69could sail right past the probepoint.) 70 71After the instruction is single-stepped, Kprobes executes the 72"post_handler," if any, that is associated with the kprobe. 73Execution then continues with the instruction following the probepoint. 74 751.2 How Does a Jprobe Work? 76 77A jprobe is implemented using a kprobe that is placed on a function's 78entry point. It employs a simple mirroring principle to allow 79seamless access to the probed function's arguments. The jprobe 80handler routine should have the same signature (arg list and return 81type) as the function being probed, and must always end by calling 82the Kprobes function jprobe_return(). 83 84Here's how it works. When the probe is hit, Kprobes makes a copy of 85the saved registers and a generous portion of the stack (see below). 86Kprobes then points the saved instruction pointer at the jprobe's 87handler routine, and returns from the trap. As a result, control 88passes to the handler, which is presented with the same register and 89stack contents as the probed function. When it is done, the handler 90calls jprobe_return(), which traps again to restore the original stack 91contents and processor state and switch to the probed function. 92 93By convention, the callee owns its arguments, so gcc may produce code 94that unexpectedly modifies that portion of the stack. This is why 95Kprobes saves a copy of the stack and restores it after the jprobe 96handler has run. Up to MAX_STACK_SIZE bytes are copied -- e.g., 9764 bytes on i386. 98 99Note that the probed function's args may be passed on the stack 100or in registers. The jprobe will work in either case, so long as the 101handler's prototype matches that of the probed function. 102 1031.3 Return Probes 104 1051.3.1 How Does a Return Probe Work? 106 107When you call register_kretprobe(), Kprobes establishes a kprobe at 108the entry to the function. When the probed function is called and this 109probe is hit, Kprobes saves a copy of the return address, and replaces 110the return address with the address of a "trampoline." The trampoline 111is an arbitrary piece of code -- typically just a nop instruction. 112At boot time, Kprobes registers a kprobe at the trampoline. 113 114When the probed function executes its return instruction, control 115passes to the trampoline and that probe is hit. Kprobes' trampoline 116handler calls the user-specified return handler associated with the 117kretprobe, then sets the saved instruction pointer to the saved return 118address, and that's where execution resumes upon return from the trap. 119 120While the probed function is executing, its return address is 121stored in an object of type kretprobe_instance. Before calling 122register_kretprobe(), the user sets the maxactive field of the 123kretprobe struct to specify how many instances of the specified 124function can be probed simultaneously. register_kretprobe() 125pre-allocates the indicated number of kretprobe_instance objects. 126 127For example, if the function is non-recursive and is called with a 128spinlock held, maxactive = 1 should be enough. If the function is 129non-recursive and can never relinquish the CPU (e.g., via a semaphore 130or preemption), NR_CPUS should be enough. If maxactive <= 0, it is 131set to a default value. If CONFIG_PREEMPT is enabled, the default 132is max(10, 2*NR_CPUS). Otherwise, the default is NR_CPUS. 133 134It's not a disaster if you set maxactive too low; you'll just miss 135some probes. In the kretprobe struct, the nmissed field is set to 136zero when the return probe is registered, and is incremented every 137time the probed function is entered but there is no kretprobe_instance 138object available for establishing the return probe. 139 1401.3.2 Kretprobe entry-handler 141 142Kretprobes also provides an optional user-specified handler which runs 143on function entry. This handler is specified by setting the entry_handler 144field of the kretprobe struct. Whenever the kprobe placed by kretprobe at the 145function entry is hit, the user-defined entry_handler, if any, is invoked. 146If the entry_handler returns 0 (success) then a corresponding return handler 147is guaranteed to be called upon function return. If the entry_handler 148returns a non-zero error then Kprobes leaves the return address as is, and 149the kretprobe has no further effect for that particular function instance. 150 151Multiple entry and return handler invocations are matched using the unique 152kretprobe_instance object associated with them. Additionally, a user 153may also specify per return-instance private data to be part of each 154kretprobe_instance object. This is especially useful when sharing private 155data between corresponding user entry and return handlers. The size of each 156private data object can be specified at kretprobe registration time by 157setting the data_size field of the kretprobe struct. This data can be 158accessed through the data field of each kretprobe_instance object. 159 160In case probed function is entered but there is no kretprobe_instance 161object available, then in addition to incrementing the nmissed count, 162the user entry_handler invocation is also skipped. 163 1642. Architectures Supported 165 166Kprobes, jprobes, and return probes are implemented on the following 167architectures: 168 169- i386 170- x86_64 (AMD-64, EM64T) 171- ppc64 172- ia64 (Does not support probes on instruction slot1.) 173- sparc64 (Return probes not yet implemented.) 174- arm 175- ppc 176 1773. Configuring Kprobes 178 179When configuring the kernel using make menuconfig/xconfig/oldconfig, 180ensure that CONFIG_KPROBES is set to "y". Under "Instrumentation 181Support", look for "Kprobes". 182 183So that you can load and unload Kprobes-based instrumentation modules, 184make sure "Loadable module support" (CONFIG_MODULES) and "Module 185unloading" (CONFIG_MODULE_UNLOAD) are set to "y". 186 187Also make sure that CONFIG_KALLSYMS and perhaps even CONFIG_KALLSYMS_ALL 188are set to "y", since kallsyms_lookup_name() is used by the in-kernel 189kprobe address resolution code. 190 191If you need to insert a probe in the middle of a function, you may find 192it useful to "Compile the kernel with debug info" (CONFIG_DEBUG_INFO), 193so you can use "objdump -d -l vmlinux" to see the source-to-object 194code mapping. 195 1964. API Reference 197 198The Kprobes API includes a "register" function and an "unregister" 199function for each type of probe. The API also includes "register_*probes" 200and "unregister_*probes" functions for (un)registering arrays of probes. 201Here are terse, mini-man-page specifications for these functions and 202the associated probe handlers that you'll write. See the files in the 203samples/kprobes/ sub-directory for examples. 204 2054.1 register_kprobe 206 207#include <linux/kprobes.h> 208int register_kprobe(struct kprobe *kp); 209 210Sets a breakpoint at the address kp->addr. When the breakpoint is 211hit, Kprobes calls kp->pre_handler. After the probed instruction 212is single-stepped, Kprobe calls kp->post_handler. If a fault 213occurs during execution of kp->pre_handler or kp->post_handler, 214or during single-stepping of the probed instruction, Kprobes calls 215kp->fault_handler. Any or all handlers can be NULL. 216 217NOTE: 2181. With the introduction of the "symbol_name" field to struct kprobe, 219the probepoint address resolution will now be taken care of by the kernel. 220The following will now work: 221 222 kp.symbol_name = "symbol_name"; 223 224(64-bit powerpc intricacies such as function descriptors are handled 225transparently) 226 2272. Use the "offset" field of struct kprobe if the offset into the symbol 228to install a probepoint is known. This field is used to calculate the 229probepoint. 230 2313. Specify either the kprobe "symbol_name" OR the "addr". If both are 232specified, kprobe registration will fail with -EINVAL. 233 2344. With CISC architectures (such as i386 and x86_64), the kprobes code 235does not validate if the kprobe.addr is at an instruction boundary. 236Use "offset" with caution. 237 238register_kprobe() returns 0 on success, or a negative errno otherwise. 239 240User's pre-handler (kp->pre_handler): 241#include <linux/kprobes.h> 242#include <linux/ptrace.h> 243int pre_handler(struct kprobe *p, struct pt_regs *regs); 244 245Called with p pointing to the kprobe associated with the breakpoint, 246and regs pointing to the struct containing the registers saved when 247the breakpoint was hit. Return 0 here unless you're a Kprobes geek. 248 249User's post-handler (kp->post_handler): 250#include <linux/kprobes.h> 251#include <linux/ptrace.h> 252void post_handler(struct kprobe *p, struct pt_regs *regs, 253 unsigned long flags); 254 255p and regs are as described for the pre_handler. flags always seems 256to be zero. 257 258User's fault-handler (kp->fault_handler): 259#include <linux/kprobes.h> 260#include <linux/ptrace.h> 261int fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr); 262 263p and regs are as described for the pre_handler. trapnr is the 264architecture-specific trap number associated with the fault (e.g., 265on i386, 13 for a general protection fault or 14 for a page fault). 266Returns 1 if it successfully handled the exception. 267 2684.2 register_jprobe 269 270#include <linux/kprobes.h> 271int register_jprobe(struct jprobe *jp) 272 273Sets a breakpoint at the address jp->kp.addr, which must be the address 274of the first instruction of a function. When the breakpoint is hit, 275Kprobes runs the handler whose address is jp->entry. 276 277The handler should have the same arg list and return type as the probed 278function; and just before it returns, it must call jprobe_return(). 279(The handler never actually returns, since jprobe_return() returns 280control to Kprobes.) If the probed function is declared asmlinkage 281or anything else that affects how args are passed, the handler's 282declaration must match. 283 284register_jprobe() returns 0 on success, or a negative errno otherwise. 285 2864.3 register_kretprobe 287 288#include <linux/kprobes.h> 289int register_kretprobe(struct kretprobe *rp); 290 291Establishes a return probe for the function whose address is 292rp->kp.addr. When that function returns, Kprobes calls rp->handler. 293You must set rp->maxactive appropriately before you call 294register_kretprobe(); see "How Does a Return Probe Work?" for details. 295 296register_kretprobe() returns 0 on success, or a negative errno 297otherwise. 298 299User's return-probe handler (rp->handler): 300#include <linux/kprobes.h> 301#include <linux/ptrace.h> 302int kretprobe_handler(struct kretprobe_instance *ri, struct pt_regs *regs); 303 304regs is as described for kprobe.pre_handler. ri points to the 305kretprobe_instance object, of which the following fields may be 306of interest: 307- ret_addr: the return address 308- rp: points to the corresponding kretprobe object 309- task: points to the corresponding task struct 310- data: points to per return-instance private data; see "Kretprobe 311 entry-handler" for details. 312 313The regs_return_value(regs) macro provides a simple abstraction to 314extract the return value from the appropriate register as defined by 315the architecture's ABI. 316 317The handler's return value is currently ignored. 318 3194.4 unregister_*probe 320 321#include <linux/kprobes.h> 322void unregister_kprobe(struct kprobe *kp); 323void unregister_jprobe(struct jprobe *jp); 324void unregister_kretprobe(struct kretprobe *rp); 325 326Removes the specified probe. The unregister function can be called 327at any time after the probe has been registered. 328 329NOTE: 330If the functions find an incorrect probe (ex. an unregistered probe), 331they clear the addr field of the probe. 332 3334.5 register_*probes 334 335#include <linux/kprobes.h> 336int register_kprobes(struct kprobe **kps, int num); 337int register_kretprobes(struct kretprobe **rps, int num); 338int register_jprobes(struct jprobe **jps, int num); 339 340Registers each of the num probes in the specified array. If any 341error occurs during registration, all probes in the array, up to 342the bad probe, are safely unregistered before the register_*probes 343function returns. 344- kps/rps/jps: an array of pointers to *probe data structures 345- num: the number of the array entries. 346 347NOTE: 348You have to allocate(or define) an array of pointers and set all 349of the array entries before using these functions. 350 3514.6 unregister_*probes 352 353#include <linux/kprobes.h> 354void unregister_kprobes(struct kprobe **kps, int num); 355void unregister_kretprobes(struct kretprobe **rps, int num); 356void unregister_jprobes(struct jprobe **jps, int num); 357 358Removes each of the num probes in the specified array at once. 359 360NOTE: 361If the functions find some incorrect probes (ex. unregistered 362probes) in the specified array, they clear the addr field of those 363incorrect probes. However, other probes in the array are 364unregistered correctly. 365 3665. Kprobes Features and Limitations 367 368Kprobes allows multiple probes at the same address. Currently, 369however, there cannot be multiple jprobes on the same function at 370the same time. 371 372In general, you can install a probe anywhere in the kernel. 373In particular, you can probe interrupt handlers. Known exceptions 374are discussed in this section. 375 376The register_*probe functions will return -EINVAL if you attempt 377to install a probe in the code that implements Kprobes (mostly 378kernel/kprobes.c and arch/*/kernel/kprobes.c, but also functions such 379as do_page_fault and notifier_call_chain). 380 381If you install a probe in an inline-able function, Kprobes makes 382no attempt to chase down all inline instances of the function and 383install probes there. gcc may inline a function without being asked, 384so keep this in mind if you're not seeing the probe hits you expect. 385 386A probe handler can modify the environment of the probed function 387-- e.g., by modifying kernel data structures, or by modifying the 388contents of the pt_regs struct (which are restored to the registers 389upon return from the breakpoint). So Kprobes can be used, for example, 390to install a bug fix or to inject faults for testing. Kprobes, of 391course, has no way to distinguish the deliberately injected faults 392from the accidental ones. Don't drink and probe. 393 394Kprobes makes no attempt to prevent probe handlers from stepping on 395each other -- e.g., probing printk() and then calling printk() from a 396probe handler. If a probe handler hits a probe, that second probe's 397handlers won't be run in that instance, and the kprobe.nmissed member 398of the second probe will be incremented. 399 400As of Linux v2.6.15-rc1, multiple handlers (or multiple instances of 401the same handler) may run concurrently on different CPUs. 402 403Kprobes does not use mutexes or allocate memory except during 404registration and unregistration. 405 406Probe handlers are run with preemption disabled. Depending on the 407architecture, handlers may also run with interrupts disabled. In any 408case, your handler should not yield the CPU (e.g., by attempting to 409acquire a semaphore). 410 411Since a return probe is implemented by replacing the return 412address with the trampoline's address, stack backtraces and calls 413to __builtin_return_address() will typically yield the trampoline's 414address instead of the real return address for kretprobed functions. 415(As far as we can tell, __builtin_return_address() is used only 416for instrumentation and error reporting.) 417 418If the number of times a function is called does not match the number 419of times it returns, registering a return probe on that function may 420produce undesirable results. In such a case, a line: 421kretprobe BUG!: Processing kretprobe d000000000041aa8 @ c00000000004f48c 422gets printed. With this information, one will be able to correlate the 423exact instance of the kretprobe that caused the problem. We have the 424do_exit() case covered. do_execve() and do_fork() are not an issue. 425We're unaware of other specific cases where this could be a problem. 426 427If, upon entry to or exit from a function, the CPU is running on 428a stack other than that of the current task, registering a return 429probe on that function may produce undesirable results. For this 430reason, Kprobes doesn't support return probes (or kprobes or jprobes) 431on the x86_64 version of __switch_to(); the registration functions 432return -EINVAL. 433 4346. Probe Overhead 435 436On a typical CPU in use in 2005, a kprobe hit takes 0.5 to 1.0 437microseconds to process. Specifically, a benchmark that hits the same 438probepoint repeatedly, firing a simple handler each time, reports 1-2 439million hits per second, depending on the architecture. A jprobe or 440return-probe hit typically takes 50-75% longer than a kprobe hit. 441When you have a return probe set on a function, adding a kprobe at 442the entry to that function adds essentially no overhead. 443 444Here are sample overhead figures (in usec) for different architectures. 445k = kprobe; j = jprobe; r = return probe; kr = kprobe + return probe 446on same function; jr = jprobe + return probe on same function 447 448i386: Intel Pentium M, 1495 MHz, 2957.31 bogomips 449k = 0.57 usec; j = 1.00; r = 0.92; kr = 0.99; jr = 1.40 450 451x86_64: AMD Opteron 246, 1994 MHz, 3971.48 bogomips 452k = 0.49 usec; j = 0.76; r = 0.80; kr = 0.82; jr = 1.07 453 454ppc64: POWER5 (gr), 1656 MHz (SMT disabled, 1 virtual CPU per physical CPU) 455k = 0.77 usec; j = 1.31; r = 1.26; kr = 1.45; jr = 1.99 456 4577. TODO 458 459a. SystemTap (http://sourceware.org/systemtap): Provides a simplified 460programming interface for probe-based instrumentation. Try it out. 461b. Kernel return probes for sparc64. 462c. Support for other architectures. 463d. User-space probes. 464e. Watchpoint probes (which fire on data references). 465 4668. Kprobes Example 467 468See samples/kprobes/kprobe_example.c 469 4709. Jprobes Example 471 472See samples/kprobes/jprobe_example.c 473 47410. Kretprobes Example 475 476See samples/kprobes/kretprobe_example.c 477 478For additional information on Kprobes, refer to the following URLs: 479http://www-106.ibm.com/developerworks/library/l-kprobes.html?ca=dgr-lnxw42Kprobe 480http://www.redhat.com/magazine/005mar05/features/kprobes/ 481http://www-users.cs.umn.edu/~boutcher/kprobes/ 482http://www.linuxsymposium.org/2006/linuxsymposium_procv2.pdf (pages 101-115) 483 484 485Appendix A: The kprobes debugfs interface 486 487With recent kernels (> 2.6.20) the list of registered kprobes is visible 488under the /debug/kprobes/ directory (assuming debugfs is mounted at /debug). 489 490/debug/kprobes/list: Lists all registered probes on the system 491 492c015d71a k vfs_read+0x0 493c011a316 j do_fork+0x0 494c03dedc5 r tcp_v4_rcv+0x0 495 496The first column provides the kernel address where the probe is inserted. 497The second column identifies the type of probe (k - kprobe, r - kretprobe 498and j - jprobe), while the third column specifies the symbol+offset of 499the probe. If the probed function belongs to a module, the module name 500is also specified. Following columns show probe status. If the probe is on 501a virtual address that is no longer valid (module init sections, module 502virtual addresses that correspond to modules that've been unloaded), 503such probes are marked with [GONE]. 504 505/debug/kprobes/enabled: Turn kprobes ON/OFF 506 507Provides a knob to globally turn registered kprobes ON or OFF. By default, 508all kprobes are enabled. By echoing "0" to this file, all registered probes 509will be disarmed, till such time a "1" is echoed to this file. 510