1 ==========================
2 Kprobe-based Event Tracing
3 ==========================
5 :Author: Masami Hiramatsu
9 These events are similar to tracepoint-based events. Instead of tracepoints,
10 this is based on kprobes (kprobe and kretprobe). So it can probe wherever
11 kprobes can probe (this means, all functions except those with
12 __kprobes/nokprobe_inline annotation and those marked NOKPROBE_SYMBOL).
13 Unlike the tracepoint-based event, this can be added and removed
14 dynamically, on the fly.
16 To enable this feature, build your kernel with CONFIG_KPROBE_EVENTS=y.
18 Similar to the event tracer, this doesn't need to be activated via
19 current_tracer. Instead of that, add probe points via
20 /sys/kernel/tracing/kprobe_events, and enable it via
21 /sys/kernel/tracing/events/kprobes/<EVENT>/enable.
23 You can also use /sys/kernel/tracing/dynamic_events instead of
24 kprobe_events. That interface will provide unified access to other
27 Synopsis of kprobe_events
28 -------------------------
31 p[:[GRP/][EVENT]] [MOD:]SYM[+offs]|MEMADDR [FETCHARGS] : Set a probe
32 r[MAXACTIVE][:[GRP/][EVENT]] [MOD:]SYM[+0] [FETCHARGS] : Set a return probe
33 p[:[GRP/][EVENT]] [MOD:]SYM[+0]%return [FETCHARGS] : Set a return probe
34 -:[GRP/][EVENT] : Clear a probe
36 GRP : Group name. If omitted, use "kprobes" for it.
37 EVENT : Event name. If omitted, the event name is generated
38 based on SYM+offs or MEMADDR.
39 MOD : Module name which has given SYM.
40 SYM[+offs] : Symbol+offset where the probe is inserted.
41 SYM%return : Return address of the symbol
42 MEMADDR : Address where the probe is inserted.
43 MAXACTIVE : Maximum number of instances of the specified function that
44 can be probed simultaneously, or 0 for the default value
45 as defined in Documentation/trace/kprobes.rst section 1.3.1.
47 FETCHARGS : Arguments. Each probe can have up to 128 args.
48 %REG : Fetch register REG
49 @ADDR : Fetch memory at ADDR (ADDR should be in kernel)
50 @SYM[+|-offs] : Fetch memory at SYM +|- offs (SYM should be a data symbol)
51 $stackN : Fetch Nth entry of stack (N >= 0)
52 $stack : Fetch stack address.
53 $argN : Fetch the Nth function argument. (N >= 1) (\*1)
54 $retval : Fetch return value.(\*2)
55 $comm : Fetch current task comm.
56 +|-[u]OFFS(FETCHARG) : Fetch memory at FETCHARG +|- OFFS address.(\*3)(\*4)
57 \IMM : Store an immediate value to the argument.
58 NAME=FETCHARG : Set NAME as the argument name of FETCHARG.
59 FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types
60 (u8/u16/u32/u64/s8/s16/s32/s64), hexadecimal types
61 (x8/x16/x32/x64), "char", "string", "ustring", "symbol", "symstr"
62 and bitfield are supported.
64 (\*1) only for the probe on function entry (offs == 0).
65 (\*2) only for return probe.
66 (\*3) this is useful for fetching a field of data structures.
67 (\*4) "u" means user-space dereference. See :ref:`user_mem_access`.
71 Several types are supported for fetchargs. Kprobe tracer will access memory
72 by given type. Prefix 's' and 'u' means those types are signed and unsigned
73 respectively. 'x' prefix implies it is unsigned. Traced arguments are shown
74 in decimal ('s' and 'u') or hexadecimal ('x'). Without type casting, 'x32'
75 or 'x64' is used depends on the architecture (e.g. x86-32 uses x32, and
78 These value types can be an array. To record array data, you can add '[N]'
79 (where N is a fixed number, less than 64) to the base type.
80 E.g. 'x16[4]' means an array of x16 (2-byte hex) with 4 elements.
81 Note that the array can be applied to memory type fetchargs, you can not
82 apply it to registers/stack-entries etc. (for example, '$stack1:x8[8]' is
83 wrong, but '+8($stack):x8[8]' is OK.)
85 Char type can be used to show the character value of traced arguments.
87 String type is a special type, which fetches a "null-terminated" string from
88 kernel space. This means it will fail and store NULL if the string container
89 has been paged out. "ustring" type is an alternative of string for user-space.
90 See :ref:`user_mem_access` for more info.
92 The string array type is a bit different from other types. For other base
93 types, <base-type>[1] is equal to <base-type> (e.g. +0(%di):x32[1] is same
94 as +0(%di):x32.) But string[1] is not equal to string. The string type itself
95 represents "char array", but string array type represents "char * array".
96 So, for example, +0(%di):string[1] is equal to +0(+0(%di)):string.
97 Bitfield is another special type, which takes 3 parameters, bit-width, bit-
98 offset, and container-size (usually 32). The syntax is::
100 b<bit-width>@<bit-offset>/<container-size>
102 Symbol type('symbol') is an alias of u32 or u64 type (depends on BITS_PER_LONG)
103 which shows given pointer in "symbol+offset" style.
104 On the other hand, symbol-string type ('symstr') converts the given address to
105 "symbol+offset/symbolsize" style and stores it as a null-terminated string.
106 With 'symstr' type, you can filter the event with wildcard pattern of the
107 symbols, and you don't need to solve symbol name by yourself.
108 For $comm, the default type is "string"; any other type is invalid.
114 Kprobe events supports user-space memory access. For that purpose, you can use
115 either user-space dereference syntax or 'ustring' type.
117 The user-space dereference syntax allows you to access a field of a data
118 structure in user-space. This is done by adding the "u" prefix to the
119 dereference syntax. For example, +u4(%si) means it will read memory from the
120 address in the register %si offset by 4, and the memory is expected to be in
121 user-space. You can use this for strings too, e.g. +u0(%si):string will read
122 a string from the address in the register %si that is expected to be in user-
123 space. 'ustring' is a shortcut way of performing the same task. That is,
124 +0(%si):ustring is equivalent to +u0(%si):string.
126 Note that kprobe-event provides the user-memory access syntax but it doesn't
127 use it transparently. This means if you use normal dereference or string type
128 for user memory, it might fail, and may always fail on some architectures. The
129 user has to carefully check if the target data is in kernel or user space.
131 Per-Probe Event Filtering
132 -------------------------
133 Per-probe event filtering feature allows you to set different filter on each
134 probe and gives you what arguments will be shown in trace buffer. If an event
135 name is specified right after 'p:' or 'r:' in kprobe_events, it adds an event
136 under tracing/events/kprobes/<EVENT>, at the directory you can see 'id',
137 'enable', 'format', 'filter' and 'trigger'.
140 You can enable/disable the probe by writing 1 or 0 on it.
143 This shows the format of this probe event.
146 You can write filtering rules of this event.
149 This shows the id of this probe event.
152 This allows to install trigger commands which are executed when the event is
153 hit (for details, see Documentation/trace/events.rst, section 6).
157 You can check the total number of probe hits and probe miss-hits via
158 /sys/kernel/tracing/kprobe_profile.
159 The first column is event name, the second is the number of probe hits,
160 the third is the number of probe miss-hits.
162 Kernel Boot Parameter
163 ---------------------
164 You can add and enable new kprobe events when booting up the kernel by
165 "kprobe_event=" parameter. The parameter accepts a semicolon-delimited
166 kprobe events, which format is similar to the kprobe_events.
167 The difference is that the probe definition parameters are comma-delimited
168 instead of space. For example, adding myprobe event on do_sys_open like below::
170 p:myprobe do_sys_open dfd=%ax filename=%dx flags=%cx mode=+4($stack)
172 should be below for kernel boot parameter (just replace spaces with comma)::
174 p:myprobe,do_sys_open,dfd=%ax,filename=%dx,flags=%cx,mode=+4($stack)
179 To add a probe as a new event, write a new definition to kprobe_events
182 echo 'p:myprobe do_sys_open dfd=%ax filename=%dx flags=%cx mode=+4($stack)' > /sys/kernel/tracing/kprobe_events
184 This sets a kprobe on the top of do_sys_open() function with recording
185 1st to 4th arguments as "myprobe" event. Note, which register/stack entry is
186 assigned to each function argument depends on arch-specific ABI. If you unsure
187 the ABI, please try to use probe subcommand of perf-tools (you can find it
189 As this example shows, users can choose more familiar names for each arguments.
192 echo 'r:myretprobe do_sys_open $retval' >> /sys/kernel/tracing/kprobe_events
194 This sets a kretprobe on the return point of do_sys_open() function with
195 recording return value as "myretprobe" event.
196 You can see the format of these events via
197 /sys/kernel/tracing/events/kprobes/<EVENT>/format.
200 cat /sys/kernel/tracing/events/kprobes/myprobe/format
204 field:unsigned short common_type; offset:0; size:2; signed:0;
205 field:unsigned char common_flags; offset:2; size:1; signed:0;
206 field:unsigned char common_preempt_count; offset:3; size:1;signed:0;
207 field:int common_pid; offset:4; size:4; signed:1;
209 field:unsigned long __probe_ip; offset:12; size:4; signed:0;
210 field:int __probe_nargs; offset:16; size:4; signed:1;
211 field:unsigned long dfd; offset:20; size:4; signed:0;
212 field:unsigned long filename; offset:24; size:4; signed:0;
213 field:unsigned long flags; offset:28; size:4; signed:0;
214 field:unsigned long mode; offset:32; size:4; signed:0;
217 print fmt: "(%lx) dfd=%lx filename=%lx flags=%lx mode=%lx", REC->__probe_ip,
218 REC->dfd, REC->filename, REC->flags, REC->mode
220 You can see that the event has 4 arguments as in the expressions you specified.
223 echo > /sys/kernel/tracing/kprobe_events
225 This clears all probe points.
230 echo -:myprobe >> kprobe_events
232 This clears probe points selectively.
234 Right after definition, each event is disabled by default. For tracing these
235 events, you need to enable it.
238 echo 1 > /sys/kernel/tracing/events/kprobes/myprobe/enable
239 echo 1 > /sys/kernel/tracing/events/kprobes/myretprobe/enable
241 Use the following command to start tracing in an interval.
244 # echo 1 > tracing_on
246 # echo 0 > tracing_on
248 And you can see the traced information via /sys/kernel/tracing/trace.
251 cat /sys/kernel/tracing/trace
254 # TASK-PID CPU# TIMESTAMP FUNCTION
256 <...>-1447 [001] 1038282.286875: myprobe: (do_sys_open+0x0/0xd6) dfd=3 filename=7fffd1ec4440 flags=8000 mode=0
257 <...>-1447 [001] 1038282.286878: myretprobe: (sys_openat+0xc/0xe <- do_sys_open) $retval=fffffffffffffffe
258 <...>-1447 [001] 1038282.286885: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=40413c flags=8000 mode=1b6
259 <...>-1447 [001] 1038282.286915: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3
260 <...>-1447 [001] 1038282.286969: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=4041c6 flags=98800 mode=10
261 <...>-1447 [001] 1038282.286976: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3
264 Each line shows when the kernel hits an event, and <- SYMBOL means kernel
265 returns from SYMBOL(e.g. "sys_open+0x1b/0x1d <- do_sys_open" means kernel
266 returns from do_sys_open to sys_open+0x1b).