1==========================
2Kprobe-based Event Tracing
3==========================
4
5:Author: Masami Hiramatsu
6
7Overview
8--------
9These events are similar to tracepoint based events. Instead of Tracepoint,
10this is based on kprobes (kprobe and kretprobe). So it can probe wherever
11kprobes can probe (this means, all functions except those with
12__kprobes/nokprobe_inline annotation and those marked NOKPROBE_SYMBOL).
13Unlike the Tracepoint based event, this can be added and removed
14dynamically, on the fly.
15
16To enable this feature, build your kernel with CONFIG_KPROBE_EVENTS=y.
17
18Similar to the events tracer, this doesn't need to be activated via
19current_tracer. Instead of that, add probe points via
20/sys/kernel/debug/tracing/kprobe_events, and enable it via
21/sys/kernel/debug/tracing/events/kprobes/<EVENT>/enable.
22
23You can also use /sys/kernel/debug/tracing/dynamic_events instead of
24kprobe_events. That interface will provide unified access to other
25dynamic events too.
26
27Synopsis of kprobe_events
28-------------------------
29::
30
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
35
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.
46
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  $comm		: Fetch current task comm.
56  +|-[u]OFFS(FETCHARG) : Fetch memory at FETCHARG +|- OFFS address.(\*3)(\*4)
57  \IMM		: Store an immediate value to the argument.
58  NAME=FETCHARG : Set NAME as the argument name of FETCHARG.
59  FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types
60		  (u8/u16/u32/u64/s8/s16/s32/s64), hexadecimal types
61		  (x8/x16/x32/x64), "string", "ustring", "symbol", "symstr"
62                  and bitfield are supported.
63
64  (\*1) only for the probe on function entry (offs == 0).
65  (\*2) only for return probe.
66  (\*3) this is useful for fetching a field of data structures.
67  (\*4) "u" means user-space dereference. See :ref:`user_mem_access`.
68
69Types
70-----
71Several types are supported for fetch-args. Kprobe tracer will access memory
72by given type. Prefix 's' and 'u' means those types are signed and unsigned
73respectively. 'x' prefix implies it is unsigned. Traced arguments are shown
74in decimal ('s' and 'u') or hexadecimal ('x'). Without type casting, 'x32'
75or 'x64' is used depends on the architecture (e.g. x86-32 uses x32, and
76x86-64 uses x64).
77These value types can be an array. To record array data, you can add '[N]'
78(where N is a fixed number, less than 64) to the base type.
79E.g. 'x16[4]' means an array of x16 (2bytes hex) with 4 elements.
80Note that the array can be applied to memory type fetchargs, you can not
81apply it to registers/stack-entries etc. (for example, '$stack1:x8[8]' is
82wrong, but '+8($stack):x8[8]' is OK.)
83String type is a special type, which fetches a "null-terminated" string from
84kernel space. This means it will fail and store NULL if the string container
85has been paged out. "ustring" type is an alternative of string for user-space.
86See :ref:`user_mem_access` for more info..
87The string array type is a bit different from other types. For other base
88types, <base-type>[1] is equal to <base-type> (e.g. +0(%di):x32[1] is same
89as +0(%di):x32.) But string[1] is not equal to string. The string type itself
90represents "char array", but string array type represents "char * array".
91So, for example, +0(%di):string[1] is equal to +0(+0(%di)):string.
92Bitfield is another special type, which takes 3 parameters, bit-width, bit-
93offset, and container-size (usually 32). The syntax is::
94
95 b<bit-width>@<bit-offset>/<container-size>
96
97Symbol type('symbol') is an alias of u32 or u64 type (depends on BITS_PER_LONG)
98which shows given pointer in "symbol+offset" style.
99On the other hand, symbol-string type ('symstr') converts the given address to
100"symbol+offset/symbolsize" style and stores it as a null-terminated string.
101With 'symstr' type, you can filter the event with wildcard pattern of the
102symbols, and you don't need to solve symbol name by yourself.
103For $comm, the default type is "string"; any other type is invalid.
104
105.. _user_mem_access:
106
107User Memory Access
108------------------
109Kprobe events supports user-space memory access. For that purpose, you can use
110either user-space dereference syntax or 'ustring' type.
111
112The user-space dereference syntax allows you to access a field of a data
113structure in user-space. This is done by adding the "u" prefix to the
114dereference syntax. For example, +u4(%si) means it will read memory from the
115address in the register %si offset by 4, and the memory is expected to be in
116user-space. You can use this for strings too, e.g. +u0(%si):string will read
117a string from the address in the register %si that is expected to be in user-
118space. 'ustring' is a shortcut way of performing the same task. That is,
119+0(%si):ustring is equivalent to +u0(%si):string.
120
121Note that kprobe-event provides the user-memory access syntax but it doesn't
122use it transparently. This means if you use normal dereference or string type
123for user memory, it might fail, and may always fail on some archs. The user
124has to carefully check if the target data is in kernel or user space.
125
126Per-Probe Event Filtering
127-------------------------
128Per-probe event filtering feature allows you to set different filter on each
129probe and gives you what arguments will be shown in trace buffer. If an event
130name is specified right after 'p:' or 'r:' in kprobe_events, it adds an event
131under tracing/events/kprobes/<EVENT>, at the directory you can see 'id',
132'enable', 'format', 'filter' and 'trigger'.
133
134enable:
135  You can enable/disable the probe by writing 1 or 0 on it.
136
137format:
138  This shows the format of this probe event.
139
140filter:
141  You can write filtering rules of this event.
142
143id:
144  This shows the id of this probe event.
145
146trigger:
147  This allows to install trigger commands which are executed when the event is
148  hit (for details, see Documentation/trace/events.rst, section 6).
149
150Event Profiling
151---------------
152You can check the total number of probe hits and probe miss-hits via
153/sys/kernel/debug/tracing/kprobe_profile.
154The first column is event name, the second is the number of probe hits,
155the third is the number of probe miss-hits.
156
157Kernel Boot Parameter
158---------------------
159You can add and enable new kprobe events when booting up the kernel by
160"kprobe_event=" parameter. The parameter accepts a semicolon-delimited
161kprobe events, which format is similar to the kprobe_events.
162The difference is that the probe definition parameters are comma-delimited
163instead of space. For example, adding myprobe event on do_sys_open like below
164
165  p:myprobe do_sys_open dfd=%ax filename=%dx flags=%cx mode=+4($stack)
166
167should be below for kernel boot parameter (just replace spaces with comma)
168
169  p:myprobe,do_sys_open,dfd=%ax,filename=%dx,flags=%cx,mode=+4($stack)
170
171
172Usage examples
173--------------
174To add a probe as a new event, write a new definition to kprobe_events
175as below::
176
177  echo 'p:myprobe do_sys_open dfd=%ax filename=%dx flags=%cx mode=+4($stack)' > /sys/kernel/debug/tracing/kprobe_events
178
179This sets a kprobe on the top of do_sys_open() function with recording
1801st to 4th arguments as "myprobe" event. Note, which register/stack entry is
181assigned to each function argument depends on arch-specific ABI. If you unsure
182the ABI, please try to use probe subcommand of perf-tools (you can find it
183under tools/perf/).
184As this example shows, users can choose more familiar names for each arguments.
185::
186
187  echo 'r:myretprobe do_sys_open $retval' >> /sys/kernel/debug/tracing/kprobe_events
188
189This sets a kretprobe on the return point of do_sys_open() function with
190recording return value as "myretprobe" event.
191You can see the format of these events via
192/sys/kernel/debug/tracing/events/kprobes/<EVENT>/format.
193::
194
195  cat /sys/kernel/debug/tracing/events/kprobes/myprobe/format
196  name: myprobe
197  ID: 780
198  format:
199          field:unsigned short common_type;       offset:0;       size:2; signed:0;
200          field:unsigned char common_flags;       offset:2;       size:1; signed:0;
201          field:unsigned char common_preempt_count;       offset:3; size:1;signed:0;
202          field:int common_pid;   offset:4;       size:4; signed:1;
203
204          field:unsigned long __probe_ip; offset:12;      size:4; signed:0;
205          field:int __probe_nargs;        offset:16;      size:4; signed:1;
206          field:unsigned long dfd;        offset:20;      size:4; signed:0;
207          field:unsigned long filename;   offset:24;      size:4; signed:0;
208          field:unsigned long flags;      offset:28;      size:4; signed:0;
209          field:unsigned long mode;       offset:32;      size:4; signed:0;
210
211
212  print fmt: "(%lx) dfd=%lx filename=%lx flags=%lx mode=%lx", REC->__probe_ip,
213  REC->dfd, REC->filename, REC->flags, REC->mode
214
215You can see that the event has 4 arguments as in the expressions you specified.
216::
217
218  echo > /sys/kernel/debug/tracing/kprobe_events
219
220This clears all probe points.
221
222Or,
223::
224
225  echo -:myprobe >> kprobe_events
226
227This clears probe points selectively.
228
229Right after definition, each event is disabled by default. For tracing these
230events, you need to enable it.
231::
232
233  echo 1 > /sys/kernel/debug/tracing/events/kprobes/myprobe/enable
234  echo 1 > /sys/kernel/debug/tracing/events/kprobes/myretprobe/enable
235
236Use the following command to start tracing in an interval.
237::
238
239    # echo 1 > tracing_on
240    Open something...
241    # echo 0 > tracing_on
242
243And you can see the traced information via /sys/kernel/debug/tracing/trace.
244::
245
246  cat /sys/kernel/debug/tracing/trace
247  # tracer: nop
248  #
249  #           TASK-PID    CPU#    TIMESTAMP  FUNCTION
250  #              | |       |          |         |
251             <...>-1447  [001] 1038282.286875: myprobe: (do_sys_open+0x0/0xd6) dfd=3 filename=7fffd1ec4440 flags=8000 mode=0
252             <...>-1447  [001] 1038282.286878: myretprobe: (sys_openat+0xc/0xe <- do_sys_open) $retval=fffffffffffffffe
253             <...>-1447  [001] 1038282.286885: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=40413c flags=8000 mode=1b6
254             <...>-1447  [001] 1038282.286915: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3
255             <...>-1447  [001] 1038282.286969: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=4041c6 flags=98800 mode=10
256             <...>-1447  [001] 1038282.286976: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3
257
258
259Each line shows when the kernel hits an event, and <- SYMBOL means kernel
260returns from SYMBOL(e.g. "sys_open+0x1b/0x1d <- do_sys_open" means kernel
261returns from do_sys_open to sys_open+0x1b).
262