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  -:[GRP/]EVENT						: Clear a probe
34
35 GRP		: Group name. If omitted, use "kprobes" for it.
36 EVENT		: Event name. If omitted, the event name is generated
37		  based on SYM+offs or MEMADDR.
38 MOD		: Module name which has given SYM.
39 SYM[+offs]	: Symbol+offset where the probe is inserted.
40 MEMADDR	: Address where the probe is inserted.
41 MAXACTIVE	: Maximum number of instances of the specified function that
42		  can be probed simultaneously, or 0 for the default value
43		  as defined in Documentation/kprobes.txt section 1.3.1.
44
45 FETCHARGS	: Arguments. Each probe can have up to 128 args.
46  %REG		: Fetch register REG
47  @ADDR		: Fetch memory at ADDR (ADDR should be in kernel)
48  @SYM[+|-offs]	: Fetch memory at SYM +|- offs (SYM should be a data symbol)
49  $stackN	: Fetch Nth entry of stack (N >= 0)
50  $stack	: Fetch stack address.
51  $argN		: Fetch the Nth function argument. (N >= 1) (\*1)
52  $retval	: Fetch return value.(\*2)
53  $comm		: Fetch current task comm.
54  +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(\*3)
55  NAME=FETCHARG : Set NAME as the argument name of FETCHARG.
56  FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types
57		  (u8/u16/u32/u64/s8/s16/s32/s64), hexadecimal types
58		  (x8/x16/x32/x64), "string" and bitfield are supported.
59
60  (\*1) only for the probe on function entry (offs == 0).
61  (\*2) only for return probe.
62  (\*3) this is useful for fetching a field of data structures.
63
64Types
65-----
66Several types are supported for fetch-args. Kprobe tracer will access memory
67by given type. Prefix 's' and 'u' means those types are signed and unsigned
68respectively. 'x' prefix implies it is unsigned. Traced arguments are shown
69in decimal ('s' and 'u') or hexadecimal ('x'). Without type casting, 'x32'
70or 'x64' is used depends on the architecture (e.g. x86-32 uses x32, and
71x86-64 uses x64).
72These value types can be an array. To record array data, you can add '[N]'
73(where N is a fixed number, less than 64) to the base type.
74E.g. 'x16[4]' means an array of x16 (2bytes hex) with 4 elements.
75Note that the array can be applied to memory type fetchargs, you can not
76apply it to registers/stack-entries etc. (for example, '$stack1:x8[8]' is
77wrong, but '+8($stack):x8[8]' is OK.)
78String type is a special type, which fetches a "null-terminated" string from
79kernel space. This means it will fail and store NULL if the string container
80has been paged out.
81The string array type is a bit different from other types. For other base
82types, <base-type>[1] is equal to <base-type> (e.g. +0(%di):x32[1] is same
83as +0(%di):x32.) But string[1] is not equal to string. The string type itself
84represents "char array", but string array type represents "char * array".
85So, for example, +0(%di):string[1] is equal to +0(+0(%di)):string.
86Bitfield is another special type, which takes 3 parameters, bit-width, bit-
87offset, and container-size (usually 32). The syntax is::
88
89 b<bit-width>@<bit-offset>/<container-size>
90
91Symbol type('symbol') is an alias of u32 or u64 type (depends on BITS_PER_LONG)
92which shows given pointer in "symbol+offset" style.
93For $comm, the default type is "string"; any other type is invalid.
94
95
96Per-Probe Event Filtering
97-------------------------
98Per-probe event filtering feature allows you to set different filter on each
99probe and gives you what arguments will be shown in trace buffer. If an event
100name is specified right after 'p:' or 'r:' in kprobe_events, it adds an event
101under tracing/events/kprobes/<EVENT>, at the directory you can see 'id',
102'enable', 'format', 'filter' and 'trigger'.
103
104enable:
105  You can enable/disable the probe by writing 1 or 0 on it.
106
107format:
108  This shows the format of this probe event.
109
110filter:
111  You can write filtering rules of this event.
112
113id:
114  This shows the id of this probe event.
115
116trigger:
117  This allows to install trigger commands which are executed when the event is
118  hit (for details, see Documentation/trace/events.rst, section 6).
119
120Event Profiling
121---------------
122You can check the total number of probe hits and probe miss-hits via
123/sys/kernel/debug/tracing/kprobe_profile.
124The first column is event name, the second is the number of probe hits,
125the third is the number of probe miss-hits.
126
127
128Usage examples
129--------------
130To add a probe as a new event, write a new definition to kprobe_events
131as below::
132
133  echo 'p:myprobe do_sys_open dfd=%ax filename=%dx flags=%cx mode=+4($stack)' > /sys/kernel/debug/tracing/kprobe_events
134
135This sets a kprobe on the top of do_sys_open() function with recording
1361st to 4th arguments as "myprobe" event. Note, which register/stack entry is
137assigned to each function argument depends on arch-specific ABI. If you unsure
138the ABI, please try to use probe subcommand of perf-tools (you can find it
139under tools/perf/).
140As this example shows, users can choose more familiar names for each arguments.
141::
142
143  echo 'r:myretprobe do_sys_open $retval' >> /sys/kernel/debug/tracing/kprobe_events
144
145This sets a kretprobe on the return point of do_sys_open() function with
146recording return value as "myretprobe" event.
147You can see the format of these events via
148/sys/kernel/debug/tracing/events/kprobes/<EVENT>/format.
149::
150
151  cat /sys/kernel/debug/tracing/events/kprobes/myprobe/format
152  name: myprobe
153  ID: 780
154  format:
155          field:unsigned short common_type;       offset:0;       size:2; signed:0;
156          field:unsigned char common_flags;       offset:2;       size:1; signed:0;
157          field:unsigned char common_preempt_count;       offset:3; size:1;signed:0;
158          field:int common_pid;   offset:4;       size:4; signed:1;
159
160          field:unsigned long __probe_ip; offset:12;      size:4; signed:0;
161          field:int __probe_nargs;        offset:16;      size:4; signed:1;
162          field:unsigned long dfd;        offset:20;      size:4; signed:0;
163          field:unsigned long filename;   offset:24;      size:4; signed:0;
164          field:unsigned long flags;      offset:28;      size:4; signed:0;
165          field:unsigned long mode;       offset:32;      size:4; signed:0;
166
167
168  print fmt: "(%lx) dfd=%lx filename=%lx flags=%lx mode=%lx", REC->__probe_ip,
169  REC->dfd, REC->filename, REC->flags, REC->mode
170
171You can see that the event has 4 arguments as in the expressions you specified.
172::
173
174  echo > /sys/kernel/debug/tracing/kprobe_events
175
176This clears all probe points.
177
178Or,
179::
180
181  echo -:myprobe >> kprobe_events
182
183This clears probe points selectively.
184
185Right after definition, each event is disabled by default. For tracing these
186events, you need to enable it.
187::
188
189  echo 1 > /sys/kernel/debug/tracing/events/kprobes/myprobe/enable
190  echo 1 > /sys/kernel/debug/tracing/events/kprobes/myretprobe/enable
191
192And you can see the traced information via /sys/kernel/debug/tracing/trace.
193::
194
195  cat /sys/kernel/debug/tracing/trace
196  # tracer: nop
197  #
198  #           TASK-PID    CPU#    TIMESTAMP  FUNCTION
199  #              | |       |          |         |
200             <...>-1447  [001] 1038282.286875: myprobe: (do_sys_open+0x0/0xd6) dfd=3 filename=7fffd1ec4440 flags=8000 mode=0
201             <...>-1447  [001] 1038282.286878: myretprobe: (sys_openat+0xc/0xe <- do_sys_open) $retval=fffffffffffffffe
202             <...>-1447  [001] 1038282.286885: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=40413c flags=8000 mode=1b6
203             <...>-1447  [001] 1038282.286915: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3
204             <...>-1447  [001] 1038282.286969: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=4041c6 flags=98800 mode=10
205             <...>-1447  [001] 1038282.286976: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3
206
207
208Each line shows when the kernel hits an event, and <- SYMBOL means kernel
209returns from SYMBOL(e.g. "sys_open+0x1b/0x1d <- do_sys_open" means kernel
210returns from do_sys_open to sys_open+0x1b).
211
212