1 ========================================= 1 =========================================
2 Uprobe-tracer: Uprobe-based Event Tracing 2 Uprobe-tracer: Uprobe-based Event Tracing
3 ========================================= 3 =========================================
4 4
5 :Author: Srikar Dronamraju 5 :Author: Srikar Dronamraju
6 6
7 7
8 Overview 8 Overview
9 -------- 9 --------
10 Uprobe based trace events are similar to kprob 10 Uprobe based trace events are similar to kprobe based trace events.
11 To enable this feature, build your kernel with 11 To enable this feature, build your kernel with CONFIG_UPROBE_EVENTS=y.
12 12
13 Similar to the kprobe-event tracer, this doesn 13 Similar to the kprobe-event tracer, this doesn't need to be activated via
14 current_tracer. Instead of that, add probe poi 14 current_tracer. Instead of that, add probe points via
15 /sys/kernel/tracing/uprobe_events, and enable 15 /sys/kernel/tracing/uprobe_events, and enable it via
16 /sys/kernel/tracing/events/uprobes/<EVENT>/ena 16 /sys/kernel/tracing/events/uprobes/<EVENT>/enable.
17 17
18 However unlike kprobe-event tracer, the uprobe 18 However unlike kprobe-event tracer, the uprobe event interface expects the
19 user to calculate the offset of the probepoint 19 user to calculate the offset of the probepoint in the object.
20 20
21 You can also use /sys/kernel/tracing/dynamic_e 21 You can also use /sys/kernel/tracing/dynamic_events instead of
22 uprobe_events. That interface will provide uni 22 uprobe_events. That interface will provide unified access to other
23 dynamic events too. 23 dynamic events too.
24 24
25 Synopsis of uprobe_tracer 25 Synopsis of uprobe_tracer
26 ------------------------- 26 -------------------------
27 :: 27 ::
28 28
29 p[:[GRP/][EVENT]] PATH:OFFSET [FETCHARGS] : 29 p[:[GRP/][EVENT]] PATH:OFFSET [FETCHARGS] : Set a uprobe
30 r[:[GRP/][EVENT]] PATH:OFFSET [FETCHARGS] : 30 r[:[GRP/][EVENT]] PATH:OFFSET [FETCHARGS] : Set a return uprobe (uretprobe)
31 p[:[GRP/][EVENT]] PATH:OFFSET%return [FETCHA 31 p[:[GRP/][EVENT]] PATH:OFFSET%return [FETCHARGS] : Set a return uprobe (uretprobe)
32 -:[GRP/][EVENT] : 32 -:[GRP/][EVENT] : Clear uprobe or uretprobe event
33 33
34 GRP : Group name. If omitted, "upr 34 GRP : Group name. If omitted, "uprobes" is the default value.
35 EVENT : Event name. If omitted, the 35 EVENT : Event name. If omitted, the event name is generated based
36 on PATH+OFFSET. 36 on PATH+OFFSET.
37 PATH : Path to an executable or a l 37 PATH : Path to an executable or a library.
38 OFFSET : Offset where the probe is in 38 OFFSET : Offset where the probe is inserted.
39 OFFSET%return : Offset where the return prob 39 OFFSET%return : Offset where the return probe is inserted.
40 40
41 FETCHARGS : Arguments. Each probe can ha 41 FETCHARGS : Arguments. Each probe can have up to 128 args.
42 %REG : Fetch register REG 42 %REG : Fetch register REG
43 @ADDR : Fetch memory at ADDR (ADDR s 43 @ADDR : Fetch memory at ADDR (ADDR should be in userspace)
44 @+OFFSET : Fetch memory at OFFSET (OFFS 44 @+OFFSET : Fetch memory at OFFSET (OFFSET from same file as PATH)
45 $stackN : Fetch Nth entry of stack (N 45 $stackN : Fetch Nth entry of stack (N >= 0)
46 $stack : Fetch stack address. 46 $stack : Fetch stack address.
47 $retval : Fetch return value.(\*1) 47 $retval : Fetch return value.(\*1)
48 $comm : Fetch current task comm. 48 $comm : Fetch current task comm.
49 +|-[u]OFFS(FETCHARG) : Fetch memory at FETC 49 +|-[u]OFFS(FETCHARG) : Fetch memory at FETCHARG +|- OFFS address.(\*2)(\*3)
50 \IMM : Store an immediate value to 50 \IMM : Store an immediate value to the argument.
51 NAME=FETCHARG : Set NAME as the argumen 51 NAME=FETCHARG : Set NAME as the argument name of FETCHARG.
52 FETCHARG:TYPE : Set TYPE as the type of 52 FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types
53 (u8/u16/u32/u64/s8/s16/ 53 (u8/u16/u32/u64/s8/s16/s32/s64), hexadecimal types
54 (x8/x16/x32/x64), "stri 54 (x8/x16/x32/x64), "string" and bitfield are supported.
55 55
56 (\*1) only for return probe. 56 (\*1) only for return probe.
57 (\*2) this is useful for fetching a field of 57 (\*2) this is useful for fetching a field of data structures.
58 (\*3) Unlike kprobe event, "u" prefix will j 58 (\*3) Unlike kprobe event, "u" prefix will just be ignored, because uprobe
59 events can access only user-space memo 59 events can access only user-space memory.
60 60
61 Types 61 Types
62 ----- 62 -----
63 Several types are supported for fetch-args. Up 63 Several types are supported for fetch-args. Uprobe tracer will access memory
64 by given type. Prefix 's' and 'u' means those 64 by given type. Prefix 's' and 'u' means those types are signed and unsigned
65 respectively. 'x' prefix implies it is unsigne 65 respectively. 'x' prefix implies it is unsigned. Traced arguments are shown
66 in decimal ('s' and 'u') or hexadecimal ('x'). 66 in decimal ('s' and 'u') or hexadecimal ('x'). Without type casting, 'x32'
67 or 'x64' is used depends on the architecture ( 67 or 'x64' is used depends on the architecture (e.g. x86-32 uses x32, and
68 x86-64 uses x64). 68 x86-64 uses x64).
69 String type is a special type, which fetches a 69 String type is a special type, which fetches a "null-terminated" string from
70 user space. 70 user space.
71 Bitfield is another special type, which takes 71 Bitfield is another special type, which takes 3 parameters, bit-width, bit-
72 offset, and container-size (usually 32). The s 72 offset, and container-size (usually 32). The syntax is::
73 73
74 b<bit-width>@<bit-offset>/<container-size> 74 b<bit-width>@<bit-offset>/<container-size>
75 75
76 For $comm, the default type is "string"; any o 76 For $comm, the default type is "string"; any other type is invalid.
77 77
78 78
79 Event Profiling 79 Event Profiling
80 --------------- 80 ---------------
81 You can check the total number of probe hits p 81 You can check the total number of probe hits per event via
82 /sys/kernel/tracing/uprobe_profile. The first 82 /sys/kernel/tracing/uprobe_profile. The first column is the filename,
83 the second is the event name, the third is the 83 the second is the event name, the third is the number of probe hits.
84 84
85 Usage examples 85 Usage examples
86 -------------- 86 --------------
87 * Add a probe as a new uprobe event, write a 87 * Add a probe as a new uprobe event, write a new definition to uprobe_events
88 as below (sets a uprobe at an offset of 0x4 88 as below (sets a uprobe at an offset of 0x4245c0 in the executable /bin/bash)::
89 89
90 echo 'p /bin/bash:0x4245c0' > /sys/kernel/ 90 echo 'p /bin/bash:0x4245c0' > /sys/kernel/tracing/uprobe_events
91 91
92 * Add a probe as a new uretprobe event:: 92 * Add a probe as a new uretprobe event::
93 93
94 echo 'r /bin/bash:0x4245c0' > /sys/kernel/ 94 echo 'r /bin/bash:0x4245c0' > /sys/kernel/tracing/uprobe_events
95 95
96 * Unset registered event:: 96 * Unset registered event::
97 97
98 echo '-:p_bash_0x4245c0' >> /sys/kernel/tr 98 echo '-:p_bash_0x4245c0' >> /sys/kernel/tracing/uprobe_events
99 99
100 * Print out the events that are registered:: 100 * Print out the events that are registered::
101 101
102 cat /sys/kernel/tracing/uprobe_events 102 cat /sys/kernel/tracing/uprobe_events
103 103
104 * Clear all events:: 104 * Clear all events::
105 105
106 echo > /sys/kernel/tracing/uprobe_events 106 echo > /sys/kernel/tracing/uprobe_events
107 107
108 Following example shows how to dump the instru 108 Following example shows how to dump the instruction pointer and %ax register
109 at the probed text address. Probe zfree functi 109 at the probed text address. Probe zfree function in /bin/zsh::
110 110
111 # cd /sys/kernel/tracing/ 111 # cd /sys/kernel/tracing/
112 # cat /proc/`pgrep zsh`/maps | grep /bin/z 112 # cat /proc/`pgrep zsh`/maps | grep /bin/zsh | grep r-xp
113 00400000-0048a000 r-xp 00000000 08:03 1309 113 00400000-0048a000 r-xp 00000000 08:03 130904 /bin/zsh
114 # objdump -T /bin/zsh | grep -w zfree 114 # objdump -T /bin/zsh | grep -w zfree
115 0000000000446420 g DF .text 0000000000 115 0000000000446420 g DF .text 0000000000000012 Base zfree
116 116
117 0x46420 is the offset of zfree in object /bin/ 117 0x46420 is the offset of zfree in object /bin/zsh that is loaded at
118 0x00400000. Hence the command to uprobe would 118 0x00400000. Hence the command to uprobe would be::
119 119
120 # echo 'p:zfree_entry /bin/zsh:0x46420 %ip 120 # echo 'p:zfree_entry /bin/zsh:0x46420 %ip %ax' > uprobe_events
121 121
122 And the same for the uretprobe would be:: 122 And the same for the uretprobe would be::
123 123
124 # echo 'r:zfree_exit /bin/zsh:0x46420 %ip 124 # echo 'r:zfree_exit /bin/zsh:0x46420 %ip %ax' >> uprobe_events
125 125
126 .. note:: User has to explicitly calculate the 126 .. note:: User has to explicitly calculate the offset of the probe-point
127 in the object. 127 in the object.
128 128
129 We can see the events that are registered by l 129 We can see the events that are registered by looking at the uprobe_events file.
130 :: 130 ::
131 131
132 # cat uprobe_events 132 # cat uprobe_events
133 p:uprobes/zfree_entry /bin/zsh:0x00046420 133 p:uprobes/zfree_entry /bin/zsh:0x00046420 arg1=%ip arg2=%ax
134 r:uprobes/zfree_exit /bin/zsh:0x00046420 a 134 r:uprobes/zfree_exit /bin/zsh:0x00046420 arg1=%ip arg2=%ax
135 135
136 Format of events can be seen by viewing the fi 136 Format of events can be seen by viewing the file events/uprobes/zfree_entry/format.
137 :: 137 ::
138 138
139 # cat events/uprobes/zfree_entry/format 139 # cat events/uprobes/zfree_entry/format
140 name: zfree_entry 140 name: zfree_entry
141 ID: 922 141 ID: 922
142 format: 142 format:
143 field:unsigned short common_type; 143 field:unsigned short common_type; offset:0; size:2; signed:0;
144 field:unsigned char common_flags; 144 field:unsigned char common_flags; offset:2; size:1; signed:0;
145 field:unsigned char common_preempt_co 145 field:unsigned char common_preempt_count; offset:3; size:1; signed:0;
146 field:int common_pid; 146 field:int common_pid; offset:4; size:4; signed:1;
147 field:int common_padding; 147 field:int common_padding; offset:8; size:4; signed:1;
148 148
149 field:unsigned long __probe_ip; 149 field:unsigned long __probe_ip; offset:12; size:4; signed:0;
150 field:u32 arg1; 150 field:u32 arg1; offset:16; size:4; signed:0;
151 field:u32 arg2; 151 field:u32 arg2; offset:20; size:4; signed:0;
152 152
153 print fmt: "(%lx) arg1=%lx arg2=%lx", REC- 153 print fmt: "(%lx) arg1=%lx arg2=%lx", REC->__probe_ip, REC->arg1, REC->arg2
154 154
155 Right after definition, each event is disabled 155 Right after definition, each event is disabled by default. For tracing these
156 events, you need to enable it by:: 156 events, you need to enable it by::
157 157
158 # echo 1 > events/uprobes/enable 158 # echo 1 > events/uprobes/enable
159 159
160 Lets start tracing, sleep for some time and st 160 Lets start tracing, sleep for some time and stop tracing.
161 :: 161 ::
162 162
163 # echo 1 > tracing_on 163 # echo 1 > tracing_on
164 # sleep 20 164 # sleep 20
165 # echo 0 > tracing_on 165 # echo 0 > tracing_on
166 166
167 Also, you can disable the event by:: 167 Also, you can disable the event by::
168 168
169 # echo 0 > events/uprobes/enable 169 # echo 0 > events/uprobes/enable
170 170
171 And you can see the traced information via /sy 171 And you can see the traced information via /sys/kernel/tracing/trace.
172 :: 172 ::
173 173
174 # cat trace 174 # cat trace
175 # tracer: nop 175 # tracer: nop
176 # 176 #
177 # TASK-PID CPU# TIMESTAMP 177 # TASK-PID CPU# TIMESTAMP FUNCTION
178 # | | | | 178 # | | | | |
179 zsh-24842 [006] 258544.995456 179 zsh-24842 [006] 258544.995456: zfree_entry: (0x446420) arg1=446420 arg2=79
180 zsh-24842 [007] 258545.000270 180 zsh-24842 [007] 258545.000270: zfree_exit: (0x446540 <- 0x446420) arg1=446540 arg2=0
181 zsh-24842 [002] 258545.043929 181 zsh-24842 [002] 258545.043929: zfree_entry: (0x446420) arg1=446420 arg2=79
182 zsh-24842 [004] 258547.046129 182 zsh-24842 [004] 258547.046129: zfree_exit: (0x446540 <- 0x446420) arg1=446540 arg2=0
183 183
184 Output shows us uprobe was triggered for a pid 184 Output shows us uprobe was triggered for a pid 24842 with ip being 0x446420
185 and contents of ax register being 79. And uret 185 and contents of ax register being 79. And uretprobe was triggered with ip at
186 0x446540 with counterpart function entry at 0x 186 0x446540 with counterpart function entry at 0x446420.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.