1 Kprobe-based Event Tracing 2 ========================== 3 4 Documentation is written by Masami Hiramatsu 5 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 body except for __kprobes 12functions). Unlike the Tracepoint based event, this can be added and removed 13dynamically, on the fly. 14 15To enable this feature, build your kernel with CONFIG_KPROBE_TRACING=y. 16 17Similar to the events tracer, this doesn't need to be activated via 18current_tracer. Instead of that, add probe points via 19/sys/kernel/debug/tracing/kprobe_events, and enable it via 20/sys/kernel/debug/tracing/events/kprobes/<EVENT>/enabled. 21 22 23Synopsis of kprobe_events 24------------------------- 25 p[:[GRP/]EVENT] SYMBOL[+offs]|MEMADDR [FETCHARGS] : Set a probe 26 r[:[GRP/]EVENT] SYMBOL[+0] [FETCHARGS] : Set a return probe 27 -:[GRP/]EVENT : Clear a probe 28 29 GRP : Group name. If omitted, use "kprobes" for it. 30 EVENT : Event name. If omitted, the event name is generated 31 based on SYMBOL+offs or MEMADDR. 32 SYMBOL[+offs] : Symbol+offset where the probe is inserted. 33 MEMADDR : Address where the probe is inserted. 34 35 FETCHARGS : Arguments. Each probe can have up to 128 args. 36 %REG : Fetch register REG 37 @ADDR : Fetch memory at ADDR (ADDR should be in kernel) 38 @SYM[+|-offs] : Fetch memory at SYM +|- offs (SYM should be a data symbol) 39 $stackN : Fetch Nth entry of stack (N >= 0) 40 $stack : Fetch stack address. 41 $retval : Fetch return value.(*) 42 +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(**) 43 NAME=FETCHARG : Set NAME as the argument name of FETCHARG. 44 FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types 45 (u8/u16/u32/u64/s8/s16/s32/s64) and string are supported. 46 47 (*) only for return probe. 48 (**) this is useful for fetching a field of data structures. 49 50 51Per-Probe Event Filtering 52------------------------- 53 Per-probe event filtering feature allows you to set different filter on each 54probe and gives you what arguments will be shown in trace buffer. If an event 55name is specified right after 'p:' or 'r:' in kprobe_events, it adds an event 56under tracing/events/kprobes/<EVENT>, at the directory you can see 'id', 57'enabled', 'format' and 'filter'. 58 59enabled: 60 You can enable/disable the probe by writing 1 or 0 on it. 61 62format: 63 This shows the format of this probe event. 64 65filter: 66 You can write filtering rules of this event. 67 68id: 69 This shows the id of this probe event. 70 71 72Event Profiling 73--------------- 74 You can check the total number of probe hits and probe miss-hits via 75/sys/kernel/debug/tracing/kprobe_profile. 76 The first column is event name, the second is the number of probe hits, 77the third is the number of probe miss-hits. 78 79 80Usage examples 81-------------- 82To add a probe as a new event, write a new definition to kprobe_events 83as below. 84 85 echo 'p:myprobe do_sys_open dfd=%ax filename=%dx flags=%cx mode=+4($stack)' > /sys/kernel/debug/tracing/kprobe_events 86 87 This sets a kprobe on the top of do_sys_open() function with recording 881st to 4th arguments as "myprobe" event. Note, which register/stack entry is 89assigned to each function argument depends on arch-specific ABI. If you unsure 90the ABI, please try to use probe subcommand of perf-tools (you can find it 91under tools/perf/). 92As this example shows, users can choose more familiar names for each arguments. 93 94 echo 'r:myretprobe do_sys_open $retval' >> /sys/kernel/debug/tracing/kprobe_events 95 96 This sets a kretprobe on the return point of do_sys_open() function with 97recording return value as "myretprobe" event. 98 You can see the format of these events via 99/sys/kernel/debug/tracing/events/kprobes/<EVENT>/format. 100 101 cat /sys/kernel/debug/tracing/events/kprobes/myprobe/format 102name: myprobe 103ID: 780 104format: 105 field:unsigned short common_type; offset:0; size:2; signed:0; 106 field:unsigned char common_flags; offset:2; size:1; signed:0; 107 field:unsigned char common_preempt_count; offset:3; size:1;signed:0; 108 field:int common_pid; offset:4; size:4; signed:1; 109 field:int common_lock_depth; offset:8; size:4; signed:1; 110 111 field:unsigned long __probe_ip; offset:12; size:4; signed:0; 112 field:int __probe_nargs; offset:16; size:4; signed:1; 113 field:unsigned long dfd; offset:20; size:4; signed:0; 114 field:unsigned long filename; offset:24; size:4; signed:0; 115 field:unsigned long flags; offset:28; size:4; signed:0; 116 field:unsigned long mode; offset:32; size:4; signed:0; 117 118 119print fmt: "(%lx) dfd=%lx filename=%lx flags=%lx mode=%lx", REC->__probe_ip, 120REC->dfd, REC->filename, REC->flags, REC->mode 121 122 You can see that the event has 4 arguments as in the expressions you specified. 123 124 echo > /sys/kernel/debug/tracing/kprobe_events 125 126 This clears all probe points. 127 128 Or, 129 130 echo -:myprobe >> kprobe_events 131 132 This clears probe points selectively. 133 134 Right after definition, each event is disabled by default. For tracing these 135events, you need to enable it. 136 137 echo 1 > /sys/kernel/debug/tracing/events/kprobes/myprobe/enable 138 echo 1 > /sys/kernel/debug/tracing/events/kprobes/myretprobe/enable 139 140 And you can see the traced information via /sys/kernel/debug/tracing/trace. 141 142 cat /sys/kernel/debug/tracing/trace 143# tracer: nop 144# 145# TASK-PID CPU# TIMESTAMP FUNCTION 146# | | | | | 147 <...>-1447 [001] 1038282.286875: myprobe: (do_sys_open+0x0/0xd6) dfd=3 filename=7fffd1ec4440 flags=8000 mode=0 148 <...>-1447 [001] 1038282.286878: myretprobe: (sys_openat+0xc/0xe <- do_sys_open) $retval=fffffffffffffffe 149 <...>-1447 [001] 1038282.286885: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=40413c flags=8000 mode=1b6 150 <...>-1447 [001] 1038282.286915: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3 151 <...>-1447 [001] 1038282.286969: myprobe: (do_sys_open+0x0/0xd6) dfd=ffffff9c filename=4041c6 flags=98800 mode=10 152 <...>-1447 [001] 1038282.286976: myretprobe: (sys_open+0x1b/0x1d <- do_sys_open) $retval=3 153 154 155 Each line shows when the kernel hits an event, and <- SYMBOL means kernel 156returns from SYMBOL(e.g. "sys_open+0x1b/0x1d <- do_sys_open" means kernel 157returns from do_sys_open to sys_open+0x1b). 158 159