From: Jovi Zhangwei <jovi.zhangwei@gmail.com>
To: Ingo Molnar <mingo@redhat.org>, Steven Rostedt <rostedt@goodmis.org>
Cc: linux-kernel@vger.kernel.org,
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Jovi Zhangwei <jovi.zhangwei@gmail.com>
Subject: [PATCH 03/28] ktap: add sample scripts
Date: Fri, 28 Mar 2014 09:47:24 -0400 [thread overview]
Message-ID: <1396014469-5937-4-git-send-email-jovi.zhangwei@gmail.com> (raw)
In-Reply-To: <1396014469-5937-1-git-send-email-jovi.zhangwei@gmail.com>
The samples directory is organized by different subsystems,
similar with Dtrace toolkit(http://www.brendangregg.com/dtracetoolkit.html)
It contains:
- helloworld.kp: simple hello world program
- ansi: ansi library for screen display
- basic: some simple examples
- game: tetris game wrote by ktap
- interrupt: collect hardirq and softirq time
- io: open and io related scripts
- mem: kernel memory allocation scripts
- network: networking scripts
- profiling: stack profile and function profile
- schedule: schedule scripts.
schedtimes.kp was inspired by Systemtap schedtimes.stp
- syscalls: tracing syscalls
syslatl.kp/opensnoop.kp/syslist.kp was contributed by Brendan Gregg(from Dtrace toolkit)
- tracepoints: kernel tracepoint count and histogram
- userspace: uprobe(include SDT) scripts
Besides these samples, "ktap Examples" posted by Brendan Gregg
(http://www.brendangregg.com/ktap.html) also have many sample
scripts, he mentioned how to can make flame graph by ktap.
Signed-off-by: Jovi Zhangwei <jovi.zhangwei@gmail.com>
---
tools/ktap/samples/ansi/ansi_color_demo.kp | 22 ++
tools/ktap/samples/basic/backtrace.kp | 6 +
tools/ktap/samples/basic/event_trigger.kp | 27 ++
tools/ktap/samples/basic/event_trigger_ftrace.kp | 27 ++
tools/ktap/samples/basic/ftrace.kp | 8 +
tools/ktap/samples/basic/function_time.kp | 62 +++++
tools/ktap/samples/basic/kretprobe.kp | 6 +
tools/ktap/samples/basic/memcpy_memset.kp | 23 ++
tools/ktap/samples/game/tetris.kp | 297 +++++++++++++++++++++
tools/ktap/samples/helloworld.kp | 3 +
tools/ktap/samples/interrupt/hardirq_time.kp | 25 ++
tools/ktap/samples/interrupt/softirq_time.kp | 24 ++
tools/ktap/samples/io/kprobes-do-sys-open.kp | 20 ++
tools/ktap/samples/io/traceio.kp | 61 +++++
tools/ktap/samples/mem/kmalloc-stack.kp | 12 +
tools/ktap/samples/mem/kmem_count.kp | 29 ++
tools/ktap/samples/network/tcp_ipaddr.kp | 20 ++
tools/ktap/samples/profiling/function_profiler.kp | 41 +++
.../profiling/kprobe_all_kernel_functions.kp | 13 +
tools/ktap/samples/profiling/stack_profile.kp | 27 ++
tools/ktap/samples/schedule/sched_transition.kp | 5 +
tools/ktap/samples/schedule/schedtimes.kp | 131 +++++++++
tools/ktap/samples/syscalls/errinfo.kp | 145 ++++++++++
tools/ktap/samples/syscalls/execve.kp | 9 +
tools/ktap/samples/syscalls/opensnoop.kp | 31 +++
tools/ktap/samples/syscalls/sctop.kp | 13 +
tools/ktap/samples/syscalls/syscalls.kp | 6 +
tools/ktap/samples/syscalls/syscalls_count.kp | 54 ++++
.../samples/syscalls/syscalls_count_by_proc.kp | 22 ++
tools/ktap/samples/syscalls/syslatl.kp | 33 +++
tools/ktap/samples/syscalls/syslist.kp | 31 +++
tools/ktap/samples/tracepoints/eventcount.kp | 210 +++++++++++++++
.../ktap/samples/tracepoints/eventcount_by_proc.kp | 57 ++++
tools/ktap/samples/tracepoints/raw_tracepoint.kp | 15 ++
tools/ktap/samples/tracepoints/tracepoints.kp | 6 +
tools/ktap/samples/userspace/gcc_unwind.kp | 9 +
tools/ktap/samples/userspace/glibc_func_hist.kp | 44 +++
tools/ktap/samples/userspace/glibc_sdt.kp | 11 +
tools/ktap/samples/userspace/glibc_trace.kp | 11 +
tools/ktap/samples/userspace/malloc_free.kp | 20 ++
tools/ktap/samples/userspace/malloc_size_hist.kp | 22 ++
tools/ktap/samples/userspace/pthread.kp | 8 +
42 files changed, 1646 insertions(+)
create mode 100644 tools/ktap/samples/ansi/ansi_color_demo.kp
create mode 100644 tools/ktap/samples/basic/backtrace.kp
create mode 100644 tools/ktap/samples/basic/event_trigger.kp
create mode 100644 tools/ktap/samples/basic/event_trigger_ftrace.kp
create mode 100644 tools/ktap/samples/basic/ftrace.kp
create mode 100644 tools/ktap/samples/basic/function_time.kp
create mode 100644 tools/ktap/samples/basic/kretprobe.kp
create mode 100644 tools/ktap/samples/basic/memcpy_memset.kp
create mode 100644 tools/ktap/samples/game/tetris.kp
create mode 100644 tools/ktap/samples/helloworld.kp
create mode 100644 tools/ktap/samples/interrupt/hardirq_time.kp
create mode 100644 tools/ktap/samples/interrupt/softirq_time.kp
create mode 100644 tools/ktap/samples/io/kprobes-do-sys-open.kp
create mode 100644 tools/ktap/samples/io/traceio.kp
create mode 100644 tools/ktap/samples/mem/kmalloc-stack.kp
create mode 100644 tools/ktap/samples/mem/kmem_count.kp
create mode 100644 tools/ktap/samples/network/tcp_ipaddr.kp
create mode 100644 tools/ktap/samples/profiling/function_profiler.kp
create mode 100644 tools/ktap/samples/profiling/kprobe_all_kernel_functions.kp
create mode 100644 tools/ktap/samples/profiling/stack_profile.kp
create mode 100644 tools/ktap/samples/schedule/sched_transition.kp
create mode 100644 tools/ktap/samples/schedule/schedtimes.kp
create mode 100644 tools/ktap/samples/syscalls/errinfo.kp
create mode 100644 tools/ktap/samples/syscalls/execve.kp
create mode 100644 tools/ktap/samples/syscalls/opensnoop.kp
create mode 100644 tools/ktap/samples/syscalls/sctop.kp
create mode 100644 tools/ktap/samples/syscalls/syscalls.kp
create mode 100644 tools/ktap/samples/syscalls/syscalls_count.kp
create mode 100644 tools/ktap/samples/syscalls/syscalls_count_by_proc.kp
create mode 100644 tools/ktap/samples/syscalls/syslatl.kp
create mode 100644 tools/ktap/samples/syscalls/syslist.kp
create mode 100644 tools/ktap/samples/tracepoints/eventcount.kp
create mode 100644 tools/ktap/samples/tracepoints/eventcount_by_proc.kp
create mode 100644 tools/ktap/samples/tracepoints/raw_tracepoint.kp
create mode 100644 tools/ktap/samples/tracepoints/tracepoints.kp
create mode 100644 tools/ktap/samples/userspace/gcc_unwind.kp
create mode 100644 tools/ktap/samples/userspace/glibc_func_hist.kp
create mode 100644 tools/ktap/samples/userspace/glibc_sdt.kp
create mode 100644 tools/ktap/samples/userspace/glibc_trace.kp
create mode 100644 tools/ktap/samples/userspace/malloc_free.kp
create mode 100644 tools/ktap/samples/userspace/malloc_size_hist.kp
create mode 100644 tools/ktap/samples/userspace/pthread.kp
diff --git a/tools/ktap/samples/ansi/ansi_color_demo.kp b/tools/ktap/samples/ansi/ansi_color_demo.kp
new file mode 100644
index 0000000..2998fe0
--- /dev/null
+++ b/tools/ktap/samples/ansi/ansi_color_demo.kp
@@ -0,0 +1,22 @@
+#!/usr/bin/env ktap
+
+#this script demonstrate how to use ktap to output color text.
+
+ansi.clear_screen()
+
+ansi.set_color(32)
+printf("this line should be Green color\n")
+
+ansi.set_color(31)
+printf("this line should be Red color\n")
+
+ansi.set_color2(34, 43)
+printf("this line should be Blue color, with Yellow background\n")
+
+ansi.reset_color()
+ansi.set_color3(34, 46, 4)
+printf("this line should be Blue color, with Cyan background, underline single attribute\n")
+
+ansi.reset_color()
+ansi.new_line()
+
diff --git a/tools/ktap/samples/basic/backtrace.kp b/tools/ktap/samples/basic/backtrace.kp
new file mode 100644
index 0000000..4c13c1a
--- /dev/null
+++ b/tools/ktap/samples/basic/backtrace.kp
@@ -0,0 +1,6 @@
+#!/usr/bin/env ktap
+
+trace sched:sched_switch {
+ print(stack())
+}
+
diff --git a/tools/ktap/samples/basic/event_trigger.kp b/tools/ktap/samples/basic/event_trigger.kp
new file mode 100644
index 0000000..7bf4720
--- /dev/null
+++ b/tools/ktap/samples/basic/event_trigger.kp
@@ -0,0 +1,27 @@
+#!/usr/bin/env ktap
+
+#This ktap script will output all tracepoint events between
+#sys_enter_open and sys_exit_open, in one cpu.
+
+var soft_disabled = 1
+var this_cpu = 0
+
+trace syscalls:sys_enter_open {
+ print(argstr)
+ soft_disabled = 0
+ this_cpu = cpu
+}
+
+trace *:* {
+ if (soft_disabled == 0 && cpu == this_cpu) {
+ print(argstr)
+ }
+}
+
+trace syscalls:sys_exit_open {
+ print(argstr)
+ if (cpu == this_cpu) {
+ exit()
+ }
+}
+
diff --git a/tools/ktap/samples/basic/event_trigger_ftrace.kp b/tools/ktap/samples/basic/event_trigger_ftrace.kp
new file mode 100644
index 0000000..f2ebfa5
--- /dev/null
+++ b/tools/ktap/samples/basic/event_trigger_ftrace.kp
@@ -0,0 +1,27 @@
+#!/usr/bin/env ktap
+
+#This ktap script will output all function calling between
+#sys_enter_open and sys_exit_open, in one cpu.
+
+var soft_disabled = 1
+var this_cpu = 0
+
+trace syscalls:sys_enter_open {
+ print(argstr)
+ soft_disabled = 0
+ this_cpu = cpu
+}
+
+trace ftrace:function {
+ if (soft_disabled == 0 && cpu == this_cpu) {
+ print(argstr)
+ }
+}
+
+trace syscalls:sys_exit_open {
+ print(argstr)
+ if (cpu == this_cpu) {
+ exit()
+ }
+}
+
diff --git a/tools/ktap/samples/basic/ftrace.kp b/tools/ktap/samples/basic/ftrace.kp
new file mode 100644
index 0000000..22cff5d
--- /dev/null
+++ b/tools/ktap/samples/basic/ftrace.kp
@@ -0,0 +1,8 @@
+#!/usr/bin/env ktap
+
+#Description: output all mutex* function event
+
+trace ftrace:function /ip==mutex*/ {
+ print(cpu, pid, execname, argstr)
+}
+
diff --git a/tools/ktap/samples/basic/function_time.kp b/tools/ktap/samples/basic/function_time.kp
new file mode 100644
index 0000000..51d0f89
--- /dev/null
+++ b/tools/ktap/samples/basic/function_time.kp
@@ -0,0 +1,62 @@
+#!/usr/bin/env ktap
+
+#Demo for thread-local variable
+#
+#Note this kind of function time tracing already handled concurrent issue,
+#but not aware on the recursion problem, user need to aware this limitation,
+#so don't use this script to trace function which could be called recursive.
+
+var self = {}
+var count_max = 0
+var count_min = 0
+var count_num = 0
+var total_time = 0
+
+printf("measure time(us) of function vfs_read\n");
+
+trace probe:vfs_read {
+ if (execname == "ktap") {
+ return
+ }
+
+ self[tid] = gettimeofday_us()
+}
+
+trace probe:vfs_read%return {
+ if (execname == "ktap") {
+ return
+ }
+
+ if (self[tid] == nil) {
+ return
+ }
+
+ var durtion = gettimeofday_us() - self[tid]
+ if (durtion > count_max) {
+ count_max = durtion
+ }
+ var min = count_min
+ if (min == 0 || durtion < min) {
+ count_min = durtion
+ }
+
+ count_num = count_num + 1
+ total_time = total_time + durtion
+
+ self[tid] = nil
+}
+
+trace_end {
+ var avg
+ if (count_num == 0) {
+ avg = 0
+ } else {
+ avg = total_time/count_num
+ }
+
+ printf("avg\tmax\tmin\n");
+ printf("-------------------\n")
+ printf("%d\t%d\t%d\n", avg, count_max, count_min)
+}
+
+
diff --git a/tools/ktap/samples/basic/kretprobe.kp b/tools/ktap/samples/basic/kretprobe.kp
new file mode 100644
index 0000000..e03a7a2
--- /dev/null
+++ b/tools/ktap/samples/basic/kretprobe.kp
@@ -0,0 +1,6 @@
+#!/usr/bin/env ktap
+
+trace probe:vfs_read%return fd=$retval {
+ print(execname, argstr);
+}
+
diff --git a/tools/ktap/samples/basic/memcpy_memset.kp b/tools/ktap/samples/basic/memcpy_memset.kp
new file mode 100644
index 0000000..18008cc
--- /dev/null
+++ b/tools/ktap/samples/basic/memcpy_memset.kp
@@ -0,0 +1,23 @@
+#!/usr/bin/env ktap
+
+# This script collect kernel memcpy/memset size histgoram output.
+
+var h_memcpy = {}
+var h_memset = {}
+
+trace probe:memcpy size=%dx {
+ h_memcpy[arg1] += 1
+}
+
+trace probe:memset size=%dx {
+ h_memset[arg1] += 1
+}
+
+trace_end {
+ print("memcpy size hist:")
+ print_hist(h_memcpy)
+
+ print()
+ print("memset size hist:")
+ print_hist(h_memset)
+}
diff --git a/tools/ktap/samples/game/tetris.kp b/tools/ktap/samples/game/tetris.kp
new file mode 100644
index 0000000..f7fbbf2
--- /dev/null
+++ b/tools/ktap/samples/game/tetris.kp
@@ -0,0 +1,297 @@
+#!/usr/bin/env ktap
+
+#
+# Tetris KTAP Script
+#
+# Copyright (C) 2013/OCT/05 Tadaki SAKAI
+#
+# based on stapgames (Systemtap Game Collection)
+# https://github.com/mhiramat/stapgames/blob/master/games/tetris.stp
+#
+# - Requirements
+# Kernel Configuration: CONFIG_KPROBE_EVENT=y
+# CONFIG_EVENT_TRACING=y
+# CONFIG_PERF_EVENTS=y
+# CONFIG_DEBUG_FS=y
+# CPU Architecture : x86_64
+#
+# - Setup
+# $ sudo mount -t debugfs none /sys/kernel/debug/
+#
+# $ git clone https://github.com/ktap/ktap
+# $ cd ktap
+# $ make 2>&1 | tee ../make.log
+# $ sudo make load
+# $ sudo sh -c 'echo 50000 > /sys/module/ktapvm/parameters/max_exec_count'
+#
+# - Run Tetris
+# $ sudo ./ktap samples/game/tetris.kp
+#
+
+#
+# global value
+#
+
+var empty = -1
+
+var key_code = 0
+var point = 0
+var block_number = 0
+var height = 0
+var height_update = 0
+
+var destination_position = {}
+
+var block_data0 = {}
+var block_data1 = {}
+var block_data2 = {}
+var block_data3 = {}
+var block_data4 = {}
+var block_data5 = {}
+var block_data6 = {}
+var block_table = {}
+
+#
+# utils
+#
+
+function rand(max) {
+ var r = gettimeofday_us()
+ if (r < 0) {
+ r = r * -1
+ }
+ return r % max
+}
+
+var display_buffer = {}
+
+function update_display() {
+ var tmp
+ for (i = 0, 239, 1) {
+ if ((i % 12 - 11) != 0) {
+ tmp = ""
+ } else {
+ tmp = "\n"
+ }
+
+ if (display_buffer[240 + i] == empty) {
+ printf(" %s", tmp)
+ } else {
+ var color = display_buffer[240 + i] + 40
+ ansi.set_color2(color, color)
+ printf(" %s", tmp)
+ ansi.reset_color()
+ }
+
+ # clear the display buffer
+ display_buffer[240 + i] = display_buffer[i]
+ }
+
+ printf("%d\n",point)
+}
+
+
+
+#
+# Initialize
+#
+
+# Create blocks
+# block is represented by the position from the center.
+# Every block has "L" part in the center except for a bar.
+block_data0[0] = -11 # non-"L" part for each block
+block_data1[0] = -24
+block_data2[0] = 2
+block_data3[0] = 13
+block_data4[0] = -13
+block_data5[0] = -1
+block_data6[0] = 2
+
+block_table[0] = block_data0
+block_table[1] = block_data1
+block_table[2] = block_data2
+block_table[3] = block_data3
+block_table[4] = block_data4
+block_table[5] = block_data5
+block_table[6] = block_data6
+
+for (i = 0, len(block_table) - 1, 1) {
+ # common "L" part
+ block_table[i][1] = 0
+ block_table[i][2] = 1
+ block_table[i][3] = -12
+}
+
+block_table[6][3] = -1 # bar is not common
+# Position: 1 row has 12 columns,
+# and (x, y) is represented by h = x + y * 12.p
+height = 17 # First block position (center)
+
+for (i = 0, 240, 1) {
+ var tmp
+ # Wall and Floor (sentinel)
+ if (((i % 12) < 2) || (i > 228)) {
+ tmp = 7 # White
+ } else {
+ tmp = empty
+ }
+ display_buffer[i - 1] = tmp
+ display_buffer[240 + i - 1] = tmp
+}
+
+block_number = rand(7)
+
+ansi.clear_screen()
+
+
+#
+# Key Input
+#
+
+trace probe:kbd_event handle=%di event_type=%si event_code=%dx value=%cx {
+ # Only can run it in x86_64
+ #
+ # Register follow x86_64 call conversion:
+ #
+ # x86_64:
+ # %rcx 4 argument
+ # %rdx 3 argument
+ # %rsi 2 argument
+ # %rdi 1 argument
+
+ var event_code = arg4
+ var value = arg5
+
+ if (value != 0) {
+ if ((event_code - 4) != 0) {
+ key_code = event_code
+ }
+ }
+}
+
+
+#
+# timer
+#
+
+tick-200ms {
+ ansi.clear_screen()
+
+ var f = 0 # move/rotate flag
+ var d
+
+ if (key_code != 0) { # if key is pressed
+ if(key_code != 103) { #move left or right
+ # d: movement direction
+ if ((key_code - 105) != 0) {
+ if ((key_code - 106) != 0) {
+ d = 0
+ } else {
+ d = 1
+ }
+ } else {
+ d = -1
+ }
+
+ for (i = 0, 3, 1) { # check if the block can be moved
+ # destination is free
+ if (display_buffer[height +
+ block_table[block_number][i] + d]
+ != empty) {
+ f = 1
+ }
+ }
+ # move if destinations of every block are free
+ if (f == 0) {
+ height = height + d
+ }
+ } else { # rotate
+ for (i = 0, 3, 1) { # check if block can be rotated
+ # each block position
+ var p = block_table[block_number][i]
+
+ # destination x pos(p/12 rounded)
+ var v = (p * 2 + 252) / 24 - 10
+ var w = p - v * 12 # destination y pos
+
+ # destination position
+ destination_position[i] = w * 12 - v
+
+ # check if desetination is free
+ if (display_buffer[height +
+ destination_position[i]] != empty) {
+ f = 1
+ }
+ }
+
+ if (f == 0) {
+ # rotate if destinations of every block
+ # are free
+ for (i = 0, 3, 1) {
+ block_table[block_number][i] =
+ destination_position[i]
+ }
+ }
+ }
+ }
+ key_code = 0 # clear the input key
+
+ f = 0
+ for (i = 0, 3, 1) { # drop 1 row
+ # check if destination is free
+ var p = height + block_table[block_number][i]
+ if (display_buffer[12 + p] != empty) {
+ f = 1
+ }
+
+ # copy the moving block to display buffer
+ display_buffer[240 + p] = block_number
+ }
+
+ if ((f == 1) && (height == 17)) {
+ update_display()
+ exit() # exit if there are block at initial position
+ }
+
+ height_update = !height_update
+ if (height_update != 0) {
+ if(f != 0) { # the block can't drop anymore
+ for (i = 0, 3, 1) {
+ # fix the block
+ display_buffer[height +
+ block_table[block_number][i]] = block_number
+ }
+ # determin the next block
+ block_number = rand(7)
+ height = 17 # make the block to initial position
+ } else {
+ height = height + 12 # drop the block 1 row
+ }
+ }
+
+ var k = 1
+ for (i = 18, 0, -1) { #check if line is filled
+ # search for filled line
+ var j = 10
+ while ((j > 0) &&
+ (display_buffer[i * 12 + j] != empty)) {
+ j = j - 1
+ }
+
+ if (j == 0) { # filled!
+ # add a point: 1 line - 1 point, ..., tetris - 10points
+ point = point + k
+ k = k + 1
+
+ # drop every upper block
+ j = (i + 1) * 12
+ i = i + 1
+ while (j > 2 * 12) {
+ j = j - 1
+ display_buffer[j] = display_buffer[j - 12]
+ }
+ }
+ }
+
+ update_display()
+}
diff --git a/tools/ktap/samples/helloworld.kp b/tools/ktap/samples/helloworld.kp
new file mode 100644
index 0000000..5673c15
--- /dev/null
+++ b/tools/ktap/samples/helloworld.kp
@@ -0,0 +1,3 @@
+#!/usr/bin/env ktap
+
+print("Hello World! I am ktap")
diff --git a/tools/ktap/samples/interrupt/hardirq_time.kp b/tools/ktap/samples/interrupt/hardirq_time.kp
new file mode 100644
index 0000000..f305a41
--- /dev/null
+++ b/tools/ktap/samples/interrupt/hardirq_time.kp
@@ -0,0 +1,25 @@
+#!/usr/bin/env ktap
+
+#this script output each average consumimg time of each hardirq
+
+var s = {}
+var map = {}
+
+trace irq:irq_handler_entry {
+ map[cpu] = gettimeofday_us()
+}
+
+trace irq:irq_handler_exit {
+ var entry_time = map[cpu]
+ if (entry_time == nil) {
+ return;
+ }
+
+ s[arg0] += gettimeofday_us() - entry_time
+ map[cpu] = nil
+}
+
+trace_end {
+ print_hist(s)
+}
+
diff --git a/tools/ktap/samples/interrupt/softirq_time.kp b/tools/ktap/samples/interrupt/softirq_time.kp
new file mode 100644
index 0000000..561733a
--- /dev/null
+++ b/tools/ktap/samples/interrupt/softirq_time.kp
@@ -0,0 +1,24 @@
+#!/usr/bin/env ktap
+
+#this script output each average consumimg time of each softirq line
+var s = {}
+var map = {}
+
+trace irq:softirq_entry {
+ map[cpu] = gettimeofday_us()
+}
+
+trace irq:softirq_exit {
+ var entry_time = map[cpu]
+ if (entry_time == nil) {
+ return;
+ }
+
+ s[arg0] += gettimeofday_us() - entry_time
+ map[cpu] = nil
+}
+
+trace_end {
+ print_hist(s)
+}
+
diff --git a/tools/ktap/samples/io/kprobes-do-sys-open.kp b/tools/ktap/samples/io/kprobes-do-sys-open.kp
new file mode 100644
index 0000000..692f6c4
--- /dev/null
+++ b/tools/ktap/samples/io/kprobes-do-sys-open.kp
@@ -0,0 +1,20 @@
+#!/usr/bin/env ktap
+
+#Only can run it in x86_64
+#
+#Register follow x86_64 call conversion:
+#
+#x86_64:
+# %rcx 4 argument
+# %rdx 3 argument
+# %rsi 2 argument
+# %rdi 1 argument
+
+trace probe:do_sys_open dfd=%di filename=%si flags=%dx mode=%cx {
+ printf("[do_sys_open entry]: (%s) open file (%s)\n",
+ execname, user_string(arg2))
+}
+
+trace probe:do_sys_open%return fd=$retval {
+ printf("[do_sys_open exit]: return fd (%d)\n", arg2)
+}
diff --git a/tools/ktap/samples/io/traceio.kp b/tools/ktap/samples/io/traceio.kp
new file mode 100644
index 0000000..1ef6588
--- /dev/null
+++ b/tools/ktap/samples/io/traceio.kp
@@ -0,0 +1,61 @@
+#! /usr/bin/env ktap
+
+# Based on systemtap traceio.stp
+
+var reads = {}
+var writes = {}
+var total_io = {}
+
+trace syscalls:sys_exit_read {
+ reads[execname] += arg1
+ total_io[execname] += arg1
+}
+
+trace syscalls:sys_exit_write {
+ writes[execname] += arg1
+ total_io[execname] += arg1
+}
+
+function humanread_digit(bytes) {
+ if (bytes > 1024*1024*1024) {
+ return bytes/1024/1024/1024
+ } elseif (bytes > 1024*1024) {
+ return bytes/1024/1024
+ } elseif (bytes > 1024) {
+ return bytes/1024
+ } else {
+ return bytes
+ }
+}
+
+function humanread_x(bytes) {
+ if (bytes > 1024*1024*1024) {
+ return " GiB"
+ } elseif (bytes > 1024*1024) {
+ return " MiB"
+ } elseif (bytes > 1024) {
+ return " KiB"
+ } else {
+ return " B"
+ }
+}
+
+tick-1s {
+ ansi.clear_screen()
+ for (exec, _ in pairs(total_io)) {
+ var readnum = reads[exec]
+ var writenum = writes[exec]
+
+ if (readnum == nil) {
+ readnum = 0
+ }
+ if (writenum == nil) {
+ writenum = 0
+ }
+ printf("%15s r: %12d%s w: %12d%s\n", exec,
+ humanread_digit(readnum), humanread_x(readnum),
+ humanread_digit(writenum), humanread_x(writenum))
+ }
+ printf("\n")
+}
+
diff --git a/tools/ktap/samples/mem/kmalloc-stack.kp b/tools/ktap/samples/mem/kmalloc-stack.kp
new file mode 100644
index 0000000..8406461
--- /dev/null
+++ b/tools/ktap/samples/mem/kmalloc-stack.kp
@@ -0,0 +1,12 @@
+#!/usr/bin/env ktap
+
+var s = {}
+
+trace kmem:kmalloc {
+ s[stack()] += 1
+}
+
+tick-60s {
+ print_hist(s)
+}
+
diff --git a/tools/ktap/samples/mem/kmem_count.kp b/tools/ktap/samples/mem/kmem_count.kp
new file mode 100644
index 0000000..ab5c8a9
--- /dev/null
+++ b/tools/ktap/samples/mem/kmem_count.kp
@@ -0,0 +1,29 @@
+#!/usr/bin/env ktap
+
+var count1 = 0
+trace kmem:kmalloc {
+ count1 += 1
+}
+
+var count2 = 0
+trace kmem:kfree {
+ count2 += 1
+}
+
+var count3 = 0
+trace kmem:mm_page_alloc {
+ count3 += 1
+}
+
+var count4 = 0
+trace kmem:mm_page_free {
+ count4 += 1
+}
+
+trace_end {
+ print("\n")
+ print("kmem:kmalloc:\t", count1)
+ print("kmem:kfree:\t", count2)
+ print("kmem:mm_page_alloc:", count3)
+ print("kmem:mm_page_free:", count4)
+}
diff --git a/tools/ktap/samples/network/tcp_ipaddr.kp b/tools/ktap/samples/network/tcp_ipaddr.kp
new file mode 100644
index 0000000..6363aef
--- /dev/null
+++ b/tools/ktap/samples/network/tcp_ipaddr.kp
@@ -0,0 +1,20 @@
+#!/usr/bin/env ktap
+
+#This script print source and destination IP address of received tcp message
+#
+#Tested in x86_64
+#
+#function tcp_recvmsg prototype:
+#int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+# size_t len, int nonblock, int flags, int *addr_len)
+#
+
+var ip_sock_saddr = net.ip_sock_saddr
+var ip_sock_daddr = net.ip_sock_daddr
+var format_ip_addr = net.format_ip_addr
+
+trace probe:tcp_recvmsg sock=%si {
+ var saddr = format_ip_addr(ip_sock_saddr(arg1))
+ var daddr = format_ip_addr(ip_sock_daddr(arg1))
+ printf("%s -> %s\n", daddr, saddr)
+}
diff --git a/tools/ktap/samples/profiling/function_profiler.kp b/tools/ktap/samples/profiling/function_profiler.kp
new file mode 100644
index 0000000..1414936
--- /dev/null
+++ b/tools/ktap/samples/profiling/function_profiler.kp
@@ -0,0 +1,41 @@
+#!/usr/bin/env ktap
+
+#kernel function profile
+#You can use this script to know what function is called frequently,
+#without enable CONFIG_FUNCTION_PROFILER in kernel.
+
+var s = {}
+
+trace ftrace:function {
+ s[ipof(arg0)] += 1
+}
+
+trace_end {
+ print_hist(s)
+}
+
+#sample output
+#^C
+# value ------------- Distribution ------------- count
+# sub_preempt_count | @@@@@ 34904
+# add_preempt_count | @@@@@ 33435
+# nsecs_to_jiffies64 | @@@ 19919
+# irqtime_account_process_tick... | @ 9970
+# account_idle_time | @ 9880
+# _raw_spin_lock | 5100
+# _raw_spin_unlock | 5021
+# _raw_spin_unlock_irqrestore | 4235
+# _raw_spin_lock_irqsave | 4232
+# __rcu_read_lock | 3373
+# __rcu_read_unlock | 3373
+# lookup_address | 2392
+# pfn_range_is_mapped | 2384
+# update_cfs_rq_blocked_load | 1983
+# idle_cpu | 1808
+# ktime_get | 1394
+# _raw_spin_unlock_irq | 1270
+# _raw_spin_lock_irq | 1091
+# update_curr | 950
+# irqtime_account_irq | 950
+# ... |
+#
diff --git a/tools/ktap/samples/profiling/kprobe_all_kernel_functions.kp b/tools/ktap/samples/profiling/kprobe_all_kernel_functions.kp
new file mode 100644
index 0000000..d42f7de
--- /dev/null
+++ b/tools/ktap/samples/profiling/kprobe_all_kernel_functions.kp
@@ -0,0 +1,13 @@
+#!/usr/bin/env ktap
+
+#enable kprobe on all available kernel functios in /proc/kallsyms.
+#
+#This script is very dangerous, it will softlockup your system.
+#and make your system extremely slow, but the reason is not
+#caused by ktap.
+#
+#DON'T use this script, at least for now.
+
+trace probe:* {
+ print(argstr)
+}
diff --git a/tools/ktap/samples/profiling/stack_profile.kp b/tools/ktap/samples/profiling/stack_profile.kp
new file mode 100644
index 0000000..3595a65
--- /dev/null
+++ b/tools/ktap/samples/profiling/stack_profile.kp
@@ -0,0 +1,27 @@
+#!/usr/bin/env ktap
+
+# This ktap script samples stacktrace of system per 10us,
+# you can use generated output to make a flame graph.
+#
+# Flame Graphs:
+# http://dtrace.org/blogs/brendan/2012/03/17/linux-kernel-performance-flame-graphs/
+#
+#
+# TODO: use aggregation instead of table.
+
+#pre-allocate 2000 record entries, enlarge it if it's not enough
+var s = table.new(0, 2000)
+
+profile-10us {
+ #skip 12 stack entries, and dump all remain entries.
+ s[stack(-1, 12)] += 1
+}
+
+tick-60s {
+ exit()
+}
+
+trace_end {
+ print_hist(s, 100000)
+}
+
diff --git a/tools/ktap/samples/schedule/sched_transition.kp b/tools/ktap/samples/schedule/sched_transition.kp
new file mode 100644
index 0000000..bad90e8
--- /dev/null
+++ b/tools/ktap/samples/schedule/sched_transition.kp
@@ -0,0 +1,5 @@
+#!/usr/bin/env ktap
+
+trace sched:sched_switch {
+ printf("%s ... ", arg0)
+}
diff --git a/tools/ktap/samples/schedule/schedtimes.kp b/tools/ktap/samples/schedule/schedtimes.kp
new file mode 100644
index 0000000..19de1bf
--- /dev/null
+++ b/tools/ktap/samples/schedule/schedtimes.kp
@@ -0,0 +1,131 @@
+#!/usr/vin/env ktap
+
+#schedtimer.kp
+#Initially inspired by Systemtap schedtimes.stp
+#and more bugfree compare with Systemtap's version
+#
+#Note that the time value is associate with pid, not with execname strictly,
+#sometime you will found there have sleep time for command "ls", the reason
+#is that sleep time is belong to parent process bash, so clear on this.
+
+var RUNNING = 0
+var QUEUED = 1
+var SLEEPING = 2
+var DEAD = 64
+
+var run_time = {}
+var queued_time = {}
+var sleep_time = {}
+var io_wait_time = {}
+
+var pid_state = {}
+var pid_names = {}
+var prev_timestamp = {}
+var io_wait = {}
+
+trace sched:sched_switch {
+ var prev_comm = arg0
+ var prev_pid = arg1
+ var prev_state = arg3
+ var next_comm = arg4
+ var next_pid = arg5
+ var t = gettimeofday_us()
+
+ if (pid_state[prev_pid] == nil) {
+ #do nothing
+ } elseif (pid_state[prev_pid] == RUNNING) {
+ run_time[prev_pid] += t - prev_timestamp[prev_pid]
+ } elseif (pid_state[prev_pid] == QUEUED) {
+ #found this:
+ #sched_wakeup comm=foo
+ #sched_switch prev_comm=foo
+ run_time[prev_pid] += t - prev_timestamp[prev_pid]
+ }
+
+ pid_names[prev_pid] = prev_comm
+ prev_timestamp[prev_pid] = t
+
+ if (prev_state == DEAD) {
+ pid_state[prev_pid] = DEAD
+ } elseif (prev_state > 0) {
+ if (in_iowait() == 1) {
+ io_wait[prev_pid] = 1
+ }
+ pid_state[prev_pid] = SLEEPING
+ } elseif (prev_state == 0) {
+ pid_state[prev_pid] = QUEUED
+ }
+
+ if (pid_state[next_pid] == nil) {
+ pid_state[next_pid] = RUNNING
+ } elseif (pid_state[next_pid] == QUEUED) {
+ queued_time[next_pid] += t - prev_timestamp[next_pid]
+ pid_state[next_pid] = RUNNING
+ }
+
+ pid_names[next_pid] = next_comm
+ prev_timestamp[next_pid] = t
+}
+
+trace sched:sched_wakeup, sched:sched_wakeup_new {
+ var comm = arg0
+ var wakeup_pid = arg1
+ var success = arg3
+ var t = gettimeofday_us()
+
+ if (pid_state[wakeup_pid] == nil) {
+ #do nothing
+ } elseif (pid_state[wakeup_pid] == SLEEPING) {
+ var durtion = t - prev_timestamp[wakeup_pid]
+
+ sleep_time[wakeup_pid] += durtion
+ if (io_wait[wakeup_pid] == 1) {
+ io_wait_time[wakeup_pid] += durtion
+ io_wait[wakeup_pid] = 0
+ }
+ } elseif (pid_state[wakeup_pid] == RUNNING) {
+ return
+ }
+
+ pid_names[wakeup_pid] = comm
+ prev_timestamp[wakeup_pid] = t
+ pid_state[wakeup_pid] = QUEUED
+}
+
+trace_end {
+ var t = gettimeofday_us()
+
+ for (_pid, _state in pairs(pid_state)) {
+ var durtion = t - prev_timestamp[_pid]
+ if (_state == SLEEPING) {
+ sleep_time[_pid] += durtion
+ } elseif (_state == QUEUED) {
+ queued_time[_pid] += durtion
+ } elseif (_state == RUNNING) {
+ run_time[_pid] += durtion
+ }
+ }
+
+ printf ("%16s: %6s %10s %10s %10s %10s %10s\n\n",
+ "execname", "pid", "run(us)", "sleep(us)", "io_wait(us)",
+ "queued(us)", "total(us)")
+
+ for (_pid, _time in pairs(run_time)) {
+ if (sleep_time[_pid] == nil) {
+ sleep_time[_pid] = 0
+ }
+ if (queued_time[_pid] == nil) {
+ queued_time[_pid] = 0
+ }
+
+ if (io_wait_time[_pid] == nil) {
+ io_wait_time[_pid] = 0
+ }
+
+ printf("%16s: %6d %10d %10d %10d %10d %10d\n",
+ pid_names[_pid], _pid, run_time[_pid],
+ sleep_time[_pid], io_wait_time[_pid],
+ queued_time[_pid], run_time[_pid] + sleep_time[_pid] +
+ queued_time[_pid]);
+ }
+}
diff --git a/tools/ktap/samples/syscalls/errinfo.kp b/tools/ktap/samples/syscalls/errinfo.kp
new file mode 100644
index 0000000..5a0f7a6
--- /dev/null
+++ b/tools/ktap/samples/syscalls/errinfo.kp
@@ -0,0 +1,145 @@
+#!/usr/bin/env ktap
+
+#errdesc get from include/uapi/asm-generic/errno*.h
+var errdesc = {
+ [1] = "Operation not permitted", #EPERM
+ [2] = "No such file or directory", #ENOENT
+ [3] = "No such process", #ESRCH
+ [4] = "Interrupted system call", #EINRT
+ [5] = "I/O error", #EIO
+ [6] = "No such device or address", #ENXIO
+ [7] = "Argument list too long", #E2BIG
+ [8] = "Exec format error", #ENOEXEC
+ [9] = "Bad file number", #EBADF
+ [10] = "No child processes", #ECHILD
+ [11] = "Try again", #EAGAIN
+ [12] = "Out of memory", #ENOMEM
+ [13] = "Permission denied", #EACCES
+ [14] = "Bad address", #EFAULT
+ [15] = "Block device required", #ENOTBLK
+ [16] = "Device or resource busy", #EBUSY
+ [17] = "File exists", #EEXIST
+ [18] = "Cross-device link", #EXDEV
+ [19] = "No such device", #ENODEV
+ [20] = "Not a directory", #ENOTDIR
+ [21] = "Is a directory", #EISDIR
+ [22] = "Invalid argument", #EINVAL
+ [23] = "File table overflow", #ENFILE
+ [24] = "Too many open files", #EMFILE
+ [25] = "Not a typewriter", #ENOTTY
+ [26] = "Text file busy", #ETXTBSY
+ [27] = "File too large", #EFBIG
+ [28] = "No space left on device", #ENOSPC
+ [29] = "Illegal seek", #ESPIPE
+ [30] = "Read-only file system", #EROFS
+ [31] = "Too many links", #EMLINK
+ [32] = "Broken pipe", #EPIPE
+ [33] = "Math argument out of domain of func", #EDOM
+ [34] = "Math result not representable", #ERANGE
+
+ [35] = "Resource deadlock would occur", #EDEADLK
+ [36] = "File name too long", #ENAMETOOLONG
+ [37] = "No record locks available", #ENOLCK
+ [38] = "Function not implemented", #ENOSYS
+ [39] = "Directory not empty", #ENOTEMPTY
+ [40] = "Too many symbolic links encountered", #ELOOP
+ [42] = "No message of desired type", #ENOMSG
+ [43] = "Identifier removed", #EIDRM
+ [44] = "Channel number out of range", #ECHRNG
+ [45] = "Level 2 not synchronized", #EL2NSYNC
+ [46] = "Level 3 halted", #EL3HLT
+ [47] = "Level 3 reset", #EL3RST
+ [48] = "Link number out of range", #ELNRNG
+ [49] = "Protocol driver not attached", #EUNATCH
+ [50] = "No CSI structure available", #ENOCSI
+ [51] = "Level 2 halted", #EL2HLT
+ [52] = "Invalid exchange", #EBADE
+ [53] = "Invalid request descriptor", #EBADR
+ [54] = "Exchange full", #EXFULL
+ [55] = "No anode", #ENOANO
+ [56] = "Invalid request code", #EBADRQC
+ [57] = "Invalid slot", #EBADSLT
+
+ [59] = "Bad font file format", #EBFONT
+ [60] = "Device not a stream", #ENOSTR
+ [61] = "No data available", #ENODATA
+ [62] = "Timer expired", #ETIME
+ [63] = "Out of streams resources", #ENOSR
+ [64] = "Machine is not on the network", #ENONET
+ [65] = "Package not installed", #ENOPKG
+ [66] = "Object is remote", #EREMOTE
+ [67] = "Link has been severed", #ENOLINK
+ [68] = "Advertise error", #EADV
+ [69] = "Srmount error", #ESRMNT
+ [70] = "Communication error on send", #ECOMM
+ [71] = "Protocol error", #EPROTO
+ [72] = "Multihop attempted", #EMULTIHOP
+ [73] = "RFS specific error", #EDOTDOT
+ [74] = "Not a data message", #EBADMSG
+ [75] = "Value too large for defined data type", #EOVERFLOW
+ [76] = "Name not unique on network", #ENOTUNIQ
+ [77] = "File descriptor in bad state", #EBADFD
+ [78] = "Remote address changed", #EREMCHG
+ [79] = "Can not access a needed shared library", #ELIBACC
+ [80] = "Accessing a corrupted shared library", #ELIBBAD
+ [81] = ".lib section in a.out corrupted", #ELIBSCN
+ [82] = "Attempting to link in too many shared libraries", #ELIBMAX
+ [83] = "Cannot exec a shared library directly", #ELIBEXEC
+ [84] = "Illegal byte sequence", #EILSEQ
+ [85] = "Interrupted system call should be restarted", #ERESTART
+ [86] = "Streams pipe error", #ESTRPIPE
+ [87] = "Too many users", #EUSERS
+ [88] = "Socket operation on non-socket", #ENOTSOCK
+ [89] = "Destination address required", #EDESTADDRREQ
+ [90] = "Message too long", #EMSGSIZE
+ [91] = "Protocol wrong type for socket", #EPROTOTYPE
+ [92] = "Protocol not available", #ENOPROTOOPT
+ [93] = "Protocol not supported", #EPROTONOSUPPORT
+ [94] = "Socket type not supported", #ESOCKTNOSUPPORT
+ [95] = "Operation not supported on transport endpoint", #EOPNOTSUPP
+ [96] = "Protocol family not supported", #EPFNOSUPPORT
+ [97] = "Address family not supported by protocol", #EAFNOSUPPORT
+ [98] = "Address already in use", #EADDRINUSE
+ [99] = "Cannot assign requested address", #EADDRNOTAVAIL
+ [100] = "Network is down", #ENETDOWN
+ [101] = "Network is unreachable", #ENETUNREACH
+ [102] = "Network dropped connection because of reset", #ENETRESET
+ [103] = "Software caused connection abort", #ECONNABORTED
+ [104] = "Connection reset by peer", #ECONNRESET
+ [105] = "No buffer space available", #ENOBUFS
+ [106] = "Transport endpoint is already connected", #EISCONN
+ [107] = "Transport endpoint is not connected", #ENOTCONN
+ [108] = " Cannot send after transport endpoint shutdown", #ESHUTDOWN
+ [109] = "Too many references: cannot splice", #ETOOMANYREFS
+ [110] = "Connection timed out", #ETIMEDOUT
+ [111] = "Connection refused", #ECONNREFUSED
+ [112] = "Host is down", #EHOSTDOWN
+ [113] = "No route to host", #EHOSTUNREACH
+ [114] = "Operation already in progress", #EALREADY
+ [115] = "Operation now in progress", #EINPROGRESS
+ [116] = "Stale NFS file handle", #ESTALE
+ [117] = "Structure needs cleaning", #EUCLEAN
+ [118] = "Not a XENIX named type file", #ENOTNAM
+ [119] = "No XENIX semaphores available", #ENAVAIL
+ [120] = "Is a named type file", #EISNAM
+ [121] = "Remote I/O error", #EREMOTEIO
+ [122] = "Quota exceeded", #EDQUOT
+ [123] = "No medium found", #ENOMEDIUM
+ [124] = "Wrong medium type", #EMEDIUMTYPE
+ [125] = "Operation Canceled", #ECANCELED
+ [126] = "Required key not available", #ENOKEY
+ [127] = "Key has expired", #EKEYEXPIRED
+ [128] = "Key has been revoked", #EKEYREVOKED
+ [129] = "Key was rejected by service", #EKEYREJECTED
+ [130] = "Owner died", #EOWNERDEAD
+ [131] = "State not recoverable", #ENOTRECOVERABLE
+
+}
+
+trace syscalls:sys_exit_* {
+ if (arg1 < 0) {
+ var errno = -arg1
+ printf("%-15s%-20s\t%d\t%-30s\n",
+ execname, probename, errno, errdesc[errno])
+ }
+}
diff --git a/tools/ktap/samples/syscalls/execve.kp b/tools/ktap/samples/syscalls/execve.kp
new file mode 100644
index 0000000..8b4115e
--- /dev/null
+++ b/tools/ktap/samples/syscalls/execve.kp
@@ -0,0 +1,9 @@
+#!/usr/bin/env ktap
+
+#This script trace filename of process execution
+#only tested in x86-64
+
+trace probe:do_execve filename=%di {
+ printf("[do_execve entry]: (%s) name=%s\n", execname,
+ kernel_string(arg1))
+}
diff --git a/tools/ktap/samples/syscalls/opensnoop.kp b/tools/ktap/samples/syscalls/opensnoop.kp
new file mode 100644
index 0000000..5e561b2
--- /dev/null
+++ b/tools/ktap/samples/syscalls/opensnoop.kp
@@ -0,0 +1,31 @@
+#!/usr/local/bin/ktap -q
+#
+# opensnoop.kp trace open syscalls with pathnames and basic info
+#
+# 23-Nov-2013 Brendan Gregg Created this
+
+var path = {}
+
+printf("%5s %6s %-12s %3s %3s %s\n", "UID", "PID", "COMM", "FD", "ERR", "PATH");
+
+trace syscalls:sys_enter_open {
+ path[tid] = user_string(arg1)
+}
+
+trace syscalls:sys_exit_open {
+ var fd
+ var errno
+
+ if (arg1 < 0) {
+ fd = 0
+ errno = -arg1
+ } else {
+ fd = arg1
+ errno = 0
+ }
+
+ printf("%5d %6d %-12s %3d %3d %s\n", uid, pid, execname, fd,
+ errno, path[tid])
+
+ path[tid] = 0
+}
diff --git a/tools/ktap/samples/syscalls/sctop.kp b/tools/ktap/samples/syscalls/sctop.kp
new file mode 100644
index 0000000..bd33d21
--- /dev/null
+++ b/tools/ktap/samples/syscalls/sctop.kp
@@ -0,0 +1,13 @@
+#! /usr/bin/env ktap
+
+var s = {}
+
+trace syscalls:sys_enter_* {
+ s[probename] += 1
+}
+
+tick-5s {
+ ansi.clear_screen()
+ print_hist(s)
+ delete(s)
+}
diff --git a/tools/ktap/samples/syscalls/syscalls.kp b/tools/ktap/samples/syscalls/syscalls.kp
new file mode 100644
index 0000000..4eb332a
--- /dev/null
+++ b/tools/ktap/samples/syscalls/syscalls.kp
@@ -0,0 +1,6 @@
+#!/usr/bin/env ktap
+
+trace syscalls:* {
+ print(cpu, pid, execname, argstr)
+}
+
diff --git a/tools/ktap/samples/syscalls/syscalls_count.kp b/tools/ktap/samples/syscalls/syscalls_count.kp
new file mode 100644
index 0000000..a355eef
--- /dev/null
+++ b/tools/ktap/samples/syscalls/syscalls_count.kp
@@ -0,0 +1,54 @@
+#!/usr/bin/env ktap
+
+var s = {}
+
+trace syscalls:sys_enter_* {
+ s[probename] += 1
+}
+
+trace_end {
+ print_hist(s)
+}
+
+#Result:
+#
+#[root@jovi ktap]# ./ktap samples/syscalls_count.kp
+#^C
+# value ------------- Distribution ------------- count
+# sys_enter_rt_sigprocmask |@@@@@@ 326
+# sys_enter_read |@@@@@ 287
+# sys_enter_close |@@@@ 236
+# sys_enter_open |@@@@ 222
+# sys_enter_stat64 |@@ 132
+# sys_enter_select |@@ 123
+# sys_enter_rt_sigaction |@@ 107
+# sys_enter_poll |@ 72
+# sys_enter_write |@ 70
+# sys_enter_mmap_pgoff |@ 58
+# sys_enter_fstat64 | 41
+# sys_enter_nanosleep | 23
+# sys_enter_access | 20
+# sys_enter_mprotect | 18
+# sys_enter_geteuid | 17
+# sys_enter_getegid | 16
+# sys_enter_getuid | 16
+# sys_enter_getgid | 16
+# sys_enter_brk | 15
+# sys_enter_waitpid | 11
+# sys_enter_time | 10
+# sys_enter_ioctl | 9
+# sys_enter_munmap | 9
+# sys_enter_fcntl64 | 7
+# sys_enter_dup2 | 7
+# sys_enter_clone | 6
+# sys_enter_exit_group | 6
+# sys_enter_execve | 4
+# sys_enter_pipe | 3
+# sys_enter_gettimeofday | 3
+# sys_enter_getdents | 2
+# sys_enter_getgroups | 2
+# sys_enter_statfs64 | 2
+# sys_enter_lseek | 2
+# sys_enter_openat | 1
+# sys_enter_newuname | 1
+
diff --git a/tools/ktap/samples/syscalls/syscalls_count_by_proc.kp b/tools/ktap/samples/syscalls/syscalls_count_by_proc.kp
new file mode 100644
index 0000000..fdd0eff
--- /dev/null
+++ b/tools/ktap/samples/syscalls/syscalls_count_by_proc.kp
@@ -0,0 +1,22 @@
+#!/usr/bin/env ktap
+
+var s = {}
+
+trace syscalls:sys_enter_* {
+ s[execname] += 1
+}
+
+trace_end {
+ print_hist(s)
+}
+
+#Result:
+#
+#[root@jovi ktap]# ./ktap samples/syscalls_count_by_proc.kp
+#^C
+# value ------------- Distribution ------------- count
+# sshd |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 196
+# iscsid |@@@@ 24
+# sendmail |@ 9
+
+
diff --git a/tools/ktap/samples/syscalls/syslatl.kp b/tools/ktap/samples/syscalls/syslatl.kp
new file mode 100644
index 0000000..e9cd91d
--- /dev/null
+++ b/tools/ktap/samples/syscalls/syslatl.kp
@@ -0,0 +1,33 @@
+#!/usr/bin/env ktap
+#
+# syslatl.kp syscall latency linear aggregation
+#
+# 10-Nov-2013 Brendan Gregg Created this
+
+var step = 10 # number of ms per step
+
+var self = {}
+var lats = {}
+var max = 0
+
+trace syscalls:sys_enter_* {
+ self[tid] = gettimeofday_us()
+}
+
+trace syscalls:sys_exit_* {
+ if (self[tid] == nil) { return }
+ var delta = (gettimeofday_us() - self[tid]) / (step * 1000)
+ if (delta > max) { max = delta }
+ lats[delta] += 1
+ self[tid] = nil
+}
+
+trace_end {
+ printf(" %8s %8s\n", "LAT(ms)+", "COUNT");
+ for (i = 0, max, 1) {
+ if (lats[i] == nil) {
+ lats[i] = 0
+ }
+ printf(" %8d %8d\n", i * step, lats[i]);
+ }
+}
diff --git a/tools/ktap/samples/syscalls/syslist.kp b/tools/ktap/samples/syscalls/syslist.kp
new file mode 100644
index 0000000..09f173d
--- /dev/null
+++ b/tools/ktap/samples/syscalls/syslist.kp
@@ -0,0 +1,31 @@
+#!/usr/bin/env ktap
+#
+# syslist.kp syscall latency as a list with counts
+#
+# 10-Nov-2013 Brendan Gregg Created this
+
+var self = {}
+var lats = {}
+var order = {} # a workaround for key sorting
+
+trace syscalls:sys_enter_* {
+ self[tid] = gettimeofday_us()
+}
+
+trace syscalls:sys_exit_* {
+ if (self[tid] == nil) { return }
+ var delta = gettimeofday_us() - self[tid]
+ lats[delta] += 1
+ order[delta] = delta
+ self[tid] = nil
+}
+
+trace_end {
+ printf(" %8s %8s\n", "LAT(us)", "COUNT");
+ #TODO: use a more simple way to sort keys
+
+ #for (lat, dummy in sort_pairs(order, cmp)) {
+ # printf(" %8d %8d\n", lat, lats[lat]);
+ #}
+ print_hist(lats)
+}
diff --git a/tools/ktap/samples/tracepoints/eventcount.kp b/tools/ktap/samples/tracepoints/eventcount.kp
new file mode 100644
index 0000000..a9e197b
--- /dev/null
+++ b/tools/ktap/samples/tracepoints/eventcount.kp
@@ -0,0 +1,210 @@
+#!/usr/bin/env ktap
+
+# showing all tracepoints in histogram style
+
+var s = {}
+
+trace *:* {
+ s[probename] += 1
+}
+
+trace_end {
+ print_hist(s)
+}
+
+#Results:
+#^C
+#
+# value ------------- Distribution ------------- count
+# rcu_utilization |@@@@@ 225289
+# cpu_idle |@@@ 120168
+# sched_wakeup |@@ 91950
+# timer_cancel |@@ 91232
+# timer_start |@@ 91201
+# sched_stat_sleep |@@ 90981
+# timer_expire_exit |@@ 90634
+# timer_expire_entry |@@ 90625
+# hrtimer_cancel |@ 75411
+# hrtimer_start |@ 74946
+# softirq_raise |@ 63117
+# softirq_exit |@ 63109
+# softirq_entry |@ 63094
+# sched_switch |@ 62331
+# sched_stat_wait |@ 60491
+# hrtimer_expire_exit |@ 47538
+# hrtimer_expire_entry |@ 47530
+# sched_stat_runtime | 2780
+# kmem_cache_free | 2684
+# kmem_cache_alloc | 2415
+# kfree | 2288
+# sys_exit | 2145
+# sys_enter | 2145
+# sys_exit_rt_sigprocmask | 1000
+# sys_enter_rt_sigprocmask | 1000
+# timer_init | 912
+# sched_stat_blocked | 685
+# kmalloc | 667
+# workqueue_execute_end | 621
+# workqueue_execute_start | 621
+# sys_enter_select | 566
+# sys_exit_select | 566
+# sys_enter_read | 526
+# sys_exit_read | 526
+# mm_page_free | 478
+# mm_page_alloc | 427
+# mm_page_free_batched | 382
+# net_dev_queue | 296
+# net_dev_xmit | 296
+# consume_skb | 296
+# sys_exit_write | 290
+# sys_enter_write | 290
+# kfree_skb | 289
+# kmem_cache_alloc_node | 269
+# kmalloc_node | 263
+# sys_enter_close | 249
+# sys_exit_close | 249
+# hrtimer_init | 248
+# netif_receive_skb | 242
+# sys_enter_open | 237
+# sys_exit_open | 237
+# napi_poll | 226
+# sched_migrate_task | 207
+# sys_exit_poll | 173
+# sys_enter_poll | 173
+# workqueue_queue_work | 152
+# workqueue_activate_work | 152
+# sys_enter_stat64 | 133
+# sys_exit_stat64 | 133
+# sys_exit_rt_sigaction | 133
+# sys_enter_rt_sigaction | 133
+# irq_handler_entry | 125
+# irq_handler_exit | 125
+# mm_page_alloc_zone_locked | 99
+# sys_exit_mmap_pgoff | 66
+# sys_enter_mmap_pgoff | 66
+# sys_exit_fstat64 | 54
+# sys_enter_fstat64 | 54
+# sys_enter_nanosleep | 51
+# sys_exit_nanosleep | 51
+# block_bio_queue | 46
+# block_bio_remap | 46
+# block_bio_complete | 46
+# mix_pool_bytes | 44
+# mm_page_pcpu_drain | 31
+# sys_exit_time | 23
+# sys_enter_time | 23
+# sys_exit_access | 20
+# sys_enter_access | 20
+# mix_pool_bytes_nolock | 18
+# sys_enter_mprotect | 18
+# sys_exit_mprotect | 18
+# sys_enter_geteuid | 17
+# sys_exit_geteuid | 17
+# sys_enter_munmap | 17
+# sys_exit_munmap | 17
+# block_getrq | 16
+# sys_enter_getuid | 16
+# sys_enter_getgid | 16
+# sys_exit_getgid | 16
+# sys_exit_getuid | 16
+# block_rq_issue | 16
+# scsi_dispatch_cmd_start | 16
+# block_rq_complete | 16
+# scsi_dispatch_cmd_done | 16
+# sys_enter_getegid | 16
+# sys_exit_getegid | 16
+# block_rq_insert | 16
+# skb_copy_datagram_iovec | 15
+# sys_enter_brk | 15
+# sys_exit_brk | 15
+# credit_entropy_bits | 14
+# wbc_writepage | 14
+# sys_exit_clone | 12
+# block_touch_buffer | 12
+# sched_process_wait | 11
+# sys_enter_waitpid | 11
+# sys_exit_waitpid | 11
+# writeback_written | 10
+# writeback_start | 10
+# writeback_queue_io | 10
+# ext4_es_lookup_extent_enter | 9
+# sys_enter_ioctl | 9
+# sys_exit_ioctl | 9
+# ext4_ext_map_blocks_enter | 9
+# ext4_ext_map_blocks_exit | 9
+# ext4_es_lookup_extent_exit | 9
+# ext4_es_insert_extent | 9
+# ext4_ext_show_extent | 8
+# extract_entropy | 8
+#ext4_es_find_delayed_extent_exit | 8
+# ext4_es_find_delayed_extent_... | 8
+# writeback_pages_written | 7
+# sys_exit_dup2 | 7
+# sys_enter_dup2 | 7
+# signal_generate | 7
+# sys_enter_fcntl64 | 7
+# sys_exit_fcntl64 | 7
+# global_dirty_state | 7
+# writeback_dirty_inode_start | 7
+# block_bio_backmerge | 7
+# writeback_dirty_inode | 7
+# sched_wakeup_new | 6
+# sched_process_free | 6
+# sys_enter_exit_group | 6
+# task_newtask | 6
+# sys_enter_clone | 6
+# sched_process_fork | 6
+# sched_process_exit | 6
+# sys_exit_gettimeofday | 5
+# signal_deliver | 5
+# sys_enter_gettimeofday | 5
+# writeback_single_inode | 4
+# sys_enter_execve | 4
+# task_rename | 4
+# sched_process_exec | 4
+# block_dirty_buffer | 4
+# sys_exit_execve | 4
+# block_unplug | 4
+# sched_stat_iowait | 4
+# writeback_single_inode_start | 4
+# block_plug | 4
+# writeback_write_inode | 3
+# sys_enter_pipe | 3
+# writeback_dirty_page | 3
+# writeback_write_inode_start | 3
+# ext4_mark_inode_dirty | 3
+# ext4_journal_start | 3
+# sys_exit_pipe | 3
+# jbd2_drop_transaction | 2
+# jbd2_commit_locking | 2
+# jbd2_commit_flushing | 2
+# jbd2_handle_start | 2
+# jbd2_run_stats | 2
+# sys_exit_getdents | 2
+# jbd2_checkpoint_stats | 2
+# sys_enter_getgroups | 2
+# jbd2_start_commit | 2
+# jbd2_end_commit | 2
+# ext4_da_writepages | 2
+# jbd2_handle_stats | 2
+# sys_enter_statfs64 | 2
+# sys_exit_statfs64 | 2
+# sys_exit_getgroups | 2
+# sys_exit_lseek | 2
+# sys_enter_lseek | 2
+# sys_enter_getdents | 2
+# ext4_da_write_pages | 2
+# jbd2_commit_logging | 2
+# ext4_request_blocks | 1
+# sys_exit_openat | 1
+# ext4_discard_preallocations | 1
+# ext4_mballoc_alloc | 1
+# sys_enter_openat | 1
+# ext4_da_writepages_result | 1
+# ext4_allocate_blocks | 1
+# sys_enter_newuname | 1
+# ext4_da_update_reserve_space | 1
+# ext4_get_reserved_cluster_alloc | 1
+# sys_exit_newuname | 1
+# writeback_wake_thread | 1
+
diff --git a/tools/ktap/samples/tracepoints/eventcount_by_proc.kp b/tools/ktap/samples/tracepoints/eventcount_by_proc.kp
new file mode 100644
index 0000000..1b95f19
--- /dev/null
+++ b/tools/ktap/samples/tracepoints/eventcount_by_proc.kp
@@ -0,0 +1,57 @@
+#!/usr/bin/env ktap
+
+# showing all tracepoints in histogram style
+
+var s = {}
+
+trace *:* {
+ s[execname] += 1
+}
+
+trace_end {
+ print_hist(s)
+}
+
+#Results:
+#^C
+# value ------------- Distribution ------------- count
+# swapper/0 |@@@@@@@@@@@@ 354378
+# swapper/1 |@@@@@@@@@@ 284984
+# ps |@@@@ 115697
+# ksmtuned |@@@ 95857
+# iscsid |@@ 80008
+# awk |@ 30354
+# irqbalance | 16530
+# rcu_sched | 15892
+# sendmail | 14463
+# kworker/0:1 | 10540
+# kworker/u4:2 | 9250
+# kworker/1:2 | 7943
+# sleep | 7555
+# crond | 3911
+# ksoftirqd/0 | 3817
+# sshd | 2849
+# systemd-journal | 2209
+# migration/1 | 1601
+# migration/0 | 1350
+# dhclient | 1343
+# nm-dhcp-client. | 1208
+# ksoftirqd/1 | 1064
+# watchdog/1 | 966
+# watchdog/0 | 964
+# khugepaged | 776
+# dbus-daemon | 611
+# rpcbind | 607
+# gdbus | 529
+# NetworkManager | 399
+# jbd2/dm-1-8 | 378
+# modem-manager | 184
+# abrt-watch-log | 157
+# polkitd | 156
+# rs:main Q:Reg | 153
+# avahi-daemon | 151
+# rsyslogd | 102
+# systemd | 96
+# kworker/0:1H | 45
+# smartd | 30
+
diff --git a/tools/ktap/samples/tracepoints/raw_tracepoint.kp b/tools/ktap/samples/tracepoints/raw_tracepoint.kp
new file mode 100644
index 0000000..cffda45
--- /dev/null
+++ b/tools/ktap/samples/tracepoints/raw_tracepoint.kp
@@ -0,0 +1,15 @@
+#!/usr/bin/env ktap
+
+#This script use kdebug.tracepoint, not 'trace' keyword which use perf backend.
+#
+#The overhead of kdebug.tracepoint would be much little than normal perf
+#backend tracing.
+
+kdebug.tracepoint("sys_enter_open", function () {
+ printf("sys_enter_open: (%s) open file (%s)\n",
+ execname, user_string(arg1))
+})
+
+kdebug.tracepoint("sys_exit_open", function () {
+ printf("sys_exit_open: return fd: (%d)\n", arg1)
+})
diff --git a/tools/ktap/samples/tracepoints/tracepoints.kp b/tools/ktap/samples/tracepoints/tracepoints.kp
new file mode 100644
index 0000000..3ff29b5
--- /dev/null
+++ b/tools/ktap/samples/tracepoints/tracepoints.kp
@@ -0,0 +1,6 @@
+#!/usr/bin/env ktap
+
+trace *:* {
+ print(cpu, pid, execname, argstr)
+}
+
diff --git a/tools/ktap/samples/userspace/gcc_unwind.kp b/tools/ktap/samples/userspace/gcc_unwind.kp
new file mode 100644
index 0000000..48db46e
--- /dev/null
+++ b/tools/ktap/samples/userspace/gcc_unwind.kp
@@ -0,0 +1,9 @@
+#!/usr/bin/env ktap
+
+#only tested in x86-64 system,
+#if you run this script in x86_32, change the libc path.
+
+trace sdt:/lib/x86_64-linux-gnu/libgcc_s.so.1:unwind {
+ print(execname, argstr)
+}
+
diff --git a/tools/ktap/samples/userspace/glibc_func_hist.kp b/tools/ktap/samples/userspace/glibc_func_hist.kp
new file mode 100644
index 0000000..bb3dd9f
--- /dev/null
+++ b/tools/ktap/samples/userspace/glibc_func_hist.kp
@@ -0,0 +1,44 @@
+#!/usr/bin/env ktap
+
+#This ktap script trace all glibc functions in histogram output
+
+#only tested in x86-64 system,
+#if you run this script in x86_32, change the libc path.
+
+var s = {}
+
+trace probe:/lib64/libc.so.6:* {
+ s[probename] += 1
+}
+
+trace_end {
+ print_hist(s)
+}
+
+# Example result:
+#[root@localhost ktap]# ./ktap ./glibc_func_hist.kp
+#Tracing... Ctrl-C to end.
+#^C
+# value ------------- Distribution ------------- count
+# _IO_sputbackc | 1536
+# __strncmp_sse2 | 1522
+# __GI_strncmp | 1522
+# __GI_memcpy | 1446
+# __memcpy_sse2 | 1446
+# _dl_mcount_wrapper_check | 1433
+# __GI__dl_mcount_wrapper_check | 1433
+# __gconv_transform_utf8_internal | 1429
+# __mbrtowc | 1425
+# mbrtoc32 | 1425
+# __GI___mbrtowc | 1425
+# mbrtowc | 1425
+# __GI_mbrtowc | 1425
+# strtouq | 1274
+# strtoull | 1274
+# strtoul | 1274
+# __ctype_get_mb_cur_max | 984
+# ____strtoull_l_internal | 970
+# __GI_____strtoul_l_internal | 970
+# __GI__IO_sputbackc | 960
+# ... |
+
diff --git a/tools/ktap/samples/userspace/glibc_sdt.kp b/tools/ktap/samples/userspace/glibc_sdt.kp
new file mode 100644
index 0000000..e396901
--- /dev/null
+++ b/tools/ktap/samples/userspace/glibc_sdt.kp
@@ -0,0 +1,11 @@
+#!/usr/bin/env ktap
+
+#This ktap script trace all sdt notes in glibc
+
+#only tested in x86-64 system,
+#if you run this script in x86_32, change the libc path.
+
+trace sdt:/lib64/libc.so.6:* {
+ print(execname, argstr)
+}
+
diff --git a/tools/ktap/samples/userspace/glibc_trace.kp b/tools/ktap/samples/userspace/glibc_trace.kp
new file mode 100644
index 0000000..9b8d16e
--- /dev/null
+++ b/tools/ktap/samples/userspace/glibc_trace.kp
@@ -0,0 +1,11 @@
+#!/usr/bin/env ktap
+
+#This ktap script trace all functions in glibc
+
+#only tested in x86-64 system,
+#if you run this script in x86_32, change the libc path.
+
+trace probe:/lib64/libc.so.6:* {
+ print(execname, argstr)
+}
+
diff --git a/tools/ktap/samples/userspace/malloc_free.kp b/tools/ktap/samples/userspace/malloc_free.kp
new file mode 100644
index 0000000..884d99a
--- /dev/null
+++ b/tools/ktap/samples/userspace/malloc_free.kp
@@ -0,0 +1,20 @@
+#!/usr/bin/env ktap
+
+#only tested in x86-64 system,
+#if you run this script in x86_32, change the libc path.
+
+trace probe:/lib64/libc.so.6:malloc {
+ print("malloc entry:", execname)
+}
+
+trace probe:/lib64/libc.so.6:malloc%return {
+ print("malloc exit:", execname)
+}
+
+trace probe:/lib64/libc.so.6:free {
+ print("free entry:", execname)
+}
+
+trace probe:/lib64/libc.so.6:free%return {
+ print("free exit:", execname)
+}
diff --git a/tools/ktap/samples/userspace/malloc_size_hist.kp b/tools/ktap/samples/userspace/malloc_size_hist.kp
new file mode 100644
index 0000000..41bd202
--- /dev/null
+++ b/tools/ktap/samples/userspace/malloc_size_hist.kp
@@ -0,0 +1,22 @@
+#!/usr/bin/env ktap
+
+# Aggregate system or process malloc size
+
+# only tested in x86-64 system,
+# if you run this script in x86_32, change the libc path and register name.
+#
+# Examples:
+#
+# ktap malloc_size_hist.kp
+# ktap malloc_size_hist.kp -- ls
+
+var s = {}
+
+trace probe:/lib64/libc.so.6:malloc size=%di {
+ #arg2 is argument "size" of malloc function
+ s[arg1] += 1
+}
+
+trace_end {
+ print_hist(s)
+}
diff --git a/tools/ktap/samples/userspace/pthread.kp b/tools/ktap/samples/userspace/pthread.kp
new file mode 100644
index 0000000..be693ac
--- /dev/null
+++ b/tools/ktap/samples/userspace/pthread.kp
@@ -0,0 +1,8 @@
+#!/usr/bin/env ktap
+
+# This script trace pthread_mutex* related call in libpthread
+# Tested in x86_64
+
+trace probe:/lib64/libpthread-2.17.so:pthread_mutex_* {
+ print(execname, argstr)
+}
--
1.8.1.4
next prev parent reply other threads:[~2014-03-28 13:52 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-28 13:47 [RFC PATCH 00/28] ktap: A lightweight dynamic tracing tool for Linux Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 01/28] ktap: add README file Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 02/28] ktap: add ktap tutorial Jovi Zhangwei
2014-03-28 13:47 ` Jovi Zhangwei [this message]
2014-03-28 13:47 ` [PATCH 04/28] ktap: add basic ktap types definition Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 05/28] ktap: add bytecode definition Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 06/28] ktap: add include/ktap_arch.h and error header file Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 07/28] ktap: add runtime/ktap.[c|h] Jovi Zhangwei
2014-03-28 18:38 ` Andi Kleen
2014-03-29 7:32 ` Jovi Zhangwei
2014-03-29 17:04 ` Greg Kroah-Hartman
2014-03-30 7:26 ` Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 08/28] ktap: add runtime/kp_bcread.[c|h] Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 09/28] ktap: add runtime/kp_vm.[c|h] Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 10/28] ktap: add runtime/kp_str.[c|h] and runtime/kp_mempool.[c|h] Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 11/28] ktap: add runtime/kp_tab.[c|h] Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 12/28] ktap: add runtime/kp_obj.[c|h] Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 13/28] ktap: add runtime/kp_transport.[c|h] Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 14/28] ktap: add runtime/kp_events.[c|h] Jovi Zhangwei
2014-03-31 9:10 ` Masami Hiramatsu
2014-03-31 10:14 ` Jovi Zhangwei
2014-04-01 6:59 ` Masami Hiramatsu
2014-04-01 7:28 ` Jovi Zhangwei
2014-04-01 8:05 ` Masami Hiramatsu
2014-03-28 13:47 ` [PATCH 15/28] ktap: add built-in functions and library (runtime/lib_*.c) Jovi Zhangwei
2014-03-28 18:51 ` Andi Kleen
2014-03-29 4:15 ` Jovi Zhangwei
2014-03-30 0:58 ` Andi Kleen
2014-03-31 2:01 ` Jovi Zhangwei
2014-03-31 13:13 ` Andi Kleen
2014-04-02 1:44 ` Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 16/28] ktap: add runtime/amalg.c Jovi Zhangwei
2014-03-28 18:52 ` Andi Kleen
2014-03-29 7:38 ` Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 17/28] ktap: add userspace/kp_main.c Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 18/28] ktap: add compiler(userspace/kp_lex.[c|h] and userspace/kp_parse.[c|h]) Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 19/28] ktap: add userspace/symbol.[c|h] Jovi Zhangwei
2014-04-01 7:28 ` Masami Hiramatsu
2014-03-28 13:47 ` [PATCH 20/28] ktap: add userspace/kp_parse_events.c Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 21/28] ktap: add userspace/kp_reader.c Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 22/28] ktap: add userspace/kp_bcwrite.c Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 23/28] ktap: add userspace/kp_util.c Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 24/28] ktap: add Makefile Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 25/28] ktap: add Kconfig Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 26/28] ktap: add testsuite Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 27/28] ktap: add vim syntax file Jovi Zhangwei
2014-03-28 13:47 ` [PATCH 28/28] ktap: add COPYRIGHT file Jovi Zhangwei
2014-03-28 16:08 ` [RFC PATCH 00/28] ktap: A lightweight dynamic tracing tool for Linux Greg Kroah-Hartman
2014-03-29 1:46 ` Jovi Zhangwei
2014-03-31 7:17 ` Ingo Molnar
2014-03-31 10:01 ` Jovi Zhangwei
2014-03-31 21:29 ` Alexei Starovoitov
2014-04-01 4:47 ` Jovi Zhangwei
2014-04-02 4:57 ` Alexei Starovoitov
2014-04-02 6:37 ` Jovi Zhangwei
2014-04-02 7:43 ` Ingo Molnar
2014-04-02 8:49 ` Jovi Zhangwei
2014-04-04 7:36 ` Ingo Molnar
2014-04-08 6:50 ` Jovi Zhangwei
2014-04-14 15:11 ` Ingo Molnar
2014-04-14 15:28 ` Daniel Borkmann
2014-04-02 7:42 ` Ingo Molnar
2014-04-07 13:55 ` Peter Zijlstra
2014-04-08 7:40 ` Masami Hiramatsu
2014-04-08 9:08 ` Peter Zijlstra
2014-04-02 7:36 ` Ingo Molnar
2014-03-31 20:06 ` Daniel Borkmann
2014-03-31 9:18 ` Masami Hiramatsu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1396014469-5937-4-git-send-email-jovi.zhangwei@gmail.com \
--to=jovi.zhangwei@gmail.com \
--cc=fweisbec@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=masami.hiramatsu.pt@hitachi.com \
--cc=mingo@redhat.org \
--cc=rostedt@goodmis.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.