1 Linux Magic System Request Key Hacks
2 Documentation for sysrq.c
4 * What is the magic SysRq key?
5 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
6 It is a 'magical' key combo you can hit which the kernel will respond to
7 regardless of whatever else it is doing, unless it is completely locked up.
9 * How do I enable the magic SysRq key?
10 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
11 You need to say "yes" to 'Magic SysRq key (CONFIG_MAGIC_SYSRQ)' when
12 configuring the kernel. When running a kernel with SysRq compiled in,
13 /proc/sys/kernel/sysrq controls the functions allowed to be invoked via
14 the SysRq key. The default value in this file is set by the
15 CONFIG_MAGIC_SYSRQ_DEFAULT_ENABLE config symbol, which itself defaults
16 to 1. Here is the list of possible values in /proc/sys/kernel/sysrq:
17 0 - disable sysrq completely
18 1 - enable all functions of sysrq
19 >1 - bitmask of allowed sysrq functions (see below for detailed function
21 2 = 0x2 - enable control of console logging level
22 4 = 0x4 - enable control of keyboard (SAK, unraw)
23 8 = 0x8 - enable debugging dumps of processes etc.
24 16 = 0x10 - enable sync command
25 32 = 0x20 - enable remount read-only
26 64 = 0x40 - enable signalling of processes (term, kill, oom-kill)
27 128 = 0x80 - allow reboot/poweroff
28 256 = 0x100 - allow nicing of all RT tasks
30 You can set the value in the file by the following command:
31 echo "number" >/proc/sys/kernel/sysrq
33 The number may be written here either as decimal or as hexadecimal
34 with the 0x prefix. CONFIG_MAGIC_SYSRQ_DEFAULT_ENABLE must always be
35 written in hexadecimal.
37 Note that the value of /proc/sys/kernel/sysrq influences only the invocation
38 via a keyboard. Invocation of any operation via /proc/sysrq-trigger is always
39 allowed (by a user with admin privileges).
41 * How do I use the magic SysRq key?
42 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
43 On x86 - You press the key combo 'ALT-SysRq-<command key>'. Note - Some
44 keyboards may not have a key labeled 'SysRq'. The 'SysRq' key is
45 also known as the 'Print Screen' key. Also some keyboards cannot
46 handle so many keys being pressed at the same time, so you might
47 have better luck with "press Alt", "press SysRq", "release SysRq",
48 "press <command key>", release everything.
50 On SPARC - You press 'ALT-STOP-<command key>', I believe.
52 On the serial console (PC style standard serial ports only) -
53 You send a BREAK, then within 5 seconds a command key. Sending
54 BREAK twice is interpreted as a normal BREAK.
56 On PowerPC - Press 'ALT - Print Screen (or F13) - <command key>,
57 Print Screen (or F13) - <command key> may suffice.
59 On other - If you know of the key combos for other architectures, please
60 let me know so I can add them to this section.
62 On all - write a character to /proc/sysrq-trigger, e.g.:
63 echo t > /proc/sysrq-trigger
65 On all - Enable network SysRq by writing a cookie to icmp_echo_sysrq, e.g.
66 echo 0x01020304 >/proc/sys/net/ipv4/icmp_echo_sysrq
67 Send an ICMP echo request with this pattern plus the particular
68 SysRq command key. Example:
69 # ping -c1 -s57 -p0102030468
70 will trigger the SysRq-H (help) command.
73 * What are the 'command' keys?
74 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
75 'b' - Will immediately reboot the system without syncing or unmounting
78 'c' - Will perform a system crash by a NULL pointer dereference.
79 A crashdump will be taken if configured.
81 'd' - Shows all locks that are held.
83 'e' - Send a SIGTERM to all processes, except for init.
85 'f' - Will call oom_kill to kill a memory hog process.
87 'g' - Used by kgdb (kernel debugger)
89 'h' - Will display help (actually any other key than those listed
90 here will display help. but 'h' is easy to remember :-)
92 'i' - Send a SIGKILL to all processes, except for init.
94 'j' - Forcibly "Just thaw it" - filesystems frozen by the FIFREEZE ioctl.
96 'k' - Secure Access Key (SAK) Kills all programs on the current virtual
97 console. NOTE: See important comments below in SAK section.
99 'l' - Shows a stack backtrace for all active CPUs.
101 'm' - Will dump current memory info to your console.
103 'n' - Used to make RT tasks nice-able
105 'o' - Will shut your system off (if configured and supported).
107 'p' - Will dump the current registers and flags to your console.
109 'q' - Will dump per CPU lists of all armed hrtimers (but NOT regular
110 timer_list timers) and detailed information about all
113 'r' - Turns off keyboard raw mode and sets it to XLATE.
115 's' - Will attempt to sync all mounted filesystems.
117 't' - Will dump a list of current tasks and their information to your
120 'u' - Will attempt to remount all mounted filesystems read-only.
122 'v' - Forcefully restores framebuffer console
123 'v' - Causes ETM buffer dump [ARM-specific]
125 'w' - Dumps tasks that are in uninterruptable (blocked) state.
127 'x' - Used by xmon interface on ppc/powerpc platforms.
128 Show global PMU Registers on sparc64.
130 'y' - Show global CPU Registers [SPARC-64 specific]
132 'z' - Dump the ftrace buffer
134 '0'-'9' - Sets the console log level, controlling which kernel messages
135 will be printed to your console. ('0', for example would make
136 it so that only emergency messages like PANICs or OOPSes would
137 make it to your console.)
139 * Okay, so what can I use them for?
140 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
141 Well, unraw(r) is very handy when your X server or a svgalib program crashes.
143 sak(k) (Secure Access Key) is useful when you want to be sure there is no
144 trojan program running at console which could grab your password
145 when you would try to login. It will kill all programs on given console,
146 thus letting you make sure that the login prompt you see is actually
147 the one from init, not some trojan program.
148 IMPORTANT: In its true form it is not a true SAK like the one in a :IMPORTANT
149 IMPORTANT: c2 compliant system, and it should not be mistaken as :IMPORTANT
150 IMPORTANT: such. :IMPORTANT
151 It seems others find it useful as (System Attention Key) which is
152 useful when you want to exit a program that will not let you switch consoles.
153 (For example, X or a svgalib program.)
155 reboot(b) is good when you're unable to shut down. But you should also
156 sync(s) and umount(u) first.
158 crash(c) can be used to manually trigger a crashdump when the system is hung.
159 Note that this just triggers a crash if there is no dump mechanism available.
161 sync(s) is great when your system is locked up, it allows you to sync your
162 disks and will certainly lessen the chance of data loss and fscking. Note
163 that the sync hasn't taken place until you see the "OK" and "Done" appear
164 on the screen. (If the kernel is really in strife, you may not ever get the
165 OK or Done message...)
167 umount(u) is basically useful in the same ways as sync(s). I generally sync(s),
168 umount(u), then reboot(b) when my system locks. It's saved me many a fsck.
169 Again, the unmount (remount read-only) hasn't taken place until you see the
170 "OK" and "Done" message appear on the screen.
172 The loglevels '0'-'9' are useful when your console is being flooded with
173 kernel messages you do not want to see. Selecting '0' will prevent all but
174 the most urgent kernel messages from reaching your console. (They will
175 still be logged if syslogd/klogd are alive, though.)
177 term(e) and kill(i) are useful if you have some sort of runaway process you
178 are unable to kill any other way, especially if it's spawning other
181 "just thaw it(j)" is useful if your system becomes unresponsive due to a frozen
182 (probably root) filesystem via the FIFREEZE ioctl.
184 * Sometimes SysRq seems to get 'stuck' after using it, what can I do?
185 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
186 That happens to me, also. I've found that tapping shift, alt, and control
187 on both sides of the keyboard, and hitting an invalid sysrq sequence again
188 will fix the problem. (i.e., something like alt-sysrq-z). Switching to another
189 virtual console (ALT+Fn) and then back again should also help.
191 * I hit SysRq, but nothing seems to happen, what's wrong?
192 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
193 There are some keyboards that produce a different keycode for SysRq than the
194 pre-defined value of 99 (see KEY_SYSRQ in include/linux/input.h), or which
195 don't have a SysRq key at all. In these cases, run 'showkey -s' to find an
196 appropriate scancode sequence, and use 'setkeycodes <sequence> 99' to map
197 this sequence to the usual SysRq code (e.g., 'setkeycodes e05b 99'). It's
198 probably best to put this command in a boot script. Oh, and by the way, you
199 exit 'showkey' by not typing anything for ten seconds.
201 * I want to add SysRQ key events to a module, how does it work?
202 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
203 In order to register a basic function with the table, you must first include
204 the header 'include/linux/sysrq.h', this will define everything else you need.
205 Next, you must create a sysrq_key_op struct, and populate it with A) the key
206 handler function you will use, B) a help_msg string, that will print when SysRQ
207 prints help, and C) an action_msg string, that will print right before your
208 handler is called. Your handler must conform to the prototype in 'sysrq.h'.
210 After the sysrq_key_op is created, you can call the kernel function
211 register_sysrq_key(int key, struct sysrq_key_op *op_p); this will
212 register the operation pointed to by 'op_p' at table key 'key',
213 if that slot in the table is blank. At module unload time, you must call
214 the function unregister_sysrq_key(int key, struct sysrq_key_op *op_p), which
215 will remove the key op pointed to by 'op_p' from the key 'key', if and only if
216 it is currently registered in that slot. This is in case the slot has been
217 overwritten since you registered it.
219 The Magic SysRQ system works by registering key operations against a key op
220 lookup table, which is defined in 'drivers/char/sysrq.c'. This key table has
221 a number of operations registered into it at compile time, but is mutable,
222 and 2 functions are exported for interface to it:
223 register_sysrq_key and unregister_sysrq_key.
224 Of course, never ever leave an invalid pointer in the table. I.e., when
225 your module that called register_sysrq_key() exits, it must call
226 unregister_sysrq_key() to clean up the sysrq key table entry that it used.
227 Null pointers in the table are always safe. :)
229 If for some reason you feel the need to call the handle_sysrq function from
230 within a function called by handle_sysrq, you must be aware that you are in
231 a lock (you are also in an interrupt handler, which means don't sleep!), so
232 you must call __handle_sysrq_nolock instead.
234 * When I hit a SysRq key combination only the header appears on the console?
235 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
236 Sysrq output is subject to the same console loglevel control as all
237 other console output. This means that if the kernel was booted 'quiet'
238 as is common on distro kernels the output may not appear on the actual
239 console, even though it will appear in the dmesg buffer, and be accessible
240 via the dmesg command and to the consumers of /proc/kmsg. As a specific
241 exception the header line from the sysrq command is passed to all console
242 consumers as if the current loglevel was maximum. If only the header
243 is emitted it is almost certain that the kernel loglevel is too low.
244 Should you require the output on the console channel then you will need
245 to temporarily up the console loglevel using alt-sysrq-8 or:
247 echo 8 > /proc/sysrq-trigger
249 Remember to return the loglevel to normal after triggering the sysrq
250 command you are interested in.
252 * I have more questions, who can I ask?
253 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
254 Just ask them on the linux-kernel mailing list:
255 linux-kernel@vger.kernel.org
258 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
259 Written by Mydraal <vulpyne@vulpyne.net>
260 Updated by Adam Sulmicki <adam@cfar.umd.edu>
261 Updated by Jeremy M. Dolan <jmd@turbogeek.org> 2001/01/28 10:15:59
262 Added to by Crutcher Dunnavant <crutcher+kernel@datastacks.com>