Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
Documentation/sysrq.txt: convert to ReST markup
- Fix document title;
- use a table for the valid commands;
- use quote blocks where needed;
- use monotonic fonts for config options and file names;
- adjust whitespaces and blank lines;
- add it to the user's book.
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
+149
-117
+149
-117
Documentation/sysrq.txt
+149
-117
Documentation/sysrq.txt
···
1
1
Linux Magic System Request Key Hacks
2
+
====================================
3
+
2
4
Documentation for sysrq.c
3
5
4
-
* What is the magic SysRq key?
5
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
6
+
What is the magic SysRq key?
7
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
8
+
6
9
It is a 'magical' key combo you can hit which the kernel will respond to
7
10
regardless of whatever else it is doing, unless it is completely locked up.
8
11
9
-
* How do I enable the magic SysRq key?
10
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
12
+
How do I enable the magic SysRq key?
13
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
14
+
11
15
You need to say "yes" to 'Magic SysRq key (CONFIG_MAGIC_SYSRQ)' when
12
16
configuring the kernel. When running a kernel with SysRq compiled in,
13
17
/proc/sys/kernel/sysrq controls the functions allowed to be invoked via
14
18
the SysRq key. The default value in this file is set by the
15
19
CONFIG_MAGIC_SYSRQ_DEFAULT_ENABLE config symbol, which itself defaults
16
20
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
20
-
description):
21
+
22
+
- 0 - disable sysrq completely
23
+
- 1 - enable all functions of sysrq
24
+
- >1 - bitmask of allowed sysrq functions (see below for detailed function
25
+
description)::
26
+
21
27
2 = 0x2 - enable control of console logging level
22
28
4 = 0x4 - enable control of keyboard (SAK, unraw)
23
29
8 = 0x8 - enable debugging dumps of processes etc.
···
33
27
128 = 0x80 - allow reboot/poweroff
34
28
256 = 0x100 - allow nicing of all RT tasks
35
29
36
-
You can set the value in the file by the following command:
30
+
You can set the value in the file by the following command::
31
+
37
32
echo "number" >/proc/sys/kernel/sysrq
38
33
39
34
The number may be written here either as decimal or as hexadecimal
40
35
with the 0x prefix. CONFIG_MAGIC_SYSRQ_DEFAULT_ENABLE must always be
41
36
written in hexadecimal.
42
37
43
-
Note that the value of /proc/sys/kernel/sysrq influences only the invocation
44
-
via a keyboard. Invocation of any operation via /proc/sysrq-trigger is always
45
-
allowed (by a user with admin privileges).
38
+
Note that the value of ``/proc/sys/kernel/sysrq`` influences only the invocation
39
+
via a keyboard. Invocation of any operation via ``/proc/sysrq-trigger`` is
40
+
always allowed (by a user with admin privileges).
46
41
47
-
* How do I use the magic SysRq key?
48
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
49
-
On x86 - You press the key combo 'ALT-SysRq-<command key>'. Note - Some
42
+
How do I use the magic SysRq key?
43
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
44
+
45
+
On x86 - You press the key combo :kbd:`ALT-SysRq-<command key>`.
46
+
47
+
.. note::
48
+
Some
50
49
keyboards may not have a key labeled 'SysRq'. The 'SysRq' key is
51
50
also known as the 'Print Screen' key. Also some keyboards cannot
52
51
handle so many keys being pressed at the same time, so you might
53
-
have better luck with "press Alt", "press SysRq", "release SysRq",
54
-
"press <command key>", release everything.
52
+
have better luck with press :kbd:`Alt`, press :kbd:`SysRq`,
53
+
release :kbd:`SysRq`, press :kbd:`<command key>`, release everything.
55
54
56
-
On SPARC - You press 'ALT-STOP-<command key>', I believe.
55
+
On SPARC - You press :kbd:`ALT-STOP-<command key>`, I believe.
57
56
58
-
On the serial console (PC style standard serial ports only) -
59
-
You send a BREAK, then within 5 seconds a command key. Sending
60
-
BREAK twice is interpreted as a normal BREAK.
57
+
On the serial console (PC style standard serial ports only)
58
+
You send a ``BREAK``, then within 5 seconds a command key. Sending
59
+
``BREAK`` twice is interpreted as a normal BREAK.
61
60
62
-
On PowerPC - Press 'ALT - Print Screen (or F13) - <command key>,
63
-
Print Screen (or F13) - <command key> may suffice.
61
+
On PowerPC
62
+
Press :kbd:`ALT - Print Screen` (or :kbd:`F13`) - :kbd:`<command key>`,
63
+
:kbd:`Print Screen` (or :kbd:`F13`) - :kbd:`<command key>` may suffice.
64
64
65
-
On other - If you know of the key combos for other architectures, please
66
-
let me know so I can add them to this section.
65
+
On other
66
+
If you know of the key combos for other architectures, please
67
+
let me know so I can add them to this section.
67
68
68
-
On all - write a character to /proc/sysrq-trigger. e.g.:
69
+
On all
70
+
write a character to /proc/sysrq-trigger. e.g.::
69
71
70
72
echo t > /proc/sysrq-trigger
71
73
72
-
* What are the 'command' keys?
73
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
74
-
'b' - Will immediately reboot the system without syncing or unmounting
75
-
your disks.
74
+
What are the 'command' keys?
75
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
76
76
77
-
'c' - Will perform a system crash by a NULL pointer dereference.
78
-
A crashdump will be taken if configured.
77
+
=========== ===================================================================
78
+
Command Function
79
+
=========== ===================================================================
80
+
``b`` Will immediately reboot the system without syncing or unmounting
81
+
your disks.
79
82
80
-
'd' - Shows all locks that are held.
83
+
``c`` Will perform a system crash by a NULL pointer dereference.
84
+
A crashdump will be taken if configured.
81
85
82
-
'e' - Send a SIGTERM to all processes, except for init.
86
+
``d`` Shows all locks that are held.
83
87
84
-
'f' - Will call the oom killer to kill a memory hog process, but do not
85
-
panic if nothing can be killed.
88
+
``e`` Send a SIGTERM to all processes, except for init.
86
89
87
-
'g' - Used by kgdb (kernel debugger)
90
+
``f`` Will call the oom killer to kill a memory hog process, but do not
91
+
panic if nothing can be killed.
88
92
89
-
'h' - Will display help (actually any other key than those listed
90
-
here will display help. but 'h' is easy to remember :-)
93
+
``g`` Used by kgdb (kernel debugger)
91
94
92
-
'i' - Send a SIGKILL to all processes, except for init.
95
+
``h`` Will display help (actually any other key than those listed
96
+
here will display help. but ``h`` is easy to remember :-)
93
97
94
-
'j' - Forcibly "Just thaw it" - filesystems frozen by the FIFREEZE ioctl.
98
+
``i`` Send a SIGKILL to all processes, except for init.
95
99
96
-
'k' - Secure Access Key (SAK) Kills all programs on the current virtual
97
-
console. NOTE: See important comments below in SAK section.
100
+
``j`` Forcibly "Just thaw it" - filesystems frozen by the FIFREEZE ioctl.
98
101
99
-
'l' - Shows a stack backtrace for all active CPUs.
102
+
``k`` Secure Access Key (SAK) Kills all programs on the current virtual
103
+
console. NOTE: See important comments below in SAK section.
100
104
101
-
'm' - Will dump current memory info to your console.
105
+
``l`` Shows a stack backtrace for all active CPUs.
102
106
103
-
'n' - Used to make RT tasks nice-able
107
+
``m`` Will dump current memory info to your console.
104
108
105
-
'o' - Will shut your system off (if configured and supported).
109
+
``n`` Used to make RT tasks nice-able
106
110
107
-
'p' - Will dump the current registers and flags to your console.
111
+
``o`` Will shut your system off (if configured and supported).
108
112
109
-
'q' - Will dump per CPU lists of all armed hrtimers (but NOT regular
110
-
timer_list timers) and detailed information about all
111
-
clockevent devices.
113
+
``p`` Will dump the current registers and flags to your console.
112
114
113
-
'r' - Turns off keyboard raw mode and sets it to XLATE.
115
+
``q`` Will dump per CPU lists of all armed hrtimers (but NOT regular
116
+
timer_list timers) and detailed information about all
117
+
clockevent devices.
114
118
115
-
's' - Will attempt to sync all mounted filesystems.
119
+
``r`` Turns off keyboard raw mode and sets it to XLATE.
116
120
117
-
't' - Will dump a list of current tasks and their information to your
118
-
console.
121
+
``s`` Will attempt to sync all mounted filesystems.
119
122
120
-
'u' - Will attempt to remount all mounted filesystems read-only.
123
+
``t`` Will dump a list of current tasks and their information to your
124
+
console.
121
125
122
-
'v' - Forcefully restores framebuffer console
123
-
'v' - Causes ETM buffer dump [ARM-specific]
126
+
``u`` Will attempt to remount all mounted filesystems read-only.
124
127
125
-
'w' - Dumps tasks that are in uninterruptable (blocked) state.
128
+
``v`` Forcefully restores framebuffer console
129
+
``v`` Causes ETM buffer dump [ARM-specific]
126
130
127
-
'x' - Used by xmon interface on ppc/powerpc platforms.
128
-
Show global PMU Registers on sparc64.
129
-
Dump all TLB entries on MIPS.
131
+
``w`` Dumps tasks that are in uninterruptable (blocked) state.
130
132
131
-
'y' - Show global CPU Registers [SPARC-64 specific]
133
+
``x`` Used by xmon interface on ppc/powerpc platforms.
134
+
Show global PMU Registers on sparc64.
135
+
Dump all TLB entries on MIPS.
132
136
133
-
'z' - Dump the ftrace buffer
137
+
``y`` Show global CPU Registers [SPARC-64 specific]
134
138
135
-
'0'-'9' - Sets the console log level, controlling which kernel messages
136
-
will be printed to your console. ('0', for example would make
137
-
it so that only emergency messages like PANICs or OOPSes would
138
-
make it to your console.)
139
+
``z`` Dump the ftrace buffer
139
140
140
-
* Okay, so what can I use them for?
141
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
141
+
``0``-``9`` Sets the console log level, controlling which kernel messages
142
+
will be printed to your console. (``0``, for example would make
143
+
it so that only emergency messages like PANICs or OOPSes would
144
+
make it to your console.)
145
+
=========== ===================================================================
146
+
147
+
Okay, so what can I use them for?
148
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
149
+
142
150
Well, unraw(r) is very handy when your X server or a svgalib program crashes.
143
151
144
152
sak(k) (Secure Access Key) is useful when you want to be sure there is no
···
160
140
when you would try to login. It will kill all programs on given console,
161
141
thus letting you make sure that the login prompt you see is actually
162
142
the one from init, not some trojan program.
163
-
IMPORTANT: In its true form it is not a true SAK like the one in a :IMPORTANT
164
-
IMPORTANT: c2 compliant system, and it should not be mistaken as :IMPORTANT
165
-
IMPORTANT: such. :IMPORTANT
166
-
It seems others find it useful as (System Attention Key) which is
143
+
144
+
.. important::
145
+
146
+
In its true form it is not a true SAK like the one in a
147
+
c2 compliant system, and it should not be mistaken as
148
+
such.
149
+
150
+
It seems others find it useful as (System Attention Key) which is
167
151
useful when you want to exit a program that will not let you switch consoles.
168
152
(For example, X or a svgalib program.)
169
153
170
-
reboot(b) is good when you're unable to shut down. But you should also
171
-
sync(s) and umount(u) first.
154
+
``reboot(b)`` is good when you're unable to shut down. But you should also
155
+
``sync(s)`` and ``umount(u)`` first.
172
156
173
-
crash(c) can be used to manually trigger a crashdump when the system is hung.
157
+
``crash(c)`` can be used to manually trigger a crashdump when the system is hung.
174
158
Note that this just triggers a crash if there is no dump mechanism available.
175
159
176
-
sync(s) is great when your system is locked up, it allows you to sync your
160
+
``sync(s)`` is great when your system is locked up, it allows you to sync your
177
161
disks and will certainly lessen the chance of data loss and fscking. Note
178
162
that the sync hasn't taken place until you see the "OK" and "Done" appear
179
163
on the screen. (If the kernel is really in strife, you may not ever get the
180
164
OK or Done message...)
181
165
182
-
umount(u) is basically useful in the same ways as sync(s). I generally sync(s),
183
-
umount(u), then reboot(b) when my system locks. It's saved me many a fsck.
184
-
Again, the unmount (remount read-only) hasn't taken place until you see the
185
-
"OK" and "Done" message appear on the screen.
166
+
``umount(u)`` is basically useful in the same ways as ``sync(s)``. I generally
167
+
``sync(s)``, ``umount(u)``, then ``reboot(b)`` when my system locks. It's saved
168
+
me many a fsck. Again, the unmount (remount read-only) hasn't taken place until
169
+
you see the "OK" and "Done" message appear on the screen.
186
170
187
-
The loglevels '0'-'9' are useful when your console is being flooded with
188
-
kernel messages you do not want to see. Selecting '0' will prevent all but
171
+
The loglevels ``0``-``9`` are useful when your console is being flooded with
172
+
kernel messages you do not want to see. Selecting ``0`` will prevent all but
189
173
the most urgent kernel messages from reaching your console. (They will
190
174
still be logged if syslogd/klogd are alive, though.)
191
175
192
-
term(e) and kill(i) are useful if you have some sort of runaway process you
193
-
are unable to kill any other way, especially if it's spawning other
176
+
``term(e)`` and ``kill(i)`` are useful if you have some sort of runaway process
177
+
you are unable to kill any other way, especially if it's spawning other
194
178
processes.
195
179
196
-
"just thaw it(j)" is useful if your system becomes unresponsive due to a frozen
197
-
(probably root) filesystem via the FIFREEZE ioctl.
180
+
"just thaw ``it(j)``" is useful if your system becomes unresponsive due to a
181
+
frozen (probably root) filesystem via the FIFREEZE ioctl.
198
182
199
-
* Sometimes SysRq seems to get 'stuck' after using it, what can I do?
200
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
183
+
Sometimes SysRq seems to get 'stuck' after using it, what can I do?
184
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
185
+
201
186
That happens to me, also. I've found that tapping shift, alt, and control
202
187
on both sides of the keyboard, and hitting an invalid sysrq sequence again
203
-
will fix the problem. (i.e., something like alt-sysrq-z). Switching to another
204
-
virtual console (ALT+Fn) and then back again should also help.
188
+
will fix the problem. (i.e., something like :kbd:`alt-sysrq-z`). Switching to
189
+
another virtual console (:kbd:`ALT+Fn`) and then back again should also help.
205
190
206
-
* I hit SysRq, but nothing seems to happen, what's wrong?
207
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
191
+
I hit SysRq, but nothing seems to happen, what's wrong?
192
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
193
+
208
194
There are some keyboards that produce a different keycode for SysRq than the
209
-
pre-defined value of 99 (see KEY_SYSRQ in include/linux/input.h), or which
210
-
don't have a SysRq key at all. In these cases, run 'showkey -s' to find an
211
-
appropriate scancode sequence, and use 'setkeycodes <sequence> 99' to map
212
-
this sequence to the usual SysRq code (e.g., 'setkeycodes e05b 99'). It's
195
+
pre-defined value of 99 (see ``KEY_SYSRQ`` in ``include/linux/input.h``), or
196
+
which don't have a SysRq key at all. In these cases, run ``showkey -s`` to find
197
+
an appropriate scancode sequence, and use ``setkeycodes <sequence> 99`` to map
198
+
this sequence to the usual SysRq code (e.g., ``setkeycodes e05b 99``). It's
213
199
probably best to put this command in a boot script. Oh, and by the way, you
214
-
exit 'showkey' by not typing anything for ten seconds.
200
+
exit ``showkey`` by not typing anything for ten seconds.
215
201
216
-
* I want to add SysRQ key events to a module, how does it work?
217
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
202
+
I want to add SysRQ key events to a module, how does it work?
203
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
204
+
218
205
In order to register a basic function with the table, you must first include
219
-
the header 'include/linux/sysrq.h', this will define everything else you need.
220
-
Next, you must create a sysrq_key_op struct, and populate it with A) the key
206
+
the header ``include/linux/sysrq.h``, this will define everything else you need.
207
+
Next, you must create a ``sysrq_key_op`` struct, and populate it with A) the key
221
208
handler function you will use, B) a help_msg string, that will print when SysRQ
222
209
prints help, and C) an action_msg string, that will print right before your
223
210
handler is called. Your handler must conform to the prototype in 'sysrq.h'.
224
211
225
-
After the sysrq_key_op is created, you can call the kernel function
226
-
register_sysrq_key(int key, struct sysrq_key_op *op_p); this will
227
-
register the operation pointed to by 'op_p' at table key 'key',
212
+
After the ``sysrq_key_op`` is created, you can call the kernel function
213
+
``register_sysrq_key(int key, struct sysrq_key_op *op_p);`` this will
214
+
register the operation pointed to by ``op_p`` at table key 'key',
228
215
if that slot in the table is blank. At module unload time, you must call
229
-
the function unregister_sysrq_key(int key, struct sysrq_key_op *op_p), which
216
+
the function ``unregister_sysrq_key(int key, struct sysrq_key_op *op_p)``, which
230
217
will remove the key op pointed to by 'op_p' from the key 'key', if and only if
231
218
it is currently registered in that slot. This is in case the slot has been
232
219
overwritten since you registered it.
···
241
214
The Magic SysRQ system works by registering key operations against a key op
242
215
lookup table, which is defined in 'drivers/tty/sysrq.c'. This key table has
243
216
a number of operations registered into it at compile time, but is mutable,
244
-
and 2 functions are exported for interface to it:
217
+
and 2 functions are exported for interface to it::
218
+
245
219
register_sysrq_key and unregister_sysrq_key.
220
+
246
221
Of course, never ever leave an invalid pointer in the table. I.e., when
247
222
your module that called register_sysrq_key() exits, it must call
248
223
unregister_sysrq_key() to clean up the sysrq key table entry that it used.
···
253
224
If for some reason you feel the need to call the handle_sysrq function from
254
225
within a function called by handle_sysrq, you must be aware that you are in
255
226
a lock (you are also in an interrupt handler, which means don't sleep!), so
256
-
you must call __handle_sysrq_nolock instead.
227
+
you must call ``__handle_sysrq_nolock`` instead.
257
228
258
-
* When I hit a SysRq key combination only the header appears on the console?
259
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
229
+
When I hit a SysRq key combination only the header appears on the console?
230
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
231
+
260
232
Sysrq output is subject to the same console loglevel control as all
261
233
other console output. This means that if the kernel was booted 'quiet'
262
234
as is common on distro kernels the output may not appear on the actual
263
235
console, even though it will appear in the dmesg buffer, and be accessible
264
-
via the dmesg command and to the consumers of /proc/kmsg. As a specific
236
+
via the dmesg command and to the consumers of ``/proc/kmsg``. As a specific
265
237
exception the header line from the sysrq command is passed to all console
266
238
consumers as if the current loglevel was maximum. If only the header
267
239
is emitted it is almost certain that the kernel loglevel is too low.
268
240
Should you require the output on the console channel then you will need
269
-
to temporarily up the console loglevel using alt-sysrq-8 or:
241
+
to temporarily up the console loglevel using :kbd:`alt-sysrq-8` or::
270
242
271
243
echo 8 > /proc/sysrq-trigger
272
244
273
245
Remember to return the loglevel to normal after triggering the sysrq
274
246
command you are interested in.
275
247
276
-
* I have more questions, who can I ask?
277
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
248
+
I have more questions, who can I ask?
249
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
250
+
278
251
Just ask them on the linux-kernel mailing list:
279
252
linux-kernel@vger.kernel.org
280
253
281
-
* Credits
282
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
254
+
Credits
255
+
~~~~~~~
256
+
283
257
Written by Mydraal <vulpyne@vulpyne.net>
284
258
Updated by Adam Sulmicki <adam@cfar.umd.edu>
285
259
Updated by Jeremy M. Dolan <jmd@turbogeek.org> 2001/01/28 10:15:59