aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/sysrq.txt
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 18:20:36 -0400
committerLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 18:20:36 -0400
commit1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 (patch)
tree0bba044c4ce775e45a88a51686b5d9f90697ea9d /Documentation/sysrq.txt
Linux-2.6.12-rc2v2.6.12-rc2
Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
Diffstat (limited to 'Documentation/sysrq.txt')
-rw-r--r--Documentation/sysrq.txt213
1 files changed, 213 insertions, 0 deletions
diff --git a/Documentation/sysrq.txt b/Documentation/sysrq.txt
new file mode 100644
index 000000000000..f98c2e31c143
--- /dev/null
+++ b/Documentation/sysrq.txt
@@ -0,0 +1,213 @@
1Linux Magic System Request Key Hacks
2Documentation for sysrq.c version 1.15
3Last update: $Date: 2001/01/28 10:15:59 $
4
5* What is the magic SysRq key?
6~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
7It is a 'magical' key combo you can hit which the kernel will respond to
8regardless of whatever else it is doing, unless it is completely locked up.
9
10* How do I enable the magic SysRq key?
11~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
12You need to say "yes" to 'Magic SysRq key (CONFIG_MAGIC_SYSRQ)' when
13configuring the kernel. When running a kernel with SysRq compiled in,
14/proc/sys/kernel/sysrq controls the functions allowed to be invoked via
15the SysRq key. By default the file contains 1 which means that every
16possible SysRq request is allowed (in older versions SysRq was disabled
17by default, and you were required to specifically enable it at run-time
18but this is not the case any more). Here is the list of possible values
19in /proc/sys/kernel/sysrq:
20 0 - disable sysrq completely
21 1 - enable all functions of sysrq
22 >1 - bitmask of allowed sysrq functions (see below for detailed function
23 description):
24 2 - enable control of console logging level
25 4 - enable control of keyboard (SAK, unraw)
26 8 - enable debugging dumps of processes etc.
27 16 - enable sync command
28 32 - enable remount read-only
29 64 - enable signalling of processes (term, kill, oom-kill)
30 128 - allow reboot/poweroff
31 256 - allow nicing of all RT tasks
32
33You can set the value in the file by the following command:
34 echo "number" >/proc/sys/kernel/sysrq
35
36Note that the value of /proc/sys/kernel/sysrq influences only the invocation
37via a keyboard. Invocation of any operation via /proc/sysrq-trigger is always
38allowed.
39
40* How do I use the magic SysRq key?
41~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
42On x86 - You press the key combo 'ALT-SysRq-<command key>'. Note - Some
43 keyboards may not have a key labeled 'SysRq'. The 'SysRq' key is
44 also known as the 'Print Screen' key. Also some keyboards cannot
45 handle so many keys being pressed at the same time, so you might
46 have better luck with "press Alt", "press SysRq", "release Alt",
47 "press <command key>", release everything.
48
49On SPARC - You press 'ALT-STOP-<command key>', I believe.
50
51On the serial console (PC style standard serial ports only) -
52 You send a BREAK, then within 5 seconds a command key. Sending
53 BREAK twice is interpreted as a normal BREAK.
54
55On PowerPC - Press 'ALT - Print Screen (or F13) - <command key>,
56 Print Screen (or F13) - <command key> may suffice.
57
58On other - If you know of the key combos for other architectures, please
59 let me know so I can add them to this section.
60
61On all - write a character to /proc/sysrq-trigger. eg:
62
63 echo t > /proc/sysrq-trigger
64
65* What are the 'command' keys?
66~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
67'r' - Turns off keyboard raw mode and sets it to XLATE.
68
69'k' - Secure Access Key (SAK) Kills all programs on the current virtual
70 console. NOTE: See important comments below in SAK section.
71
72'b' - Will immediately reboot the system without syncing or unmounting
73 your disks.
74
75'o' - Will shut your system off (if configured and supported).
76
77's' - Will attempt to sync all mounted filesystems.
78
79'u' - Will attempt to remount all mounted filesystems read-only.
80
81'p' - Will dump the current registers and flags to your console.
82
83't' - Will dump a list of current tasks and their information to your
84 console.
85
86'm' - Will dump current memory info to your console.
87
88'v' - Dumps Voyager SMP processor info to your console.
89
90'0'-'9' - Sets the console log level, controlling which kernel messages
91 will be printed to your console. ('0', for example would make
92 it so that only emergency messages like PANICs or OOPSes would
93 make it to your console.)
94
95'f' - Will call oom_kill to kill a memory hog process
96
97'e' - Send a SIGTERM to all processes, except for init.
98
99'i' - Send a SIGKILL to all processes, except for init.
100
101'l' - Send a SIGKILL to all processes, INCLUDING init. (Your system
102 will be non-functional after this.)
103
104'h' - Will display help ( actually any other key than those listed
105 above will display help. but 'h' is easy to remember :-)
106
107* Okay, so what can I use them for?
108~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
109Well, un'R'aw is very handy when your X server or a svgalib program crashes.
110
111sa'K' (Secure Access Key) is useful when you want to be sure there are no
112trojan program is running at console and which could grab your password
113when you would try to login. It will kill all programs on given console
114and thus letting you make sure that the login prompt you see is actually
115the one from init, not some trojan program.
116IMPORTANT:In its true form it is not a true SAK like the one in :IMPORTANT
117IMPORTANT:c2 compliant systems, and it should be mistook as such. :IMPORTANT
118 It seems other find it useful as (System Attention Key) which is
119useful when you want to exit a program that will not let you switch consoles.
120(For example, X or a svgalib program.)
121
122re'B'oot is good when you're unable to shut down. But you should also 'S'ync
123and 'U'mount first.
124
125'S'ync is great when your system is locked up, it allows you to sync your
126disks and will certainly lessen the chance of data loss and fscking. Note
127that the sync hasn't taken place until you see the "OK" and "Done" appear
128on the screen. (If the kernel is really in strife, you may not ever get the
129OK or Done message...)
130
131'U'mount is basically useful in the same ways as 'S'ync. I generally 'S'ync,
132'U'mount, then re'B'oot when my system locks. It's saved me many a fsck.
133Again, the unmount (remount read-only) hasn't taken place until you see the
134"OK" and "Done" message appear on the screen.
135
136The loglevel'0'-'9' is useful when your console is being flooded with
137kernel messages you do not want to see. Setting '0' will prevent all but
138the most urgent kernel messages from reaching your console. (They will
139still be logged if syslogd/klogd are alive, though.)
140
141t'E'rm and k'I'll are useful if you have some sort of runaway process you
142are unable to kill any other way, especially if it's spawning other
143processes.
144
145* Sometimes SysRq seems to get 'stuck' after using it, what can I do?
146~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
147That happens to me, also. I've found that tapping shift, alt, and control
148on both sides of the keyboard, and hitting an invalid sysrq sequence again
149will fix the problem. (ie, something like alt-sysrq-z). Switching to another
150virtual console (ALT+Fn) and then back again should also help.
151
152* I hit SysRq, but nothing seems to happen, what's wrong?
153~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
154There are some keyboards that send different scancodes for SysRq than the
155pre-defined 0x54. So if SysRq doesn't work out of the box for a certain
156keyboard, run 'showkey -s' to find out the proper scancode sequence. Then
157use 'setkeycodes <sequence> 84' to define this sequence to the usual SysRq
158code (84 is decimal for 0x54). It's probably best to put this command in a
159boot script. Oh, and by the way, you exit 'showkey' by not typing anything
160for ten seconds.
161
162* I want to add SysRQ key events to a module, how does it work?
163~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
164In order to register a basic function with the table, you must first include
165the header 'include/linux/sysrq.h', this will define everything else you need.
166Next, you must create a sysrq_key_op struct, and populate it with A) the key
167handler function you will use, B) a help_msg string, that will print when SysRQ
168prints help, and C) an action_msg string, that will print right before your
169handler is called. Your handler must conform to the protoype in 'sysrq.h'.
170
171After the sysrq_key_op is created, you can call the macro
172register_sysrq_key(int key, struct sysrq_key_op *op_p) that is defined in
173sysrq.h, this will register the operation pointed to by 'op_p' at table
174key 'key', if that slot in the table is blank. At module unload time, you must
175call the macro unregister_sysrq_key(int key, struct sysrq_key_op *op_p), which
176will remove the key op pointed to by 'op_p' from the key 'key', if and only if
177it is currently registered in that slot. This is in case the slot has been
178overwritten since you registered it.
179
180The Magic SysRQ system works by registering key operations against a key op
181lookup table, which is defined in 'drivers/char/sysrq.c'. This key table has
182a number of operations registered into it at compile time, but is mutable,
183and 4 functions are exported for interface to it: __sysrq_lock_table,
184__sysrq_unlock_table, __sysrq_get_key_op, and __sysrq_put_key_op. The
185functions __sysrq_swap_key_ops and __sysrq_swap_key_ops_nolock are defined
186in the header itself, and the REGISTER and UNREGISTER macros are built from
187these. More complex (and dangerous!) manipulations of the table are possible
188using these functions, but you must be careful to always lock the table before
189you read or write from it, and to unlock it again when you are done. (And of
190course, to never ever leave an invalid pointer in the table). Null pointers in
191the table are always safe :)
192
193If for some reason you feel the need to call the handle_sysrq function from
194within a function called by handle_sysrq, you must be aware that you are in
195a lock (you are also in an interrupt handler, which means don't sleep!), so
196you must call __handle_sysrq_nolock instead.
197
198* I have more questions, who can I ask?
199~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
200You may feel free to send email to myrdraal@deathsdoor.com, and I will
201respond as soon as possible.
202 -Myrdraal
203
204And I'll answer any questions about the registration system you got, also
205responding as soon as possible.
206 -Crutcher
207
208* Credits
209~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
210Written by Mydraal <myrdraal@deathsdoor.com>
211Updated by Adam Sulmicki <adam@cfar.umd.edu>
212Updated by Jeremy M. Dolan <jmd@turbogeek.org> 2001/01/28 10:15:59
213Added to by Crutcher Dunnavant <crutcher+kernel@datastacks.com>