1SIGNAL(7) Linux Programmer's Manual SIGNAL(7)
2
6 signal - overview of signals
7
9 Linux supports both POSIX reliable signals (hereinafter "standard sig‐
10 nals") and POSIX real-time signals.
11 Signal dispositions
13 Each signal has a current disposition, which determines how the process
14 behaves when it is delivered the signal.
15 The entries in the "Action" column of the table below specify the
17 default disposition for each signal, as follows:
18 Term Default action is to terminate the process.
20 Ign Default action is to ignore the signal.
22 Core Default action is to terminate the process and dump core (see
24 core(5)).
25 Stop Default action is to stop the process.
27 Cont Default action is to continue the process if it is currently
29 stopped.
30 A process can change the disposition of a signal using sigaction(2) or
32 signal(2). (The latter is less portable when establishing a signal
33 handler; see signal(2) for details.) Using these system calls, a
34 process can elect one of the following behaviors to occur on delivery
35 of the signal: perform the default action; ignore the signal; or catch
36 the signal with a signal handler, a programmer-defined function that is
37 automatically invoked when the signal is delivered.
38 By default, a signal handler is invoked on the normal process stack.
40 It is possible to arrange that the signal handler uses an alternate
41 stack; see sigaltstack(2) for a discussion of how to do this and when
42 it might be useful.
43 The signal disposition is a per-process attribute: in a multithreaded
45 application, the disposition of a particular signal is the same for all
46 threads.
47 A child created via fork(2) inherits a copy of its parent's signal dis‐
49 positions. During an execve(2), the dispositions of handled signals
50 are reset to the default; the dispositions of ignored signals are left
51 unchanged.
52 Sending a signal
54 The following system calls and library functions allow the caller to
55 send a signal:
56 raise(3) Sends a signal to the calling thread.
58 kill(2) Sends a signal to a specified process, to all members
60 of a specified process group, or to all processes on
61 the system.
62 killpg(3) Sends a signal to all of the members of a specified
64 process group.
65 pthread_kill(3) Sends a signal to a specified POSIX thread in the same
67 process as the caller.
68 tgkill(2) Sends a signal to a specified thread within a specific
70 process. (This is the system call used to implement
71 pthread_kill(3).)
72 sigqueue(3) Sends a real-time signal with accompanying data to a
74 specified process.
75 Waiting for a signal to be caught
77 The following system calls suspend execution of the calling thread
78 until a signal is caught (or an unhandled signal terminates the
79 process):
80 pause(2) Suspends execution until any signal is caught.
82 sigsuspend(2) Temporarily changes the signal mask (see below) and
84 suspends execution until one of the unmasked signals is
85 caught.
86 Synchronously accepting a signal
88 Rather than asynchronously catching a signal via a signal handler, it
89 is possible to synchronously accept the signal, that is, to block exe‐
90 cution until the signal is delivered, at which point the kernel returns
91 information about the signal to the caller. There are two general ways
92 to do this:
93 * sigwaitinfo(2), sigtimedwait(2), and sigwait(3) suspend execution
95 until one of the signals in a specified set is delivered. Each of
96 these calls returns information about the delivered signal.
97 * signalfd(2) returns a file descriptor that can be used to read infor‐
99 mation about signals that are delivered to the caller. Each read(2)
100 from this file descriptor blocks until one of the signals in the set
101 specified in the signalfd(2) call is delivered to the caller. The
102 buffer returned by read(2) contains a structure describing the sig‐
103 nal.
104 Signal mask and pending signals
106 A signal may be blocked, which means that it will not be delivered
107 until it is later unblocked. Between the time when it is generated and
108 when it is delivered a signal is said to be pending.
109 Each thread in a process has an independent signal mask, which indi‐
111 cates the set of signals that the thread is currently blocking. A
112 thread can manipulate its signal mask using pthread_sigmask(3). In a
113 traditional single-threaded application, sigprocmask(2) can be used to
114 manipulate the signal mask.
115 A child created via fork(2) inherits a copy of its parent's signal
117 mask; the signal mask is preserved across execve(2).
118 A signal may be process-directed or thread-directed. A process-
120 directed signal is one that is targeted at (and thus pending for) the
121 process as a whole. A signal may be process-directed because it was
122 generated by the kernel for reasons other than a hardware exception, or
123 because it was sent using kill(2) or sigqueue(3). A thread-directed
124 signal is one that is targeted at a specific thread. A signal may be
125 thread-directed because it was generated as a consequence of executing
126 a specific machine-language instruction that triggered a hardware
127 exception (e.g., SIGSEGV for an invalid memory access, or SIGFPE for a
128 math error), or because it was it was targeted at a specific thread
129 using interfaces such as tgkill(2) or pthread_kill(3).
130 A process-directed signal may be delivered to any one of the threads
132 that does not currently have the signal blocked. If more than one of
133 the threads has the signal unblocked, then the kernel chooses an arbi‐
134 trary thread to which to deliver the signal.
135 A thread can obtain the set of signals that it currently has pending
137 using sigpending(2). This set will consist of the union of the set of
138 pending process-directed signals and the set of signals pending for the
139 calling thread.
140 A child created via fork(2) initially has an empty pending signal set;
142 the pending signal set is preserved across an execve(2).
143 Standard signals
145 Linux supports the standard signals listed below. The second column of
146 the table indicates which standard (if any) specified the signal:
147 "P1990" indicates that the signal is described in the original
148 POSIX.1-1990 standard; "P2001" indicates that the signal was added in
149 SUSv2 and POSIX.1-2001.
150 Signal Standard Action Comment
152 ────────────────────────────────────────────────────────────────────────
153 SIGABRT P1990 Core Abort signal from abort(3)
154 SIGALRM P1990 Term Timer signal from alarm(2)
155 SIGBUS P2001 Core Bus error (bad memory access)
156 SIGCHLD P1990 Ign Child stopped or terminated
157 SIGCLD - Ign A synonym for SIGCHLD
158 SIGCONT P1990 Cont Continue if stopped
159 SIGEMT - Term Emulator trap
160 SIGFPE P1990 Core Floating-point exception
161 SIGHUP P1990 Term Hangup detected on controlling terminal
162 or death of controlling process
163 SIGILL P1990 Core Illegal Instruction
164 SIGINFO - A synonym for SIGPWR
165 SIGINT P1990 Term Interrupt from keyboard
166 SIGIO - Term I/O now possible (4.2BSD)
167 SIGIOT - Core IOT trap. A synonym for SIGABRT
168 SIGKILL P1990 Term Kill signal
169 SIGLOST - Term File lock lost (unused)
170 SIGPIPE P1990 Term Broken pipe: write to pipe with no
171 readers; see pipe(7)
172 SIGPOLL P2001 Term Pollable event (Sys V).
173 Synonym for SIGIO
174 SIGPROF P2001 Term Profiling timer expired
175 SIGPWR - Term Power failure (System V)
176 SIGQUIT P1990 Core Quit from keyboard
177 SIGSEGV P1990 Core Invalid memory reference
178 SIGSTKFLT - Term Stack fault on coprocessor (unused)
179 SIGSTOP P1990 Stop Stop process
180 SIGTSTP P1990 Stop Stop typed at terminal
181 SIGSYS P2001 Core Bad system call (SVr4);
182 see also seccomp(2)
183 SIGTERM P1990 Term Termination signal
184 SIGTRAP P2001 Core Trace/breakpoint trap
185 SIGTTIN P1990 Stop Terminal input for background process
186 SIGTTOU P1990 Stop Terminal output for background process
187 SIGUNUSED - Core Synonymous with SIGSYS
188 SIGURG P2001 Ign Urgent condition on socket (4.2BSD)
189 SIGUSR1 P1990 Term User-defined signal 1
190 SIGUSR2 P1990 Term User-defined signal 2
191 SIGVTALRM P2001 Term Virtual alarm clock (4.2BSD)
192 SIGXCPU P2001 Core CPU time limit exceeded (4.2BSD);
193 see setrlimit(2)
194 SIGXFSZ P2001 Core File size limit exceeded (4.2BSD);
195 see setrlimit(2)
196 SIGWINCH - Ign Window resize signal (4.3BSD, Sun)
197 The signals SIGKILL and SIGSTOP cannot be caught, blocked, or ignored.
199 Up to and including Linux 2.2, the default behavior for SIGSYS, SIGX‐
201 CPU, SIGXFSZ, and (on architectures other than SPARC and MIPS) SIGBUS
202 was to terminate the process (without a core dump). (On some other
203 UNIX systems the default action for SIGXCPU and SIGXFSZ is to terminate
204 the process without a core dump.) Linux 2.4 conforms to the
205 POSIX.1-2001 requirements for these signals, terminating the process
206 with a core dump.
207 SIGEMT is not specified in POSIX.1-2001, but nevertheless appears on
209 most other UNIX systems, where its default action is typically to ter‐
210 minate the process with a core dump.
211 SIGPWR (which is not specified in POSIX.1-2001) is typically ignored by
213 default on those other UNIX systems where it appears.
214 SIGIO (which is not specified in POSIX.1-2001) is ignored by default on
216 several other UNIX systems.
217 Queueing and delivery semantics for standard signals
219 If multiple standard signals are pending for a process, the order in
220 which the signals are delivered is unspecified.
221 Standard signals do not queue. If multiple instances of a standard
223 signal are generated while that signal is blocked, then only one
224 instance of the signal is marked as pending (and the signal will be
225 delivered just once when it is unblocked). In the case where a stan‐
226 dard signal is already pending, the siginfo_t structure (see sigac‐
227 tion(2)) associated with that signal is not overwritten on arrival of
228 subsequent instances of the same signal. Thus, the process will
229 receive the information associated with the first instance of the sig‐
230 nal.
231 Signal numbering for standard signals
233 The numeric value for each signal is given in the table below. As
234 shown in the table, many signals have different numeric values on dif‐
235 ferent architectures. The first numeric value in each table row shows
236 the signal number on x86, ARM, and most other architectures; the second
237 value is for Alpha and SPARC; the third is for MIPS; and the last is
238 for PARISC. A dash (-) denotes that a signal is absent on the corre‐
239 sponding architecture.
240 Signal x86/ARM Alpha/ MIPS PARISC Notes
242 most others SPARC
243 ─────────────────────────────────────────────────────────────────
244 SIGHUP 1 1 1 1
245 SIGINT 2 2 2 2
246 SIGQUIT 3 3 3 3
247 SIGILL 4 4 4 4
248 SIGTRAP 5 5 5 5
249 SIGABRT 6 6 6 6
250 SIGIOT 6 6 6 6
251 SIGBUS 7 10 10 10
252 SIGEMT - 7 7 -
253 SIGFPE 8 8 8 8
254 SIGKILL 9 9 9 9
255 SIGUSR1 10 30 16 16
256 SIGSEGV 11 11 11 11
257 SIGUSR2 12 31 17 17
258 SIGPIPE 13 13 13 13
259 SIGALRM 14 14 14 14
260 SIGTERM 15 15 15 15
261 SIGSTKFLT 16 - - 7
262 SIGCHLD 17 20 18 18
263 SIGCLD - - 18 -
264 SIGCONT 18 19 25 26
265 SIGSTOP 19 17 23 24
266 SIGTSTP 20 18 24 25
267 SIGTTIN 21 21 26 27
268 SIGTTOU 22 22 27 28
270 SIGURG 23 16 21 29
271 SIGXCPU 24 24 30 12
272 SIGXFSZ 25 25 31 30
273 SIGVTALRM 26 26 28 20
274 SIGPROF 27 27 29 21
275 SIGWINCH 28 28 20 23
276 SIGIO 29 23 22 22
277 SIGPOLL Same as SIGIO
278 SIGPWR 30 29/- 19 19
279 SIGINFO - 29/- - -
280 SIGLOST - -/29 - -
281 SIGSYS 31 12 12 31
282 SIGUNUSED 31 - - 31
283 Note the following:
285 * Where defined, SIGUNUSED is synonymous with SIGSYS. Since glibc
287 2.26, SIGUNUSED is no longer defined on any architecture.
288 * Signal 29 is SIGINFO/SIGPWR (synonyms for the same value) on Alpha
290 but SIGLOST on SPARC.
291 Real-time signals
293 Starting with version 2.2, Linux supports real-time signals as origi‐
294 nally defined in the POSIX.1b real-time extensions (and now included in
295 POSIX.1-2001). The range of supported real-time signals is defined by
296 the macros SIGRTMIN and SIGRTMAX. POSIX.1-2001 requires that an imple‐
297 mentation support at least _POSIX_RTSIG_MAX (8) real-time signals.
298 The Linux kernel supports a range of 33 different real-time signals,
300 numbered 32 to 64. However, the glibc POSIX threads implementation
301 internally uses two (for NPTL) or three (for LinuxThreads) real-time
302 signals (see pthreads(7)), and adjusts the value of SIGRTMIN suitably
303 (to 34 or 35). Because the range of available real-time signals varies
304 according to the glibc threading implementation (and this variation can
305 occur at run time according to the available kernel and glibc), and
306 indeed the range of real-time signals varies across UNIX systems, pro‐
307 grams should never refer to real-time signals using hard-coded numbers,
308 but instead should always refer to real-time signals using the notation
309 SIGRTMIN+n, and include suitable (run-time) checks that SIGRTMIN+n does
310 not exceed SIGRTMAX.
311 Unlike standard signals, real-time signals have no predefined meanings:
313 the entire set of real-time signals can be used for application-defined
314 purposes.
315 The default action for an unhandled real-time signal is to terminate
317 the receiving process.
318 Real-time signals are distinguished by the following:
320 1. Multiple instances of real-time signals can be queued. By con‐
322 trast, if multiple instances of a standard signal are delivered
323 while that signal is currently blocked, then only one instance is
324 queued.
325 2. If the signal is sent using sigqueue(3), an accompanying value
327 (either an integer or a pointer) can be sent with the signal. If
328 the receiving process establishes a handler for this signal using
329 the SA_SIGINFO flag to sigaction(2), then it can obtain this data
330 via the si_value field of the siginfo_t structure passed as the
331 second argument to the handler. Furthermore, the si_pid and si_uid
332 fields of this structure can be used to obtain the PID and real
333 user ID of the process sending the signal.
334 3. Real-time signals are delivered in a guaranteed order. Multiple
336 real-time signals of the same type are delivered in the order they
337 were sent. If different real-time signals are sent to a process,
338 they are delivered starting with the lowest-numbered signal.
339 (I.e., low-numbered signals have highest priority.) By contrast,
340 if multiple standard signals are pending for a process, the order
341 in which they are delivered is unspecified.
342 If both standard and real-time signals are pending for a process, POSIX
344 leaves it unspecified which is delivered first. Linux, like many other
345 implementations, gives priority to standard signals in this case.
346 According to POSIX, an implementation should permit at least
348 _POSIX_SIGQUEUE_MAX [22m(32) real-time signals to be queued to a process.
349 However, Linux does things differently. In kernels up to and including
350 2.6.7, Linux imposes a system-wide limit on the number of queued real-
351 time signals for all processes. This limit can be viewed and (with
352 privilege) changed via the /proc/sys/kernel/rtsig-max file. A related
353 file, /proc/sys/kernel/rtsig-nr, can be used to find out how many real-
354 time signals are currently queued. In Linux 2.6.8, these /proc inter‐
355 faces were replaced by the RLIMIT_SIGPENDING resource limit, which
356 specifies a per-user limit for queued signals; see setrlimit(2) for
357 further details.
358 The addition of real-time signals required the widening of the signal
360 set structure (sigset_t) from 32 to 64 bits. Consequently, various
361 system calls were superseded by new system calls that supported the
362 larger signal sets. The old and new system calls are as follows:
363 Linux 2.0 and earlier Linux 2.2 and later
365 sigaction(2) rt_sigaction(2)
366 sigpending(2) rt_sigpending(2)
367 sigprocmask(2) rt_sigprocmask(2)
368 sigreturn(2) rt_sigreturn(2)
369 sigsuspend(2) rt_sigsuspend(2)
370 sigtimedwait(2) rt_sigtimedwait(2)
371 Interruption of system calls and library functions by signal handlers
373 If a signal handler is invoked while a system call or library function
374 call is blocked, then either:
375 * the call is automatically restarted after the signal handler returns;
377 or
378 * the call fails with the error EINTR.
380 Which of these two behaviors occurs depends on the interface and
382 whether or not the signal handler was established using the SA_RESTART
383 flag (see sigaction(2)). The details vary across UNIX systems; below,
384 the details for Linux.
385 If a blocked call to one of the following interfaces is interrupted by
387 a signal handler, then the call is automatically restarted after the
388 signal handler returns if the SA_RESTART flag was used; otherwise the
389 call fails with the error EINTR:
390 * read(2), readv(2), write(2), writev(2), and ioctl(2) calls on "slow"
392 devices. A "slow" device is one where the I/O call may block for an
393 indefinite time, for example, a terminal, pipe, or socket. If an I/O
394 call on a slow device has already transferred some data by the time
395 it is interrupted by a signal handler, then the call will return a
396 success status (normally, the number of bytes transferred). Note
397 that a (local) disk is not a slow device according to this defini‐
398 tion; I/O operations on disk devices are not interrupted by signals.
399 * open(2), if it can block (e.g., when opening a FIFO; see fifo(7)).
401 * wait(2), wait3(2), wait4(2), waitid(2), and waitpid(2).
403 * Socket interfaces: accept(2), connect(2), recv(2), recvfrom(2),
405 recvmmsg(2), recvmsg(2), send(2), sendto(2), and sendmsg(2), unless a
406 timeout has been set on the socket (see below).
407 * File locking interfaces: flock(2) and the F_SETLKW and F_OFD_SETLKW
409 operations of fcntl(2)
410 * POSIX message queue interfaces: mq_receive(3), mq_timedreceive(3),
412 mq_send(3), and mq_timedsend(3).
413 * futex(2) FUTEX_WAIT (since Linux 2.6.22; beforehand, always failed
415 with EINTR).
416 * getrandom(2).
418 * pthread_mutex_lock(3), pthread_cond_wait(3), and related APIs.
420 * futex(2) FUTEX_WAIT_BITSET.
422 * POSIX semaphore interfaces: sem_wait(3) and sem_timedwait(3) (since
424 Linux 2.6.22; beforehand, always failed with EINTR).
425 * read(2) from an inotify(7) file descriptor (since Linux 3.8; before‐
427 hand, always failed with EINTR).
428 The following interfaces are never restarted after being interrupted by
430 a signal handler, regardless of the use of SA_RESTART; they always fail
431 with the error EINTR when interrupted by a signal handler:
432 * "Input" socket interfaces, when a timeout (SO_RCVTIMEO) has been set
434 on the socket using setsockopt(2): accept(2), recv(2), recvfrom(2),
435 recvmmsg(2) (also with a non-NULL timeout argument), and recvmsg(2).
436 * "Output" socket interfaces, when a timeout (SO_RCVTIMEO) has been set
438 on the socket using setsockopt(2): connect(2), send(2), sendto(2),
439 and sendmsg(2).
440 * Interfaces used to wait for signals: pause(2), sigsuspend(2), sig‐
442 timedwait(2), and sigwaitinfo(2).
443 * File descriptor multiplexing interfaces: epoll_wait(2),
445 epoll_pwait(2), poll(2), ppoll(2), select(2), and pselect(2).
446 * System V IPC interfaces: msgrcv(2), msgsnd(2), semop(2), and semtime‐
448 dop(2).
449 * Sleep interfaces: clock_nanosleep(2), nanosleep(2), and usleep(3).
451 * io_getevents(2).
453 The sleep(3) function is also never restarted if interrupted by a han‐
455 dler, but gives a success return: the number of seconds remaining to
456 sleep.
457 Interruption of system calls and library functions by stop signals
459 On Linux, even in the absence of signal handlers, certain blocking
460 interfaces can fail with the error EINTR after the process is stopped
461 by one of the stop signals and then resumed via SIGCONT. This behavior
462 is not sanctioned by POSIX.1, and doesn't occur on other systems.
463 The Linux interfaces that display this behavior are:
465 * "Input" socket interfaces, when a timeout (SO_RCVTIMEO) has been set
467 on the socket using setsockopt(2): accept(2), recv(2), recvfrom(2),
468 recvmmsg(2) (also with a non-NULL timeout argument), and recvmsg(2).
469 * "Output" socket interfaces, when a timeout (SO_RCVTIMEO) has been set
471 on the socket using setsockopt(2): connect(2), send(2), sendto(2),
472 and sendmsg(2), if a send timeout (SO_SNDTIMEO) has been set.
473 * epoll_wait(2), epoll_pwait(2).
475 * semop(2), semtimedop(2).
477 * sigtimedwait(2), sigwaitinfo(2).
479 * Linux 3.7 and earlier: read(2) from an inotify(7) file descriptor
481 * Linux 2.6.21 and earlier: futex(2) FUTEX_WAIT, sem_timedwait(3),
483 sem_wait(3).
484 * Linux 2.6.8 and earlier: msgrcv(2), msgsnd(2).
486 * Linux 2.4 and earlier: nanosleep(2).
488
490 POSIX.1, except as noted.
491
493 For a discussion of async-signal-safe functions, see signal-safety(7).
494 The /proc/[pid]/task/[tid]/status file contains various fields that
496 show the signals that a thread is blocking (SigBlk), catching (SigCgt),
497 or ignoring (SigIgn). (The set of signals that are caught or ignored
498 will be the same across all threads in a process.) Other fields show
499 the set of pending signals that are directed to the thread (SigPnd) as
500 well as the set of pending signals that are directed to the process as
501 a whole (ShdPnd). The corresponding fields in /proc/[pid]/status show
502 the information for the main thread. See proc(5) for further details.
503
505 kill(1), clone(2), getrlimit(2), kill(2), restart_syscall(2),
506 rt_sigqueueinfo(2), pidfd_send_signal(2), setitimer(2), setrlimit(2),
507 sgetmask(2), sigaction(2), sigaltstack(2), signal(2), signalfd(2), sig‐
508 pending(2), sigprocmask(2), sigreturn(2), sigsuspend(2), sigwait‐
509 info(2), abort(3), bsd_signal(3), killpg(3), longjmp(3),
510 pthread_sigqueue(3), raise(3), sigqueue(3), sigset(3), sigsetops(3),
511 sigvec(3), sigwait(3), strsignal(3), sysv_signal(3), core(5), proc(5),
512 nptl(7), pthreads(7), sigevent(7)
513
515 This page is part of release 5.04 of the Linux man-pages project. A
516 description of the project, information about reporting bugs, and the
517 latest version of this page, can be found at
518 https://www.kernel.org/doc/man-pages/.
519Linux 2019-08-02 SIGNAL(7)