1SIGACTION(2) Linux Programmer's Manual SIGACTION(2)
2
6 sigaction, rt_sigaction - examine and change a signal action
7
9 #include <signal.h>
10 int sigaction(int signum, const struct sigaction *act,
12 struct sigaction *oldact);
13 Feature Test Macro Requirements for glibc (see feature_test_macros(7)):
15 sigaction(): _POSIX_C_SOURCE
17 siginfo_t: _POSIX_C_SOURCE >= 199309L
19
21 The sigaction() system call is used to change the action taken by a
22 process on receipt of a specific signal. (See signal(7) for an over‐
23 view of signals.)
24 signum specifies the signal and can be any valid signal except SIGKILL
26 and SIGSTOP.
27 If act is non-NULL, the new action for signal signum is installed from
29 act. If oldact is non-NULL, the previous action is saved in oldact.
30 The sigaction structure is defined as something like:
32 struct sigaction {
34 void (*sa_handler)(int);
35 void (*sa_sigaction)(int, siginfo_t *, void *);
36 sigset_t sa_mask;
37 int sa_flags;
38 void (*sa_restorer)(void);
39 };
40 On some architectures a union is involved: do not assign to both
42 sa_handler and sa_sigaction.
43 The sa_restorer field is not intended for application use. (POSIX does
45 not specify a sa_restorer field.) Some further details of the purpose
46 of this field can be found in sigreturn(2).
47 sa_handler specifies the action to be associated with signum and may be
49 SIG_DFL for the default action, SIG_IGN to ignore this signal, or a
50 pointer to a signal handling function. This function receives the sig‐
51 nal number as its only argument.
52 If SA_SIGINFO is specified in sa_flags, then sa_sigaction (instead of
54 sa_handler) specifies the signal-handling function for signum. This
55 function receives three arguments, as described below.
56 sa_mask specifies a mask of signals which should be blocked (i.e.,
58 added to the signal mask of the thread in which the signal handler is
59 invoked) during execution of the signal handler. In addition, the sig‐
60 nal which triggered the handler will be blocked, unless the SA_NODEFER
61 flag is used.
62 sa_flags specifies a set of flags which modify the behavior of the sig‐
64 nal. It is formed by the bitwise OR of zero or more of the following:
65 SA_NOCLDSTOP
67 If signum is SIGCHLD, do not receive notification when child
68 processes stop (i.e., when they receive one of SIGSTOP, SIGTSTP,
69 SIGTTIN, or SIGTTOU) or resume (i.e., they receive SIGCONT) (see
70 wait(2)). This flag is meaningful only when establishing a han‐
71 dler for SIGCHLD.
72 SA_NOCLDWAIT (since Linux 2.6)
74 If signum is SIGCHLD, do not transform children into zombies
75 when they terminate. See also waitpid(2). This flag is mean‐
76 ingful only when establishing a handler for SIGCHLD, or when
77 setting that signal's disposition to SIG_DFL.
78 If the SA_NOCLDWAIT flag is set when establishing a handler for
80 SIGCHLD, POSIX.1 leaves it unspecified whether a SIGCHLD signal
81 is generated when a child process terminates. On Linux, a
82 SIGCHLD signal is generated in this case; on some other imple‐
83 mentations, it is not.
84 SA_NODEFER
86 Do not prevent the signal from being received from within its
87 own signal handler. This flag is meaningful only when estab‐
88 lishing a signal handler. SA_NOMASK is an obsolete, nonstandard
89 synonym for this flag.
90 SA_ONSTACK
92 Call the signal handler on an alternate signal stack provided by
93 sigaltstack(2). If an alternate stack is not available, the
94 default stack will be used. This flag is meaningful only when
95 establishing a signal handler.
96 SA_RESETHAND
98 Restore the signal action to the default upon entry to the sig‐
99 nal handler. This flag is meaningful only when establishing a
100 signal handler. SA_ONESHOT is an obsolete, nonstandard synonym
101 for this flag.
102 SA_RESTART
104 Provide behavior compatible with BSD signal semantics by making
105 certain system calls restartable across signals. This flag is
106 meaningful only when establishing a signal handler. See sig‐
107 nal(7) for a discussion of system call restarting.
108 SA_RESTORER
110 Not intended for application use. This flag is used by C
111 libraries to indicate that the sa_restorer field contains the
112 address of a "signal trampoline". See sigreturn(2) for more
113 details.
114 SA_SIGINFO (since Linux 2.2)
116 The signal handler takes three arguments, not one. In this
117 case, sa_sigaction should be set instead of sa_handler. This
118 flag is meaningful only when establishing a signal handler.
119 The siginfo_t argument to a SA_SIGINFO handler
121 When the SA_SIGINFO flag is specified in act.sa_flags, the signal han‐
122 dler address is passed via the act.sa_sigaction field. This handler
123 takes three arguments, as follows:
124 void
126 handler(int sig, siginfo_t *info, void *ucontext)
127 {
128 ...
129 }
130 These three arguments are as follows
132 sig The number of the signal that caused invocation of the handler.
134 info A pointer to a siginfo_t, which is a structure containing fur‐
136 ther information about the signal, as described below.
137 ucontext
139 This is a pointer to a ucontext_t structure, cast to void *.
140 The structure pointed to by this field contains signal context
141 information that was saved on the user-space stack by the ker‐
142 nel; for details, see sigreturn(2). Further information about
143 the ucontext_t structure can be found in getcontext(3). Com‐
144 monly, the handler function doesn't make any use of the third
145 argument.
146 The siginfo_t data type is a structure with the following fields:
148 siginfo_t {
150 int si_signo; /* Signal number */
151 int si_errno; /* An errno value */
152 int si_code; /* Signal code */
153 int si_trapno; /* Trap number that caused
154 hardware-generated signal
155 (unused on most architectures) */
156 pid_t si_pid; /* Sending process ID */
157 uid_t si_uid; /* Real user ID of sending process */
158 int si_status; /* Exit value or signal */
159 clock_t si_utime; /* User time consumed */
160 clock_t si_stime; /* System time consumed */
161 sigval_t si_value; /* Signal value */
162 int si_int; /* POSIX.1b signal */
163 void *si_ptr; /* POSIX.1b signal */
164 int si_overrun; /* Timer overrun count;
165 POSIX.1b timers */
166 int si_timerid; /* Timer ID; POSIX.1b timers */
167 void *si_addr; /* Memory location which caused fault */
168 long si_band; /* Band event (was int in
169 glibc 2.3.2 and earlier) */
170 int si_fd; /* File descriptor */
171 short si_addr_lsb; /* Least significant bit of address
172 (since Linux 2.6.32) */
173 void *si_lower; /* Lower bound when address violation
174 occurred (since Linux 3.19) */
175 void *si_upper; /* Upper bound when address violation
176 occurred (since Linux 3.19) */
177 int si_pkey; /* Protection key on PTE that caused
178 fault (since Linux 4.6) */
179 void *si_call_addr; /* Address of system call instruction
180 (since Linux 3.5) */
181 int si_syscall; /* Number of attempted system call
182 (since Linux 3.5) */
183 unsigned int si_arch; /* Architecture of attempted system call
184 (since Linux 3.5) */
185 }
186 si_signo, si_errno and si_code are defined for all signals. (si_errno
188 is generally unused on Linux.) The rest of the struct may be a union,
189 so that one should read only the fields that are meaningful for the
190 given signal:
191 * Signals sent with kill(2) and sigqueue(3) fill in si_pid and si_uid.
193 In addition, signals sent with sigqueue(3) fill in si_int and si_ptr
194 with the values specified by the sender of the signal; see
195 sigqueue(3) for more details.
196 * Signals sent by POSIX.1b timers (since Linux 2.6) fill in si_overrun
198 and si_timerid. The si_timerid field is an internal ID used by the
199 kernel to identify the timer; it is not the same as the timer ID
200 returned by timer_create(2). The si_overrun field is the timer over‐
201 run count; this is the same information as is obtained by a call to
202 timer_getoverrun(2). These fields are nonstandard Linux extensions.
203 * Signals sent for message queue notification (see the description of
205 SIGEV_SIGNAL in mq_notify(3)) fill in si_int/si_ptr, with the
206 sigev_value supplied to mq_notify(3); si_pid, with the process ID of
207 the message sender; and si_uid, with the real user ID of the message
208 sender.
209 * SIGCHLD fills in si_pid, si_uid, si_status, si_utime, and si_stime,
211 providing information about the child. The si_pid field is the
212 process ID of the child; si_uid is the child's real user ID. The
213 si_status field contains the exit status of the child (if si_code is
214 CLD_EXITED), or the signal number that caused the process to change
215 state. The si_utime and si_stime contain the user and system CPU
216 time used by the child process; these fields do not include the times
217 used by waited-for children (unlike getrusage(2) and times(2)). In
218 kernels up to 2.6, and since 2.6.27, these fields report CPU time in
219 units of sysconf(_SC_CLK_TCK). In 2.6 kernels before 2.6.27, a bug
220 meant that these fields reported time in units of the (configurable)
221 system jiffy (see time(7)).
222 * SIGILL, SIGFPE, SIGSEGV, SIGBUS, and SIGTRAP fill in si_addr with the
224 address of the fault. On some architectures, these signals also fill
225 in the si_trapno field.
226 Some suberrors of SIGBUS, in particular BUS_MCEERR_AO and
228 BUS_MCEERR_AR, also fill in si_addr_lsb. This field indicates the
229 least significant bit of the reported address and therefore the
230 extent of the corruption. For example, if a full page was corrupted,
231 si_addr_lsb contains log2(sysconf(_SC_PAGESIZE)). When SIGTRAP is
232 delivered in response to a ptrace(2) event (PTRACE_EVENT_foo),
233 si_addr is not populated, but si_pid and si_uid are populated with
234 the respective process ID and user ID responsible for delivering the
235 trap. In the case of seccomp(2), the tracee will be shown as deliv‐
236 ering the event. BUS_MCEERR_* and si_addr_lsb are Linux-specific
237 extensions.
238 The SEGV_BNDERR suberror of SIGSEGV populates si_lower and si_upper.
240 The SEGV_PKUERR suberror of SIGSEGV populates si_pkey.
242 * SIGIO/SIGPOLL (the two names are synonyms on Linux) fills in si_band
244 and si_fd. The si_band event is a bit mask containing the same val‐
245 ues as are filled in the revents field by poll(2). The si_fd field
246 indicates the file descriptor for which the I/O event occurred; for
247 further details, see the description of F_SETSIG in fcntl(2).
248 * SIGSYS, generated (since Linux 3.5) when a seccomp filter returns
250 SECCOMP_RET_TRAP, fills in si_call_addr, si_syscall, si_arch,
251 si_errno, and other fields as described in seccomp(2).
252 The si_code field
254 The si_code field inside the siginfo_t argument that is passed to a
255 SA_SIGINFO signal handler is a value (not a bit mask) indicating why
256 this signal was sent. For a ptrace(2) event, si_code will contain SIG‐
257 TRAP and have the ptrace event in the high byte:
258 (SIGTRAP | PTRACE_EVENT_foo << 8).
260 For a non-ptrace(2) event, the values that can appear in si_code are
262 described in the remainder of this section. Since glibc 2.20, the def‐
263 initions of most of these symbols are obtained from <signal.h> by
264 defining feature test macros (before including any header file) as fol‐
265 lows:
266 * _XOPEN_SOURCE with the value 500 or greater;
268 * _XOPEN_SOURCE and _XOPEN_SOURCE_EXTENDED; or
270 * _POSIX_C_SOURCE with the value 200809L or greater.
272 For the TRAP_* constants, the symbol definitions are provided only in
274 the first two cases. Before glibc 2.20, no feature test macros were
275 required to obtain these symbols.
276 For a regular signal, the following list shows the values which can be
278 placed in si_code for any signal, along with the reason that the signal
279 was generated.
280 SI_USER
282 kill(2).
283 SI_KERNEL
285 Sent by the kernel.
286 SI_QUEUE
288 sigqueue(3).
289 SI_TIMER
291 POSIX timer expired.
292 SI_MESGQ (since Linux 2.6.6)
294 POSIX message queue state changed; see mq_notify(3).
295 SI_ASYNCIO
297 AIO completed.
298 SI_SIGIO
300 Queued SIGIO (only in kernels up to Linux 2.2; from Linux
301 2.4 onward SIGIO/SIGPOLL fills in si_code as described
302 below).
303 SI_TKILL (since Linux 2.4.19)
305 tkill(2) or tgkill(2).
306 The following values can be placed in si_code for a SIGILL signal:
308 ILL_ILLOPC
310 Illegal opcode.
311 ILL_ILLOPN
313 Illegal operand.
314 ILL_ILLADR
316 Illegal addressing mode.
317 ILL_ILLTRP
319 Illegal trap.
320 ILL_PRVOPC
322 Privileged opcode.
323 ILL_PRVREG
325 Privileged register.
326 ILL_COPROC
328 Coprocessor error.
329 ILL_BADSTK
331 Internal stack error.
332 The following values can be placed in si_code for a SIGFPE signal:
334 FPE_INTDIV
336 Integer divide by zero.
337 FPE_INTOVF
339 Integer overflow.
340 FPE_FLTDIV
342 Floating-point divide by zero.
343 FPE_FLTOVF
345 Floating-point overflow.
346 FPE_FLTUND
348 Floating-point underflow.
349 FPE_FLTRES
351 Floating-point inexact result.
352 FPE_FLTINV
354 Floating-point invalid operation.
355 FPE_FLTSUB
357 Subscript out of range.
358 The following values can be placed in si_code for a SIGSEGV signal:
360 SEGV_MAPERR
362 Address not mapped to object.
363 SEGV_ACCERR
365 Invalid permissions for mapped object.
366 SEGV_BNDERR (since Linux 3.19)
368 Failed address bound checks.
369 SEGV_PKUERR (since Linux 4.6)
371 Access was denied by memory protection keys. See pkeys(7).
372 The protection key which applied to this access is available
373 via si_pkey.
374 The following values can be placed in si_code for a SIGBUS signal:
376 BUS_ADRALN
378 Invalid address alignment.
379 BUS_ADRERR
381 Nonexistent physical address.
382 BUS_OBJERR
384 Object-specific hardware error.
385 BUS_MCEERR_AR (since Linux 2.6.32)
387 Hardware memory error consumed on a machine check; action
388 required.
389 BUS_MCEERR_AO (since Linux 2.6.32)
391 Hardware memory error detected in process but not consumed;
392 action optional.
393 The following values can be placed in si_code for a SIGTRAP signal:
395 TRAP_BRKPT
397 Process breakpoint.
398 TRAP_TRACE
400 Process trace trap.
401 TRAP_BRANCH (since Linux 2.4, IA64 only))
403 Process taken branch trap.
404 TRAP_HWBKPT (since Linux 2.4, IA64 only))
406 Hardware breakpoint/watchpoint.
407 The following values can be placed in si_code for a SIGCHLD signal:
409 CLD_EXITED
411 Child has exited.
412 CLD_KILLED
414 Child was killed.
415 CLD_DUMPED
417 Child terminated abnormally.
418 CLD_TRAPPED
420 Traced child has trapped.
421 CLD_STOPPED
423 Child has stopped.
424 CLD_CONTINUED (since Linux 2.6.9)
426 Stopped child has continued.
427 The following values can be placed in si_code for a SIGIO/SIGPOLL sig‐
429 nal:
430 POLL_IN
432 Data input available.
433 POLL_OUT
435 Output buffers available.
436 POLL_MSG
438 Input message available.
439 POLL_ERR
441 I/O error.
442 POLL_PRI
444 High priority input available.
445 POLL_HUP
447 Device disconnected.
448 The following value can be placed in si_code for a SIGSYS signal:
450 SYS_SECCOMP (since Linux 3.5)
452 Triggered by a seccomp(2) filter rule.
453
455 sigaction() returns 0 on success; on error, -1 is returned, and errno
456 is set to indicate the error.
457
459 EFAULT act or oldact points to memory which is not a valid part of the
460 process address space.
461 EINVAL An invalid signal was specified. This will also be generated if
463 an attempt is made to change the action for SIGKILL or SIGSTOP,
464 which cannot be caught or ignored.
465
467 POSIX.1-2001, POSIX.1-2008, SVr4.
468
470 A child created via fork(2) inherits a copy of its parent's signal dis‐
471 positions. During an execve(2), the dispositions of handled signals
472 are reset to the default; the dispositions of ignored signals are left
473 unchanged.
474 According to POSIX, the behavior of a process is undefined after it
476 ignores a SIGFPE, SIGILL, or SIGSEGV signal that was not generated by
477 kill(2) or raise(3). Integer division by zero has undefined result.
478 On some architectures it will generate a SIGFPE signal. (Also dividing
479 the most negative integer by -1 may generate SIGFPE.) Ignoring this
480 signal might lead to an endless loop.
481 POSIX.1-1990 disallowed setting the action for SIGCHLD to SIG_IGN.
483 POSIX.1-2001 and later allow this possibility, so that ignoring SIGCHLD
484 can be used to prevent the creation of zombies (see wait(2)). Never‐
485 theless, the historical BSD and System V behaviors for ignoring SIGCHLD
486 differ, so that the only completely portable method of ensuring that
487 terminated children do not become zombies is to catch the SIGCHLD sig‐
488 nal and perform a wait(2) or similar.
489 POSIX.1-1990 specified only SA_NOCLDSTOP. POSIX.1-2001 added SA_NOCLD‐
491 STOP, SA_NOCLDWAIT, SA_NODEFER, SA_ONSTACK, SA_RESETHAND, SA_RESTART,
492 and SA_SIGINFO. Use of these latter values in sa_flags may be less
493 portable in applications intended for older UNIX implementations.
494 The SA_RESETHAND flag is compatible with the SVr4 flag of the same
496 name.
497 The SA_NODEFER flag is compatible with the SVr4 flag of the same name
499 under kernels 1.3.9 and newer. On older kernels the Linux implementa‐
500 tion allowed the receipt of any signal, not just the one we are
501 installing (effectively overriding any sa_mask settings).
502 sigaction() can be called with a NULL second argument to query the cur‐
504 rent signal handler. It can also be used to check whether a given sig‐
505 nal is valid for the current machine by calling it with NULL second and
506 third arguments.
507 It is not possible to block SIGKILL or SIGSTOP (by specifying them in
509 sa_mask). Attempts to do so are silently ignored.
510 See sigsetops(3) for details on manipulating signal sets.
512 See signal-safety(7) for a list of the async-signal-safe functions that
514 can be safely called inside from inside a signal handler.
515 C library/kernel differences
517 The glibc wrapper function for sigaction() gives an error (EINVAL) on
518 attempts to change the disposition of the two real-time signals used
519 internally by the NPTL threading implementation. See nptl(7) for
520 details.
521 On architectures where the signal trampoline resides in the C library,
523 the glibc wrapper function for sigaction() places the address of the
524 trampoline code in the act.sa_restorer field and sets the SA_RESTORER
525 flag in the act.sa_flags field. See sigreturn(2).
526 The original Linux system call was named sigaction(). However, with
528 the addition of real-time signals in Linux 2.2, the fixed-size, 32-bit
529 sigset_t type supported by that system call was no longer fit for pur‐
530 pose. Consequently, a new system call, rt_sigaction(), was added to
531 support an enlarged sigset_t type. The new system call takes a fourth
532 argument, size_t sigsetsize, which specifies the size in bytes of the
533 signal sets in act.sa_mask and oldact.sa_mask. This argument is cur‐
534 rently required to have the value sizeof(sigset_t) (or the error EINVAL
535 results). The glibc sigaction() wrapper function hides these details
536 from us, transparently calling rt_sigaction() when the kernel provides
537 it.
538 Undocumented
540 Before the introduction of SA_SIGINFO, it was also possible to get some
541 additional information about the signal. This was done by providing an
542 sa_handler signal handler with a second argument of type struct sigcon‐
543 text, which is the same structure as the one that is passed in the
544 uc_mcontext field of the ucontext structure that is passed (via a
545 pointer) in the third argument of the sa_sigaction handler. See the
546 relevant Linux kernel sources for details. This use is obsolete now.
547
549 When delivering a signal with a SA_SIGINFO handler, the kernel does not
550 always provide meaningful values for all of the fields of the siginfo_t
551 that are relevant for that signal.
552 In kernels up to and including 2.6.13, specifying SA_NODEFER in
554 sa_flags prevents not only the delivered signal from being masked dur‐
555 ing execution of the handler, but also the signals specified in
556 sa_mask. This bug was fixed in kernel 2.6.14.
557
559 See mprotect(2).
560
562 kill(1), kill(2), pause(2), pidfd_send_signal(2), restart_syscall(2),
563 seccomp(2), sigaltstack(2), signal(2), signalfd(2), sigpending(2), sig‐
564 procmask(2), sigreturn(2), sigsuspend(2), wait(2), killpg(3), raise(3),
565 siginterrupt(3), sigqueue(3), sigsetops(3), sigvec(3), core(5), sig‐
566 nal(7)
567
569 This page is part of release 5.07 of the Linux man-pages project. A
570 description of the project, information about reporting bugs, and the
571 latest version of this page, can be found at
572 https://www.kernel.org/doc/man-pages/.
573Linux 2020-04-11 SIGACTION(2)