sigset(3)

πŸ’– sigset(3) β€” the System V signal API, bestie edition πŸ’–

okay so. this is the man page for sigset, sighold, sigrelse and sigignore, and it lives in section 3 of the manual (that's the (3) on the end of sigset(3), mwah). fair warning before we even start: this whole squad is old old, like, they're the System V signal era, and they're marked deprecated with a capital D. but you asked, so let's get into it 🫢


✨ NAME ✨

sigset, sighold, sigrelse, sigignore β€” the System V signal API

πŸ“š LIBRARY πŸ“š

Standard C library (libc, -lc)

πŸŽ€ SYNOPSIS πŸŽ€

#include <signal.h>

typedef typeof(void (int))  *sighandler_t;

[[deprecated]] sighandler_t sigset(int sig, sighandler_t disp);

[[deprecated]] int sighold(int sig);
[[deprecated]] int sigrelse(int sig);
[[deprecated]] int sigignore(int sig);

🌸 Feature Test Macro Requirements for glibc (see feature_test_macros(7)) 🌸

  • sigset(), sighold(), sigrelse(), sigignore():
    • _XOPEN_SOURCE >= 500

πŸ’… DESCRIPTION πŸ’…

so glibc keeps these around purely as a compatibility interface for programs that use the historical System V signal API. like, it's there so your fossil program still runs, not because anyone wants you to write new stuff with it. this API is obsolete β€” new applications should be using the POSIX signal API instead: sigaction(2), sigprocmask(2), etc. no cap, that's literally what the page says.

sigset() modifies the disposition of the signal sig. the disp argument can be the address of a signal handler function, or one of these iconic constants:

  • SIG_DFL β€” reset the disposition of sig to the default. back to factory settings, we love a fresh start πŸ’«
  • SIG_IGN β€” ignore sig. not me pretending i don't see it πŸ™ˆ
  • SIG_HOLD β€” add sig to the process's signal mask, but leave the disposition of sig unchanged. so it's held, not changed, and those are two different things bestie!!

if disp specifies the address of a signal handler, then sig is added to the process's signal mask during execution of the handler.

if disp was specified as a value other than SIG_HOLD, then sig is removed from the process's signal mask.

the dispositions for SIGKILL and SIGSTOP cannot be changed. periodt. don't even try it πŸ›‘

sighold() adds sig to the calling process's signal mask.

sigrelse() removes sig from the calling process's signal mask.

sigignore() sets the disposition of sig to SIG_IGN.


πŸ”₯ RETURN VALUE πŸ”₯

on success, sigset() returns SIG_HOLD if sig was blocked before the call, or the signal's previous disposition if it was not blocked before the call. on error, sigset() returns -1, with errno set to indicate the error. (but bestie… see BUGS below. πŸ‘€)

sighold(), sigrelse(), and sigignore() return 0 on success; on error, these functions return -1 and set errno to indicate the error.

😭 ERRORS 😭

πŸ’œ ATTRIBUTES πŸ’œ

for an explanation of the terms used in this section, see attributes(7).

  • sigset(), sighold(), sigrelse(), sigignore() β€” Thread safety: MT-Safe

(that's the whole table, she's small but she's mighty 🧚)

πŸ‘‘ STANDARDS πŸ‘‘

POSIX.1-2008.

sighandler_t β€” GNU. POSIX.1 uses the same type but without a typedef.

🌈 HISTORY 🌈

glibc 2.1. SVr4, POSIX.1-2001. POSIX.1-2008 marks these functions as obsolete, recommending the use of sigaction(2), sigprocmask(2), pthread_sigmask(3), and sigsuspend(2) instead. the council has spoken πŸ“œ

πŸ¦‹ NOTES πŸ¦‹

the sigset() function provides reliable signal handling semantics (as when calling sigaction(2) with sa_mask equal to 0).

on System V, the signal() function provides unreliable semantics (as when calling sigaction(2) with sa_mask equal to SA_RESETHAND | SA_NODEFER). on BSD, signal() provides reliable semantics. POSIX.1-2001 leaves these aspects of signal() unspecified. see signal(2) for further details.

in order to wait for a signal, BSD and System V both provided a function named sigpause(3), but this function has a different argument on the two systems. see sigpause(3) for details. same name, different vibe, chaos honestly πŸ˜΅πŸ’«

πŸ’€ BUGS πŸ’€

before glibc 2.2, sigset() did not unblock sig if disp was specified as a value other than SIG_HOLD.

before glibc 2.5, sigset() does not correctly return the previous disposition of the signal in two cases. first, if disp is specified as SIG_HOLD, then a successful sigset() always returns SIG_HOLD. instead, it should return the previous disposition of the signal (unless the signal was blocked, in which case SIG_HOLD should be returned). second, if the signal is currently blocked, then the return value of a successful sigset() should be SIG_HOLD. instead, the previous disposition of the signal is returned. these problems have been fixed since glibc 2.5. so if you're on anything modern you're fine, but if you're time-travelling, beware ⚠️

🌟 SEE ALSO 🌟

kill(2), pause(2), sigaction(2), signal(2), sigprocmask(2), raise(3), sigpause(3), sigvec(3), signal(7)


okay that's the whole page, you ate that πŸ“ this was sigset(3) from Linux man-pages 6.20, dated 2026-02-08 β€” go forth, use sigaction(2) instead, and slay πŸ’…βœ¨

πŸ“¦ where this page comes from
  • arch rolling: man-pages 6.20-1 Β· BSD-2-Clause AND BSD-3-Clause AND BSD-4.3TAHOE AND BSD-4-Clause-UC AND GPL-1.0-or-later AND GPL-2.0-only AND GPL-2.0-or-later AND GPL-3.0-or-later AND LGPL-3.0-linking-exception AND LGPL-3.0-or-later AND LicenseRef-Public-Domain AND LicenseRef-UltraPermissive AND Linux-man-pages-1-para AND Linux-man-pages-copyleft AND Linux-man-pages-copyleft-2-para AND Linux-man-pages-copyleft-var AND MIT AND Spencer-94
  • nixpkgs unstable: man-pages 6.20
  • πŸ“œ the page's license: Linux-man-pages-copyleft
  • πŸ’– girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms