![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to signal.2 CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Research Unix] / researchv10dc / man / adm / man2|
Return to signal.2 CVS log | Up to [Research Unix] / researchv10dc / man / adm / man2 |
1.1 root 1: .TH SIGNAL 2
2: .CT 2 proc_man
3: .SH NAME
4: signal, kill \(mi receive and send signals
5: .SH SYNOPSIS
6: .nf
7: .B #include <signal.h>
8: .PP
9: .B SIG_TYP (*signal(sig, func))()
10: .B SIG_TYP (*func)();
11: .PP
12: .B int kill(pid, sig)
13: .fi
14: .SH DESCRIPTION
15: A signal
16: is generated by some abnormal event
17: initiated by a user at a terminal (quit, interrupt),
18: by a program error (bus error, etc.),
19: or by
20: .I kill
21: in another process.
22: Normally, most signals
23: cause termination of the receiving process,
24: but
25: .I signal
26: allows them either to be ignored
27: or to be caught by interrupting to a specified function.
28: The following signal names are defined in
29: .FR <signal.h> :
30: .LP
31: .nf
32: .ta \w'SIGMMMM 'u +\w'15* 'u
33: \fLSIGHUP\fP 1 hangup
34: \fLSIGINT\fP 2 interrupt
35: \fLSIGQUIT\fP 3* quit
36: \fLSIGILL\fP 4* illegal instruction (not reset when caught)
37: \fLSIGTRAP\fP 5* trace trap (not reset when caught)
38: \fLSIGIOT\fP 6* IOT instruction
39: \fLSIGEMT\fP 7* EMT instruction
40: \fLSIGFPE\fP 8* floating point exception
41: \fLSIGKILL\fP 9 kill (cannot be caught or ignored)
42: \fLSIGBUS\fP 10* bus error
43: \fLSIGSEGV\fP 11* segmentation violation
44: \fLSIGSYS\fP 12* bad argument to system call
45: \fLSIGPIPE\fP 13 write on a pipe with no one to read it
46: \fLSIGALRM\fP 14 alarm clock
47: \fLSIGTERM\fP 15 software termination signal
48: 16 unassigned
49: \fLSIGSTOP\fP 17+ stop (cannot be caught or ignored)
50: \fLSIGCONT\fP 19# continue a stopped process
51: \fLSIGCHLD\fP 20# child has stopped or exited
52: .sp
53: .fi
54: * places core image in file
55: .B core
56: if not caught or ignored
57: .br
58: + suspends process until
59: .B SIGCONT
60: or
61: .BR PIOCRUN ;
62: see
63: .IR proc (4)
64: .br
65: # ignored if not caught
66: .PP
67: Signals 1 through
68: .BR NSIG -1,
69: defined in the include file, exist.
70: Those not listed above have
71: no conventional meaning in this system.
72: (Berkeley systems use 1-15 and 17-25.)
73: .PP
74: .I Signal
75: specifies how signal
76: .I sig
77: will be handled.
78: If
79: .I func
80: is
81: .BR SIG_DFL ,
82: the default action listed above is reinstated.
83: If
84: .I func
85: is
86: .BR SIG_IGN ,
87: the signal will be ignored.
88: Otherwise the signal will be caught; when the signal occurs,
89: the function pointed to by
90: .IR func ,
91: say
92: .IR catcher (),
93: defined thus,
94: .EX
95: .L
96: int catcher(sig) { ... }
97: .EE
98: will be called with the
99: signal number as argument.
100: A return from the function will
101: continue the process at the point it was interrupted.
102: .PP
103: Except as indicated, a signal is reset to
104: .B SIG_DFL
105: after being caught.
106: Thus if it is desired to catch every such signal,
107: the catching routine must issue another
108: .I signal
109: call.
110: .PP
111: When a caught signal occurs
112: during certain system calls, the call terminates prematurely.
113: In particular this can occur during
114: .IR read (2)
115: or
116: .IR write
117: on a slow device (like a typewriter, but not a disk),
118: and during
119: .IR pause
120: and
121: .IR wait ;
122: see
123: .IR alarm (2)
124: and
125: .IR exit (2).
126: The interrupted system call will return error
127: .BR EINTR .
128: The user's program may then, if it wishes, re-execute the call.
129: .PP
130: .I Signal
131: returns the previous (or initial)
132: value of
133: .I func
134: for the particular signal.
135: .PP
136: After a
137: .IR fork (2)
138: the child inherits all signal settings.
139: .IR Exec (2)
140: resets all caught signals to default action.
141: .PP
142: .I Kill
143: sends signal
144: .I sig
145: to the process specified by process id
146: .I pid.
147: Signal 0
148: has no effect on the target process and may be used to
149: test the existence of a process.
150: The success of sending a signal is independent of how the receiving
151: process treats the signal.
152: .PP
153: The effective userid of the sending process must be either 0
154: or the effective userid of the receiving process.
155: .PP
156: If
157: .I pid
158: is 0, the signal is sent to all other processes in the
159: sender's process group; see
160: .IR stream (4).
161: .PP
162: If
163: .I pid
164: is \-1, and the user is the super-user,
165: the signal is broadcast universally
166: except to processes 0 (scheduler),
167: 1 (initialization)
168: and 2 (pageout); see
169: .IR init (8).
170: If
171: .I pid
172: is less than \-1,
173: it is negated
174: and taken as a process group
175: whose members should receive the signal.
176: .PP
177: Processes may send signals to themselves.
178: .SH FILES
179: .F core
180: .SH "SEE ALSO"
181: .IR kill (1),
182: .IR setjmp (3),
183: .IR stream (4)
184: .SH DIAGNOSTICS
185: .IR signal :
186: .B EINVAL
187: .br
188: .IR kill :
189: .BR EINVAL ,
190: .BR EPERM ,
191: .BR ESRCH
192: .SH BUGS
193: The reason for a trap should be distinguishable by extra arguments
194: to the signal handler.
195: .br
196: If a repeated signal arrives before the last one can be reset,
197: there is no chance to catch it.
198: .br
199: For historical reasons, the return value of
200: a catcher function is
201: .BR int ;
202: it is
203: .B void
204: in
205: .SM ANSI
206: standard C.
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.