freebsd-dev/lib/libc/sys/sigaction.2
Jilles Tjoelker b30cd8df7c sigaction(2),sigwait(2),sigwaitinfo(2): Remove [EFAULT] error condition.
Passing an invalid pointer results in undefined behaviour.

The wrappers in libthr access some of the data pointed to by the arguments
in userland, so that an invalid pointer will cause a signal and not an
[EFAULT] error return.

Furthermore, if the [EFAULT] error occurs when the kernel is writing, it is
not a proper error in the sense that the call still commits (changing the
signal disposition or accepting the signal).

MFC after:	1 week
2012-09-27 17:48:04 +00:00

660 lines
17 KiB
Groff

.\" Copyright (c) 1980, 1990, 1993
.\" The Regents of the University of California. All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\" notice, this list of conditions and the following disclaimer in the
.\" documentation and/or other materials provided with the distribution.
.\" 4. Neither the name of the University nor the names of its contributors
.\" may be used to endorse or promote products derived from this software
.\" without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" From: @(#)sigaction.2 8.2 (Berkeley) 4/3/94
.\" $FreeBSD$
.\"
.Dd September 27, 2012
.Dt SIGACTION 2
.Os
.Sh NAME
.Nm sigaction
.Nd software signal facilities
.Sh LIBRARY
.Lb libc
.Sh SYNOPSIS
.In signal.h
.Bd -literal
struct sigaction {
void (*sa_handler)(int);
void (*sa_sigaction)(int, siginfo_t *, void *);
int sa_flags; /* see signal options below */
sigset_t sa_mask; /* signal mask to apply */
};
.Ed
.Ft int
.Fo sigaction
.Fa "int sig"
.Fa "const struct sigaction * restrict act"
.Fa "struct sigaction * restrict oact"
.Fc
.Sh DESCRIPTION
The system defines a set of signals that may be delivered to a process.
Signal delivery resembles the occurrence of a hardware interrupt:
the signal is normally blocked from further occurrence, the current process
context is saved, and a new one is built.
A process may specify a
.Em handler
to which a signal is delivered, or specify that a signal is to be
.Em ignored .
A process may also specify that a default action is to be taken
by the system when a signal occurs.
A signal may also be
.Em blocked ,
in which case its delivery is postponed until it is
.Em unblocked .
The action to be taken on delivery is determined at the time
of delivery.
Normally, signal handlers execute on the current stack
of the process.
This may be changed, on a per-handler basis,
so that signals are taken on a special
.Em "signal stack" .
.Pp
Signal routines normally execute with the signal that caused their
invocation
.Em blocked ,
but other signals may yet occur.
A global
.Em "signal mask"
defines the set of signals currently blocked from delivery
to a process.
The signal mask for a process is initialized
from that of its parent (normally empty).
It may be changed with a
.Xr sigprocmask 2
call, or when a signal is delivered to the process.
.Pp
When a signal
condition arises for a process, the signal is added to a set of
signals pending for the process.
If the signal is not currently
.Em blocked
by the process then it is delivered to the process.
Signals may be delivered any time a process enters the operating system
(e.g., during a system call, page fault or trap, or clock interrupt).
If multiple signals are ready to be delivered at the same time,
any signals that could be caused by traps are delivered first.
Additional signals may be processed at the same time, with each
appearing to interrupt the handlers for the previous signals
before their first instructions.
The set of pending signals is returned by the
.Xr sigpending 2
system call.
When a caught signal
is delivered, the current state of the process is saved,
a new signal mask is calculated (as described below),
and the signal handler is invoked.
The call to the handler
is arranged so that if the signal handling routine returns
normally the process will resume execution in the context
from before the signal's delivery.
If the process wishes to resume in a different context, then it
must arrange to restore the previous context itself.
.Pp
When a signal is delivered to a process a new signal mask is
installed for the duration of the process' signal handler
(or until a
.Xr sigprocmask 2
system call is made).
This mask is formed by taking the union of the current signal mask set,
the signal to be delivered, and
the signal mask associated with the handler to be invoked.
.Pp
The
.Fn sigaction
system call
assigns an action for a signal specified by
.Fa sig .
If
.Fa act
is non-zero, it
specifies an action
.Dv ( SIG_DFL ,
.Dv SIG_IGN ,
or a handler routine) and mask
to be used when delivering the specified signal.
If
.Fa oact
is non-zero, the previous handling information for the signal
is returned to the user.
.Pp
The above declaration of
.Vt "struct sigaction"
is not literal.
It is provided only to list the accessible members.
See
.In sys/signal.h
for the actual definition.
In particular, the storage occupied by sa_handler and sa_sigaction overlaps,
and an application can not use both simultaneously.
.Pp
Once a signal handler is installed, it normally remains installed
until another
.Fn sigaction
system call is made, or an
.Xr execve 2
is performed.
A signal-specific default action may be reset by
setting
.Va sa_handler
to
.Dv SIG_DFL .
The defaults are process termination, possibly with core dump;
no action; stopping the process; or continuing the process.
See the signal list below for each signal's default action.
If
.Va sa_handler
is
.Dv SIG_DFL ,
the default action for the signal is to discard the signal,
and if a signal is pending,
the pending signal is discarded even if the signal is masked.
If
.Va sa_handler
is set to
.Dv SIG_IGN
current and pending instances
of the signal are ignored and discarded.
.Pp
Options may be specified by setting
.Va sa_flags .
The meaning of the various bits is as follows:
.Bl -tag -offset indent -width SA_RESETHANDXX
.It Dv SA_NOCLDSTOP
If this bit is set when installing a catching function
for the
.Dv SIGCHLD
signal,
the
.Dv SIGCHLD
signal will be generated only when a child process exits,
not when a child process stops.
.It Dv SA_NOCLDWAIT
If this bit is set when calling
.Fn sigaction
for the
.Dv SIGCHLD
signal, the system will not create zombie processes when children of
the calling process exit.
If the calling process subsequently issues a
.Xr wait 2
(or equivalent), it blocks until all of the calling process's child
processes terminate, and then returns a value of \-1 with
.Va errno
set to
.Er ECHILD .
The same effect of avoiding zombie creation can also be achieved by setting
.Va sa_handler
for
.Dv SIGCHLD
to
.Dv SIG_IGN .
.It Dv SA_ONSTACK
If this bit is set, the system will deliver the signal to the process
on a
.Em "signal stack" ,
specified with
.Xr sigaltstack 2 .
.It Dv SA_NODEFER
If this bit is set, further occurrences of the delivered signal are
not masked during the execution of the handler.
.It Dv SA_RESETHAND
If this bit is set, the handler is reset back to
.Dv SIG_DFL
at the moment the signal is delivered.
.It Dv SA_RESTART
See paragraph below.
.It Dv SA_SIGINFO
If this bit is set, the handler function is assumed to be pointed to by the
.Va sa_sigaction
member of
.Vt "struct sigaction"
and should match the prototype shown above or as below in
.Sx EXAMPLES .
This bit should not be set when assigning
.Dv SIG_DFL
or
.Dv SIG_IGN .
.El
.Pp
If a signal is caught during the system calls listed below,
the call may be forced to terminate
with the error
.Er EINTR ,
the call may return with a data transfer shorter than requested,
or the call may be restarted.
Restart of pending calls is requested
by setting the
.Dv SA_RESTART
bit in
.Va sa_flags .
The affected system calls include
.Xr open 2 ,
.Xr read 2 ,
.Xr write 2 ,
.Xr sendto 2 ,
.Xr recvfrom 2 ,
.Xr sendmsg 2
and
.Xr recvmsg 2
on a communications channel or a slow device (such as a terminal,
but not a regular file)
and during a
.Xr wait 2
or
.Xr ioctl 2 .
However, calls that have already committed are not restarted,
but instead return a partial success (for example, a short read count).
.Pp
After a
.Xr fork 2
or
.Xr vfork 2
all signals, the signal mask, the signal stack,
and the restart/interrupt flags are inherited by the child.
.Pp
The
.Xr execve 2
system call reinstates the default
action for all signals which were caught and
resets all signals to be caught on the user stack.
Ignored signals remain ignored;
the signal mask remains the same;
signals that restart pending system calls continue to do so.
.Pp
The following is a list of all signals
with names as in the include file
.In signal.h :
.Bl -column SIGVTALARMXX "create core imagexxx"
.It Sy "NAME Default Action Description"
.It Dv SIGHUP No " terminate process" " terminal line hangup"
.It Dv SIGINT No " terminate process" " interrupt program"
.It Dv SIGQUIT No " create core image" " quit program"
.It Dv SIGILL No " create core image" " illegal instruction"
.It Dv SIGTRAP No " create core image" " trace trap"
.It Dv SIGABRT No " create core image" Ta Xr abort 3
call (formerly
.Dv SIGIOT )
.It Dv SIGEMT No " create core image" " emulate instruction executed"
.It Dv SIGFPE No " create core image" " floating-point exception"
.It Dv SIGKILL No " terminate process" " kill program"
.It Dv SIGBUS No " create core image" " bus error"
.It Dv SIGSEGV No " create core image" " segmentation violation"
.It Dv SIGSYS No " create core image" " non-existent system call invoked"
.It Dv SIGPIPE No " terminate process" " write on a pipe with no reader"
.It Dv SIGALRM No " terminate process" " real-time timer expired"
.It Dv SIGTERM No " terminate process" " software termination signal"
.It Dv SIGURG No " discard signal" " urgent condition present on socket"
.It Dv SIGSTOP No " stop process" " stop (cannot be caught or ignored)"
.It Dv SIGTSTP No " stop process" " stop signal generated from keyboard"
.It Dv SIGCONT No " discard signal" " continue after stop"
.It Dv SIGCHLD No " discard signal" " child status has changed"
.It Dv SIGTTIN No " stop process" " background read attempted from control terminal"
.It Dv SIGTTOU No " stop process" " background write attempted to control terminal"
.It Dv SIGIO No " discard signal" Tn " I/O"
is possible on a descriptor (see
.Xr fcntl 2 )
.It Dv SIGXCPU No " terminate process" " cpu time limit exceeded (see"
.Xr setrlimit 2 )
.It Dv SIGXFSZ No " terminate process" " file size limit exceeded (see"
.Xr setrlimit 2 )
.It Dv SIGVTALRM No " terminate process" " virtual time alarm (see"
.Xr setitimer 2 )
.It Dv SIGPROF No " terminate process" " profiling timer alarm (see"
.Xr setitimer 2 )
.It Dv SIGWINCH No " discard signal" " Window size change"
.It Dv SIGINFO No " discard signal" " status request from keyboard"
.It Dv SIGUSR1 No " terminate process" " User defined signal 1"
.It Dv SIGUSR2 No " terminate process" " User defined signal 2"
.El
.Sh NOTE
The
.Va sa_mask
field specified in
.Fa act
is not allowed to block
.Dv SIGKILL
or
.Dv SIGSTOP .
Any attempt to do so will be silently ignored.
.Pp
The following functions are either reentrant or not interruptible
by signals and are async-signal safe.
Therefore applications may
invoke them, without restriction, from signal-catching functions:
.Pp
Base Interfaces:
.Pp
.Fn _exit ,
.Fn access ,
.Fn alarm ,
.Fn cfgetispeed ,
.Fn cfgetospeed ,
.Fn cfsetispeed ,
.Fn cfsetospeed ,
.Fn chdir ,
.Fn chmod ,
.Fn chown ,
.Fn close ,
.Fn creat ,
.Fn dup ,
.Fn dup2 ,
.Fn execle ,
.Fn execve ,
.Fn fcntl ,
.Fn fork ,
.Fn fpathconf ,
.Fn fstat ,
.Fn fsync ,
.Fn getegid ,
.Fn geteuid ,
.Fn getgid ,
.Fn getgroups ,
.Fn getpgrp ,
.Fn getpid ,
.Fn getppid ,
.Fn getuid ,
.Fn kill ,
.Fn link ,
.Fn lseek ,
.Fn mkdir ,
.Fn mkfifo ,
.Fn open ,
.Fn pathconf ,
.Fn pause ,
.Fn pipe ,
.Fn raise ,
.Fn read ,
.Fn rename ,
.Fn rmdir ,
.Fn setgid ,
.Fn setpgid ,
.Fn setsid ,
.Fn setuid ,
.Fn sigaction ,
.Fn sigaddset ,
.Fn sigdelset ,
.Fn sigemptyset ,
.Fn sigfillset ,
.Fn sigismember ,
.Fn signal ,
.Fn sigpending ,
.Fn sigprocmask ,
.Fn sigsuspend ,
.Fn sleep ,
.Fn stat ,
.Fn sysconf ,
.Fn tcdrain ,
.Fn tcflow ,
.Fn tcflush ,
.Fn tcgetattr ,
.Fn tcgetpgrp ,
.Fn tcsendbreak ,
.Fn tcsetattr ,
.Fn tcsetpgrp ,
.Fn time ,
.Fn times ,
.Fn umask ,
.Fn uname ,
.Fn unlink ,
.Fn utime ,
.Fn wait ,
.Fn waitpid ,
.Fn write .
.Pp
Realtime Interfaces:
.Pp
.Fn aio_error ,
.Fn clock_gettime ,
.Fn sigpause ,
.Fn timer_getoverrun ,
.Fn aio_return ,
.Fn fdatasync ,
.Fn sigqueue ,
.Fn timer_gettime ,
.Fn aio_suspend ,
.Fn sem_post ,
.Fn sigset ,
.Fn timer_settime .
.Pp
.Tn ANSI C
Interfaces:
.Pp
.Fn strcpy ,
.Fn strcat ,
.Fn strncpy ,
.Fn strncat ,
and perhaps some others.
.Pp
Extension Interfaces:
.Pp
.Fn strlcpy ,
.Fn strlcat .
.Pp
All functions not in the above lists are considered to be unsafe
with respect to signals.
That is to say, the behaviour of such
functions when called from a signal handler is undefined.
In general though, signal handlers should do little more than set a
flag; most other actions are not safe.
.Pp
Also, it is good practice to make a copy of the global variable
.Va errno
and restore it before returning from the signal handler.
This protects against the side effect of
.Va errno
being set by functions called from inside the signal handler.
.Sh RETURN VALUES
.Rv -std sigaction
.Sh EXAMPLES
There are three possible prototypes the handler may match:
.Bl -tag -offset indent -width short
.It Tn ANSI C :
.Ft void
.Fn handler int ;
.It Traditional BSD style:
.Ft void
.Fn handler int "int code" "struct sigcontext *scp" ;
.It Tn POSIX Dv SA_SIGINFO :
.Ft void
.Fn handler int "siginfo_t *info" "ucontext_t *uap" ;
.El
.Pp
The handler function should match the
.Dv SA_SIGINFO
prototype if the
.Dv SA_SIGINFO
bit is set in
.Va sa_flags .
It then should be pointed to by the
.Va sa_sigaction
member of
.Vt "struct sigaction" .
Note that you should not assign
.Dv SIG_DFL
or
.Dv SIG_IGN
this way.
.Pp
If the
.Dv SA_SIGINFO
flag is not set, the handler function should match
either the
.Tn ANSI C
or traditional
.Bx
prototype and be pointed to by
the
.Va sa_handler
member of
.Vt "struct sigaction" .
In practice,
.Fx
always sends the three arguments of the latter and since the
.Tn ANSI C
prototype is a subset, both will work.
The
.Va sa_handler
member declaration in
.Fx
include files is that of
.Tn ANSI C
(as required by
.Tn POSIX ) ,
so a function pointer of a
.Bx Ns -style
function needs to be casted to
compile without warning.
The traditional
.Bx
style is not portable and since its capabilities
are a full subset of a
.Dv SA_SIGINFO
handler,
its use is deprecated.
.Pp
The
.Fa sig
argument is the signal number, one of the
.Dv SIG...
values from
.In signal.h .
.Pp
The
.Fa code
argument of the
.Bx Ns -style
handler and the
.Va si_code
member of the
.Fa info
argument to a
.Dv SA_SIGINFO
handler contain a numeric code explaining the
cause of the signal, usually one of the
.Dv SI_...
values from
.In sys/signal.h
or codes specific to a signal, i.e., one of the
.Dv FPE_...
values for
.Dv SIGFPE .
.Pp
The
.Fa scp
argument to a
.Bx Ns -style
handler points to an instance of
.Vt "struct sigcontext" .
.Pp
The
.Fa uap
argument to a
.Tn POSIX
.Dv SA_SIGINFO
handler points to an instance of
ucontext_t.
.Sh ERRORS
The
.Fn sigaction
system call
will fail and no new signal handler will be installed if one
of the following occurs:
.Bl -tag -width Er
.It Bq Er EINVAL
The
.Fa sig
argument
is not a valid signal number.
.It Bq Er EINVAL
An attempt is made to ignore or supply a handler for
.Dv SIGKILL
or
.Dv SIGSTOP .
.El
.Sh SEE ALSO
.Xr kill 1 ,
.Xr kill 2 ,
.Xr ptrace 2 ,
.Xr sigaltstack 2 ,
.Xr sigpending 2 ,
.Xr sigprocmask 2 ,
.Xr sigsuspend 2 ,
.Xr wait 2 ,
.Xr fpsetmask 3 ,
.Xr setjmp 3 ,
.Xr siginfo 3 ,
.Xr siginterrupt 3 ,
.Xr sigsetops 3 ,
.Xr ucontext 3 ,
.Xr tty 4
.Sh STANDARDS
The
.Fn sigaction
system call is expected to conform to
.St -p1003.1-90 .
The
.Dv SA_ONSTACK
and
.Dv SA_RESTART
flags are Berkeley extensions,
as are the signals,
.Dv SIGTRAP ,
.Dv SIGEMT ,
.Dv SIGBUS ,
.Dv SIGSYS ,
.Dv SIGURG ,
.Dv SIGIO ,
.Dv SIGXCPU ,
.Dv SIGXFSZ ,
.Dv SIGVTALRM ,
.Dv SIGPROF ,
.Dv SIGWINCH ,
and
.Dv SIGINFO .
Those signals are available on most
.Bx Ns \-derived
systems.
The
.Dv SA_NODEFER
and
.Dv SA_RESETHAND
flags are intended for backwards compatibility with other operating
systems.
The
.Dv SA_NOCLDSTOP ,
and
.Dv SA_NOCLDWAIT
.\" and
.\" SA_SIGINFO
flags are featuring options commonly found in other operating systems.
The flags are approved by
.St -susv2 ,
along with the option to avoid zombie creation by ignoring
.Dv SIGCHLD .