xref: /freebsd/lib/libsys/rfork.2 (revision 8269e7673cf033aba67dab8264fe719920c70f87)

.\"
.\" This manual page is taken directly from Plan9, and modified to
.\" describe the actual BSD implementation. Permission for
.\" use of this page comes from Rob Pike <rob@plan9.att.com>.
.\"
.Dd September 25, 2019
.Dt RFORK 2
.Os
.Sh NAME
.Nm rfork
.Nd manipulate process resources
.Sh LIBRARY
.Lb libc
.Sh SYNOPSIS
.In unistd.h
.Ft pid_t
.Fn rfork "int flags"
.Sh DESCRIPTION
Forking, vforking or rforking are the only ways new processes are created.
The
.Fa flags
argument to
.Fn rfork
selects which resources of the
invoking process (parent) are shared
by the new process (child) or initialized to
their default values.
The resources include
the open file descriptor table (which, when shared, permits processes
to open and close files for other processes),
and open files.
The
.Fa flags
argument
is either
.Dv RFSPAWN
or the logical OR of some subset of:
.Bl -tag -width ".Dv RFLINUXTHPN"
.It Dv RFPROC
If set a new process is created; otherwise changes affect the
current process.
.It Dv RFNOWAIT
If set, the child process will be dissociated from the parent.
Upon
exit the child will not leave a status for the parent to collect.
See
.Xr wait 2 .
.It Dv RFFDG
If set, the invoker's file descriptor table (see
.Xr intro 2 )
is copied; otherwise the two processes share a
single table.
.It Dv RFCFDG
If set, the new process starts with a clean file descriptor table.
Is mutually exclusive with
.Dv RFFDG .
.It Dv RFTHREAD
If set, the new process shares file descriptor to process leaders table
with its parent.
Only applies when neither
.Dv RFFDG
nor
.Dv RFCFDG
are set.
.It Dv RFMEM
If set, the kernel will force sharing of the entire address space,
typically by sharing the hardware page table directly.
The child
will thus inherit and share all the segments the parent process owns,
whether they are normally shareable or not.
The stack segment is
not split (both the parent and child return on the same stack) and thus
.Fn rfork
with the RFMEM flag may not generally be called directly from high level
languages including C.
May be set only with
.Dv RFPROC .
A helper function is provided to assist with this problem and will cause
the new process to run on the provided stack.
See
.Xr rfork_thread 3
for information.
Note that a lot of code will not run correctly in such an environment.
.It Dv RFSIGSHARE
If set, the kernel will force sharing the sigacts structure between the
child and the parent.
.It Dv RFTSIGZMB
If set, the kernel will deliver a specified signal to the parent
upon the child exit, instead of default SIGCHLD.
The signal number
.Dv signum
is specified by oring the
.Dv RFTSIGFLAGS(signum)
expression into
.Fa flags .
Specifying signal number 0 disables signal delivery upon the child exit.
.It Dv RFLINUXTHPN
If set, the kernel will deliver SIGUSR1 instead of SIGCHLD upon thread
exit for the child.
This is intended to mimic certain Linux clone behaviour.
.El
.Pp
File descriptors in a shared file descriptor table are kept
open until either they are explicitly closed
or all processes sharing the table exit.
.Pp
If
.Dv RFSPAWN
is passed,
.Nm
will use
.Xr vfork 2
semantics but reset all signal actions in the child to default.
This flag is used by the
.Xr posix_spawn 3
implementation in libc.
.Pp
If
.Dv RFPROC
is set, the
value returned in the parent process
is the process id
of the child process; the value returned in the child is zero.
Without
.Dv RFPROC ,
the return value is zero.
Process id's range from 1 to the maximum integer
.Ft ( int )
value.
The
.Fn rfork
system call
will sleep, if necessary, until required process resources are available.
.Pp
The
.Fn fork
system call
can be implemented as a call to
.Fn rfork "RFFDG | RFPROC"
but is not for backwards compatibility.
.Sh RETURN VALUES
Upon successful completion,
.Fn rfork
returns a value
of 0 to the child process and returns the process ID of the child
process to the parent process.
Otherwise, a value of -1 is returned
to the parent process, no child process is created, and the global
variable
.Va errno
is set to indicate the error.
.Sh ERRORS
The
.Fn rfork
system call
will fail and no child process will be created if:
.Bl -tag -width Er
.It Bq Er EAGAIN
The system-imposed limit on the total
number of processes under execution would be exceeded.
The limit is given by the
.Xr sysctl 3
MIB variable
.Dv KERN_MAXPROC .
(The limit is actually ten less than this
except for the super user).
.It Bq Er EAGAIN
The user is not the super user, and
the system-imposed limit
on the total number of
processes under execution by a single user would be exceeded.
The limit is given by the
.Xr sysctl 3
MIB variable
.Dv KERN_MAXPROCPERUID .
.It Bq Er EAGAIN
The user is not the super user, and
the soft resource limit corresponding to the
.Fa resource
argument
.Dv RLIMIT_NOFILE
would be exceeded (see
.Xr getrlimit 2 ) .
.It Bq Er EINVAL
Both the RFFDG and the RFCFDG flags were specified.
.It Bq Er EINVAL
Any flags not listed above were specified.
.It Bq Er EINVAL
An invalid signal number was specified.
.It Bq Er ENOMEM
There is insufficient swap space for the new process.
.El
.Sh SEE ALSO
.Xr fork 2 ,
.Xr intro 2 ,
.Xr minherit 2 ,
.Xr vfork 2 ,
.Xr pthread_create 3 ,
.Xr rfork_thread 3
.Sh HISTORY
The
.Fn rfork
function first appeared in Plan9.