man2/shmget.2

.\" Copyright (c) 1993 Luigi P. Bai (lpb@softint.com) July 28, 1993
.\"
.\" %%%LICENSE_START(VERBATIM)
.\" Permission is granted to make and distribute verbatim copies of this
.\" manual provided the copyright notice and this permission notice are
.\" preserved on all copies.
.\"
.\" Permission is granted to copy and distribute modified versions of this
.\" manual under the conditions for verbatim copying, provided that the
.\" entire resulting derived work is distributed under the terms of a
.\" permission notice identical to this one.
.\"
.\" Since the Linux kernel and libraries are constantly changing, this
.\" manual page may be incorrect or out-of-date.  The author(s) assume no
.\" responsibility for errors or omissions, or for damages resulting from
.\" the use of the information contained herein.  The author(s) may not
.\" have taken the same level of care in the production of this manual,
.\" which is licensed free of charge, as they might when working
.\" professionally.
.\"
.\" Formatted or processed versions of this manual, if unaccompanied by
.\" the source, must acknowledge the copyright and authors of this work.
.\" %%%LICENSE_END
.\"
.\" Modified Wed Jul 28 10:57:35 1993, Rik Faith <faith@cs.unc.edu>
.\" Modified Sun Nov 28 16:43:30 1993, Rik Faith <faith@cs.unc.edu>
.\"          with material from Giorgio Ciucci <giorgio@crcc.it>
.\" Portions Copyright 1993 Giorgio Ciucci <giorgio@crcc.it>
.\" Modified Tue Oct 22 22:03:17 1996 by Eric S. Raymond <esr@thyrsus.com>
.\" Modified, 8 Jan 2003, Michael Kerrisk, <mtk.manpages@gmail.com>
.\"	Removed EIDRM from errors - that can't happen...
.\" Modified, 27 May 2004, Michael Kerrisk <mtk.manpages@gmail.com>
.\"     Added notes on capability requirements
.\" Modified, 11 Nov 2004, Michael Kerrisk <mtk.manpages@gmail.com>
.\"	Language and formatting clean-ups
.\"	Added notes on /proc files
.\"
.TH SHMGET 2 2020-04-11 "Linux" "Linux Programmer's Manual"
.SH NAME
shmget \- allocates a System V shared memory segment
.SH SYNOPSIS
.ad l
.B #include <sys/ipc.h>
.br
.B #include <sys/shm.h>
.PP
.BI "int shmget(key_t " key ", size_t " size ", int " shmflg );
.ad b
.SH DESCRIPTION
.BR shmget ()
returns the identifier of the System\ V shared memory segment
associated with the value of the argument
.IR key .
It may be used either to obtain the identifier of a previously created
shared memory segment (when
.I shmflg
is zero and
.I key
does not have the value
.BR IPC_PRIVATE ),
or to create a new set.
.PP
A new shared memory segment, with size equal to the value of
.I size
rounded up to a multiple of
.BR PAGE_SIZE ,
is created if
.I key
has the value
.B IPC_PRIVATE
or
.I key
isn't
.BR IPC_PRIVATE ,
no shared memory segment corresponding to
.I key
exists, and
.B IPC_CREAT
is specified in
.IR shmflg .
.PP
If
.I shmflg
specifies both
.B IPC_CREAT
and
.B IPC_EXCL
and a shared memory segment already exists for
.IR key ,
then
.BR shmget ()
fails with
.I errno
set to
.BR EEXIST .
(This is analogous to the effect of the combination
.B O_CREAT | O_EXCL
for
.BR open (2).)
.PP
The value
.I shmflg
is composed of:
.TP
.B IPC_CREAT
Create a new segment.
If this flag is not used, then
.BR shmget ()
will find the segment associated with \fIkey\fP and check to see if
the user has permission to access the segment.
.TP
.B IPC_EXCL
This flag is used with
.B IPC_CREAT
to ensure that this call creates the segment.
If the segment already exists, the call fails.
.TP
.BR SHM_HUGETLB " (since Linux 2.6)"
Allocate the segment using "huge pages."
See the Linux kernel source file
.I Documentation/admin-guide/mm/hugetlbpage.rst
for further information.
.TP
.BR SHM_HUGE_2MB ", " SHM_HUGE_1GB " (since Linux 3.8)"
.\" See https://lwn.net/Articles/533499/
Used in conjunction with
.B SHM_HUGETLB
to select alternative hugetlb page sizes (respectively, 2\ MB and 1\ GB)
on systems that support multiple hugetlb page sizes.
.IP
More generally, the desired huge page size can be configured by encoding
the base-2 logarithm of the desired page size in the six bits at the offset
.BR SHM_HUGE_SHIFT .
Thus, the above two constants are defined as:
.IP
.in +4
.EX
#define SHM_HUGE_2MB    (21 << SHM_HUGE_SHIFT)
#define SHM_HUGE_1GB    (30 << SHM_HUGE_SHIFT)
.EE
.in
.IP
For some additional details,
see the discussion of the similarly named constants in
.BR mmap (2).
.TP
.BR SHM_NORESERVE " (since Linux 2.6.15)"
This flag serves the same purpose as the
.BR mmap (2)
.B MAP_NORESERVE
flag.
Do not reserve swap space for this segment.
When swap space is reserved, one has the guarantee
that it is possible to modify the segment.
When swap space is not reserved one might get
.B SIGSEGV
upon a write
if no physical memory is available.
See also the discussion of the file
.I /proc/sys/vm/overcommit_memory
in
.BR proc (5).
.\" As at 2.6.17-rc2, this flag has no effect if SHM_HUGETLB was also
.\" specified.
.PP
In addition to the above flags, the least significant 9 bits of
.I shmflg
specify the permissions granted to the owner, group, and others.
These bits have the same format, and the same
meaning, as the
.I mode
argument of
.BR open (2).
Presently, execute permissions are not used by the system.
.PP
When a new shared memory segment is created,
its contents are initialized to zero values, and
its associated data structure,
.I shmid_ds
(see
.BR shmctl (2)),
is initialized as follows:
.IP \(bu 2
.I shm_perm.cuid
and
.I shm_perm.uid
are set to the effective user ID of the calling process.
.IP \(bu
.I shm_perm.cgid
and
.I shm_perm.gid
are set to the effective group ID of the calling process.
.IP \(bu
The least significant 9 bits of
.I shm_perm.mode
are set to the least significant 9 bit of
.IR shmflg .
.IP \(bu
.I shm_segsz
is set to the value of
.IR size .
.IP \(bu
.IR shm_lpid ,
.IR shm_nattch ,
.IR shm_atime ,
and
.I shm_dtime
are set to 0.
.IP \(bu
.I shm_ctime
is set to the current time.
.PP
If the shared memory segment already exists, the permissions are
verified, and a check is made to see if it is marked for destruction.
.SH RETURN VALUE
On success, a valid shared memory identifier is returned.
On error, \-1 is returned, and
.I errno
is set to indicate the error.
.SH ERRORS
On failure,
.I errno
is set to one of the following:
.TP
.B EACCES
The user does not have permission to access the
shared memory segment, and does not have the
.B CAP_IPC_OWNER
capability in the user namespace that governs its IPC namespace.
.TP
.B EEXIST
.BR IPC_CREAT
and
.BR IPC_EXCL
were specified in
.IR shmflg ,
but a shared memory segment already exists for
.IR key .
.TP
.B EINVAL
A new segment was to be created and
.I size
is less than
.B SHMMIN
or greater than
.BR SHMMAX .
.TP
.B EINVAL
A segment for the given
.I key
exists, but \fIsize\fP is greater than the size
of that segment.
.TP
.B ENFILE
.\" [2.6.7] shmem_zero_setup()-->shmem_file_setup()-->get_empty_filp()
The system-wide limit on the total number of open files has been reached.
.TP
.B ENOENT
No segment exists for the given \fIkey\fP, and
.B IPC_CREAT
was not specified.
.TP
.B ENOMEM
No memory could be allocated for segment overhead.
.TP
.B ENOSPC
All possible shared memory IDs have been taken
.RB ( SHMMNI ),
or allocating a segment of the requested
.I size
would cause the system to exceed the system-wide limit on shared memory
.RB ( SHMALL ).
.TP
.B EPERM
The
.B SHM_HUGETLB
flag was specified, but the caller was not privileged (did not have the
.B CAP_IPC_LOCK
capability).
.SH CONFORMING TO
POSIX.1-2001, POSIX.1-2008, SVr4.
.\" SVr4 documents an additional error condition EEXIST.
.PP
.B SHM_HUGETLB
and
.B SHM_NORESERVE
are Linux extensions.
.SH NOTES
The inclusion of
.I <sys/types.h>
and
.I <sys/ipc.h>
isn't required on Linux or by any version of POSIX.
However,
some old implementations required the inclusion of these header files,
and the SVID also documented their inclusion.
Applications intended to be portable to such old systems may need
to include these header files.
.\" Like Linux, the FreeBSD man pages still document
.\" the inclusion of these header files.
.PP
.B IPC_PRIVATE
isn't a flag field but a
.I key_t
type.
If this special value is used for
.IR key ,
the system call ignores all but the least significant 9 bits of
.I shmflg
and creates a new shared memory segment.
.\"
.SS Shared memory limits
The following limits on shared memory segment resources affect the
.BR shmget ()
call:
.TP
.B SHMALL
System-wide limit on the total amount of shared memory,
measured in units of the system page size.
.IP
On Linux, this limit can be read and modified via
.IR /proc/sys/kernel/shmall .
Since Linux 3.16,
.\" commit 060028bac94bf60a65415d1d55a359c3a17d5c31
the default value for this limit is:
.IP
    ULONG_MAX - 2^24
.IP
The effect of this value
(which is suitable for both 32-bit and 64-bit systems)
is to impose no limitation on allocations.
This value, rather than
.BR ULONG_MAX ,
was chosen as the default to prevent some cases where historical
applications simply raised the existing limit without first checking
its current value.
Such applications would cause the value to overflow if the limit was set at
.BR ULONG_MAX .
.IP
From Linux 2.4 up to Linux 3.15,
the default value for this limit was:
.IP
    SHMMAX / PAGE_SIZE * (SHMMNI / 16)
.IP
If
.B SHMMAX
and
.B SHMMNI
were not modified, then multiplying the result of this formula
by the page size (to get a value in bytes) yielded a value of 8\ GB
as the limit on the total memory used by all shared memory segments.
.TP
.B SHMMAX
Maximum size in bytes for a shared memory segment.
.IP
On Linux, this limit can be read and modified via
.IR /proc/sys/kernel/shmmax .
Since Linux 3.16,
.\" commit 060028bac94bf60a65415d1d55a359c3a17d5c31
the default value for this limit is:
.IP
    ULONG_MAX - 2^24
.IP
The effect of this value
(which is suitable for both 32-bit and 64-bit systems)
is to impose no limitation on allocations.
See the description of
.BR SHMALL
for a discussion of why this default value (rather than
.BR ULONG_MAX )
is used.
.IP
From Linux 2.2 up to Linux 3.15, the default value of
this limit was 0x2000000 (32\ MB).
.IP
Because it is not possible to map just part of a shared memory segment,
the amount of virtual memory places another limit on the maximum size of a
usable segment:
for example, on i386 the largest segments that can be mapped have a
size of around 2.8\ GB, and on x86-64 the limit is around 127 TB.
.TP
.B SHMMIN
Minimum size in bytes for a shared memory segment: implementation
dependent (currently 1 byte, though
.B PAGE_SIZE
is the effective minimum size).
.TP
.B SHMMNI
System-wide limit on the number of shared memory segments.
In Linux 2.2, the default value for this limit was 128;
since Linux 2.4, the default value is 4096.
.IP
On Linux, this limit can be read and modified via
.IR /proc/sys/kernel/shmmni .
.\" Kernels between 2.4.x and 2.6.8 had an off-by-one error that meant
.\" that we could create one more segment than SHMMNI -- MTK
.\" This /proc file is not available in Linux 2.2 and earlier -- MTK
.PP
The implementation has no specific limits for the per-process maximum
number of shared memory segments
.RB ( SHMSEG ).
.SS Linux notes
Until version 2.3.30, Linux would return
.B EIDRM
for a
.BR shmget ()
on a shared memory segment scheduled for deletion.
.SH BUGS
The name choice
.B IPC_PRIVATE
was perhaps unfortunate,
.B IPC_NEW
would more clearly show its function.
.SH EXAMPLES
See
.BR shmop (2).
.SH SEE ALSO
.BR memfd_create (2),
.BR shmat (2),
.BR shmctl (2),
.BR shmdt (2),
.BR ftok (3),
.BR capabilities (7),
.BR shm_overview (7),
.BR sysvipc (7)