man-pages/man2/shmop.2

.\" Copyright 1993 Giorgio Ciucci (giorgio@crcc.it)
.\"
.\" 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.\" manual page may be incorrect or out-of-date.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.
.\"
.\" Modified Sun Nov 28 17:06:19 1993, Rik Faith (faith@cs.unc.edu)
.\"          with material from Luigi P. Bai (lpb@softint.com)
.\" Portions Copyright 1993 Luigi P. Bai
.\" Modified Tue Oct 22 22:04:23 1996 by Eric S. Raymond <esr@thyrsus.com>
.\" Modified, 5 Jan 2002, Michael Kerrisk <mtk-manpages@gmx.net>
.\" Modified, 19 Sep 2002, Michael Kerrisk <mtk-manpages@gmx.net>
.\"	Added SHM_REMAP flag description
.\" Modified, 27 May 2004, Michael Kerrisk <mtk-manpages@gmx.net>
.\"     Added notes on capability requirements
.\" Modified, 11 Nov 2004, Michael Kerrisk <mtk-manpages@gmx.net>
.\"	Language and formatting clean-ups
.\"	Changed wording and placement of sentence regarding attachment
.\"		of segments marked for destruction
.\"
.TH SHMOP 2 2004-11-10 "Linux 2.6.9" "Linux Programmer's Manual"
.SH NAME
shmop \- shared memory operations
.SH SYNOPSIS
.nf
.B
#include <sys/types.h>
.B
#include <sys/shm.h>
.fi
.sp
.BI "void *shmat(int " shmid ,
.BI "const void *" shmaddr ,
.BI "int " shmflg );
.sp
.BI "int shmdt(const void *" shmaddr );
.SH DESCRIPTION
.BR shmat ()
attaches the shared memory segment identified by
.I shmid
to the address space of the calling process.
The attaching address is specified by
.I shmaddr
with one of the following criteria:
.LP
If
.I shmaddr
is NULL,
the system chooses a suitable (unused) address at which to attach
the segment.
.LP
If
.I shmaddr
isn't NULL
and
.B SHM_RND
is specified in
.IR shmflg ,
the attach occurs at the address equal to
.I shmaddr
rounded down to the nearest multiple of
.BR SHMLBA .
Otherwise
.I shmaddr
must be a page-aligned address at which the attach occurs.
.PP
If
.B SHM_RDONLY
is specified in
.IR shmflg ,
the segment is attached for reading and the process must have
read permission for the segment.
Otherwise the segment is attached for read and write
and the process must have read and write permission for the segment.
There is no notion of a write-only shared memory segment.
.PP
The (Linux-specific)
.B SHM_REMAP
flag may be specified in
.I shmflg
to indicate that the mapping of the segment should replace
any existing mapping in the range starting at
.I shmaddr
and continuing for the size of the segment.
(Normally an
.B EINVAL
error would result if a mapping already exists in this address range.)
In this case,
.I shmaddr
must not be NULL.
.PP
The
.BR brk (2)
value of the calling process is not altered by the attach.
The segment will automatically be detached at process exit.
The same segment may be attached as a read and as a read-write
one, and more than once, in the process's address space.
.PP
A successful
.BR shmat ()
call updates the members of the
.B shmid_ds
structure (see
.BR shmctl (2))
associated with the shared memory segment as follows:
.IP
.I shm_atime
is set to the current time.
.IP
.I shm_lpid
is set to the process-ID of the calling process.
.IP
.I shm_nattch
is incremented by one.
.PP
.BR shmdt ()
detaches the shared memory segment located at the address specified by
.I shmaddr
from the address space of the calling process.
The to\-be\-detached segment must be currently
attached with
.I shmaddr
equal to the value returned by the attaching
.BR shmat ()
call.
.PP
On a successful
.BR shmdt ()
call the system updates the members of the
.I shmid_ds
structure associated with the shared memory segment as follows:
.IP
.I shm_dtime
is set to the current time.
.IP
.I shm_lpid
is set to the process-ID of the calling process.
.IP
.I shm_nattch
is decremented by one.
If it becomes 0 and the segment is marked for deletion,
the segment is deleted.
.SH "SYSTEM CALLS"
.TP 11
.BR fork ()
After a
.BR fork ()
the child inherits the attached shared memory segments.
.TP
.BR exec ()
After an
.BR exec ()
all attached shared memory segments are detached from the process.
.TP
.BR exit ()
Upon
.BR exit ()
all attached shared memory segments are detached from the process.
.SH "RETURN VALUE"
On success
.BR shmat ()
returns the address of the attached shared memory segment; on error 
.I (void *) \-1 
is returned, and
.I errno
is set to indicate the cause of the error.

On success
.BR shmdt ()
returns 0; on error \-1 is returned, and
.I errno
is set to indicate the cause of the error.
.SH ERRORS
When
.BR shmat ()
fails,
.I errno
is set to one of the following:
.TP
.B EACCES
The calling process does not have the required permissions for
the requested attach type, and does not have the
.B CAP_IPC_OWNER
capability.
.TP
.B EINVAL
Invalid
.I shmid
value, unaligned (i.e., not page-aligned and \fBSHM_RND\fP was not
specified) or invalid
.I shmaddr
value, or failing attach at
.BR brk (),
.\" FIXME What does "failing attach at brk" mean?  (Is this phrase
.\" just junk?)
or
.B SHM_REMAP
was specified and
.I shmaddr
was NULL.
.TP
.B ENOMEM
Could not allocate memory for the descriptor or for the page tables.
.PP
When
.BR shmdt ()
fails,
.I errno
is set as follows:
.TP
.B EINVAL
There is no shared memory segment attached at
.IR shmaddr ; 
or,
.\" The following since 2.6.17-rc1:
.I shmaddr
is not aligned on a page boundary.
.SH NOTES
Using
.BR shmat ()
with
.I shmaddr
equal to NULL
is the preferred, portable way of attaching a shared memory segment.
Be aware that the shared memory segment attached in this way
may be attached at different addresses in different processes.
Therefore, any pointers maintained within the shared memory must be
made relative (typically to the starting address of the segment),
rather than absolute.
.PP
On Linux, it is possible to attach a shared memory segment even if it
is already marked to be deleted.
However, POSIX.1-2001 does not specify this behaviour and
many other implementations do not support it.
.LP
The following system parameter affects
.BR shmat ():
.TP 11
.\" FIXME A good explanation of the rationale for the existence 
.\" of SHMLBA would be useful here
.B SHMLBA
Segment low boundary address multiple.
Must be page aligned.
For the current implementation the
.B SHMLBA
value is
.BR PAGE_SIZE .
.\" FIXME That last sentence isn't true for all Linux
.\" architectures (i.e., SHMLBA != PAGE_SIZE for some architectures)
.\" -- MTK, Nov 04
.PP
The implementation places no intrinsic limit on the per\-process maximum
number of shared memory segments
.RB ( SHMSEG ).
.SH "CONFORMING TO"
SVr4, POSIX.1-2001.  
.\" SVr4 documents an additional error condition EMFILE.

In SVID 3 (or perhaps earlier) 
the type of the \fIshmaddr\fP argument was changed from
.B "char *"
into
.BR "const void *" ,
and the returned type of \fIshmat\fP() from
.B "char *"
into
.BR "void *" .
(Linux libc4 and libc5 have the
.B "char *"
prototypes; glibc2 has
.BR "void *" .)
.SH "SEE ALSO"
.BR brk (2),
.BR mmap (2),
.BR shmctl (2),
.BR shmget (2),
.BR capabilities (7),
.BR svipc (7)