.\" Copyright (C) 2003, Michael Kerrisk (mtk-manpages@gmx.net)
.\"
.\" 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.\" manual page may be incorrect or out-of-date.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.
.\"
.\" Formatted or processed versions of this manual, if unaccompanied by
.\" the source, must acknowledge the copyright and authors of this work.
.\"
.\" 2003-12-10 Initial creation, Michael Kerrisk <mtk-manpages@gmx.net>
.\" 2004-10-28 aeb, corrected prototype, prot must be 0
.\"
.TH REMAP_FILE_PAGES 2 2004-10-28 "Linux 2.6" "Linux Programmer's Manual"
.SH NAME
remap_file_pages \- create a non-linear file mapping
.SH SYNOPSIS
.nf
.B #define _GNU_SOURCE
.B #include <sys/mman.h>
.sp
.BI "int remap_file_pages(void *" start ", size_t " size ", int " prot ,
.BI "                     ssize_t " pgoff ", int " flags );
.fi
.SH DESCRIPTION
The
.BR remap_file_pages ()
system call is used to create a non-linear mapping, that is, a mapping
in which the pages of the file are mapped into a non-sequential order
in memory.
The advantage of using
.BR remap_file_pages ()
over using repeated calls to
.BR mmap (2)
is that the former approach does not require the kernel to create
additional VMA (Virtual Memory Area) data structures.

To create a non-linear mapping we perform the following steps:
.TP
\fB1.\fP
Use
.BR mmap (2)
to create a mapping (which is initially linear).
This mapping must be created with the
.B MAP_SHARED
flag.
.TP
\fB2.\fP
Use one or more calls to
.BR remap_file_pages ()
to rearrange the correspondence between the pages of the mapping
and the pages of the file.
It is possible to map the same page of a file
into multiple locations within the mapped region.
.LP
The
.I pgoff
and
.I size
arguments specify the region of the file that is to be relocated
within the mapping:
.I pgoff
is a file offset in units of the system page size;
.I size
is the length of the region in bytes.

The
.I start
argument serves two purposes.
First, it identifies the mapping whose pages we want to rearrange.
Thus,
.I start
must be an address that falls within
a region previously mapped by a call to
.BR mmap (2).
Second,
.I start
specifies the address at which the file pages
identified by
.I pgoff
and
.I size
will be placed.

The values specified in
.I start
and
.I size
should be multiples of the system page size.
If they are not, then the kernel rounds
.I both
values
.I down
to the nearest multiple of the page size.
.\" This rounding is weird, and not consistent with the treatment of
.\" the analogous arguments for munmap()/mprotect() and for mlock().
.\" MTK, 14 Sep 2005

The
.I prot
argument must be specified as 0.

The
.I flags
argument has the same meaning as for
.BR mmap (2),
but all flags other than
.B MAP_NONBLOCK
are ignored.
.SH "RETURN VALUE"
On success,
.BR remap_file_pages ()
returns 0.
On error, \-1 is returned, and
.I errno
is set appropriately.
.SH NOTES
The
.BR remap_file_pages ()
system call appeared in Linux 2.5.46.
.SH ERRORS
.TP
.B EINVAL
.I start
does not refer to a valid mapping
created with the
.B MAP_SHARED
flag.
.TP
.B EINVAL
.IR start ,
.IR size ,
.IR prot ,
or
.I pgoff
is invalid.
.\" And possibly others from vma->vm_ops->populate()
.SH "CONFORMING TO"
The
.BR remap_file_pages ()
system call is Linux specific.
.SH "SEE ALSO"
.BR getpagesize (2),
.BR mmap (2),
.BR mmap2 (2),
.BR mprotect (2),
.BR mremap (2),
.BR msync (2),
.BR feature_test_macros (7)