.\" Copyright 1998 Andries E. Brouwer (aeb@cwi.nl)
.\"
.\" May be distributed under the GNU General Public License
.\" Rewritten for 2.1.117, aeb, 981010.
.\"
.TH MKSWAP 8 "25 March 1999" "Linux 2.2.4" "Linux Programmer's Manual"
.SH NAME
mkswap \- set up a Linux swap area
.SH SYNOPSIS
.BI "mkswap [\-c] [\-v" N "] [\-f] [\-p " PSZ "] [\-L " label "] " device " [" size "]"
.SH DESCRIPTION
.B mkswap
sets up a Linux swap area on a device or in a file.
(After creating the swap area, you need the
.B swapon
command to start using it. Usually swap areas are listed in
.I /etc/fstab
so that they can be taken into use at boot time by a
.B swapon -a
command in some boot script.)
The
.I device
argument will usually be a disk partition (something like
.I /dev/hda4
or
.IR /dev/sdb7 )
but can also be a file.
The Linux kernel does not look at partition Id's, but
many installation scripts will assume that partitions
of hex type 82 (LINUX_SWAP) are meant to be swap partitions.
(Warning: Solaris also uses this type. Be careful not to kill
your Solaris partitions.)
The
.I size
parameter is superfluous but retained for backwards compatibility.
(It specifies the desired size of the swap area in 1024-byte blocks.
.B mkswap
will use the entire partition or file if it is omitted.
Specifying it is unwise - a typo may destroy your disk.)
The
.I PSZ
parameter specifies the page size to use. It is almost always
unnecessary (even unwise) to specify it, but certain old libc
versions lie about the page size, so it is possible that
.B mkswap
gets it wrong. The symptom is that a subsequent
.B swapon
fails because no swap signature is found. Typical values for
.I PSZ
are 4096 or 8192.
Linux knows about two styles of swap areas, old style and new style.
The last 10 bytes of the first page of the swap area distinguishes
them: old style has `SWAP_SPACE', new style has `SWAPSPACE2' as
signature.
In the old style, the rest of this first page was a bit map,
with a 1 bit for each usable page of the swap area.
Since the first page holds this bit map, the first bit is 0.
Also, the last 10 bytes hold the signature. So, if the page
size is S, an old style swap area can describe at most
8*(S-10)-1 pages used for swapping.
With S=4096 (as on i386), the useful area is at most 133890048 bytes
(almost 128 MiB), and the rest is wasted.
On an alpha and sparc64, with S=8192, the useful area is at most
535560992 bytes (almost 512 MiB).
The old setup wastes most of this bitmap page, because zero bits
denote bad blocks or blocks past the end of the swap space,
and a simple integer suffices to indicate the size of the swap space,
while the bad blocks, if any, can simply be listed. Nobody wants
to use a swap space with hundreds of bad blocks. (I would not even
use a swap space with 1 bad block.)
In the new style swap area this is precisely what is done.
The maximum useful size of a swap area now depends on the architecture.
It is roughly 2GiB on i386, PPC, m68k, ARM, 1GiB on sparc, 512MiB on mips,
128GiB on alpha and 3TiB on sparc64.
Note that before 2.1.117 the kernel allocated one byte for each page,
while it now allocates two bytes, so that taking a swap area of 2 GiB
in use might require 2 MiB of kernel memory.
Presently, Linux allows 32 swap areas (this was 8 before Linux 2.4.10).
The areas in use can be seen in the file
.I /proc/swaps
(since 2.1.25).
.B mkswap
refuses areas smaller than 10 pages.
If you don't know the page size that your machine uses, you may be
able to look it up with "cat /proc/cpuinfo" (or you may not -
the contents of this file depend on architecture and kernel version).
To setup a swap file, it is necessary to create that file before
initializing it with
.B mkswap ,
e.g. using a command like
.nf
.RS
# dd if=/dev/zero of=swapfile bs=1024 count=65536
.RE
.fi
Note that a swap file must not contain any holes (so, using
.BR cp (1)
to create the file is not acceptable).
.SH OPTIONS
.TP
.B \-c
Check the device (if it is a block device) for bad blocks
before creating the swap area.
If any are found, the count is printed.
.TP
.B \-f
Force - go ahead even if the command is stupid.
This allows the creation of a swap area larger than the file
or partition it resides on.
On SPARC, force creation of the swap area.
Without this option
.B mkswap
will refuse to create a v0 swap on a device with a valid SPARC superblock,
as that probably means one is going to erase the partition table.
.TP
.BI "\-p " PSZ
Specify the page size to use.
.TP
.BI "\-L " label
Specify a label, to allow swapon by label.
(Only for new style swap areas.)
.TP
.B \-v0
Create an old style swap area.
.TP
.B \-v1
Create a new style swap area.
.LP
If no \-v option is given,
.B mkswap
will default to new style, but use old style if the current kernel
is older than 2.1.117 (and also if PAGE_SIZE is less than 2048).
The new style header does not touch the first block, so may be
preferable, in case you have a boot loader or disk label there.
If you need to use both 2.0 and 2.2 kernels, use the \-v0 option
when creating the swapspace.
.SH "SEE ALSO"
.BR fdisk (8),
.BR swapon (8)
