path: root/source/a/mkinitrd/mkinitrd.8

.\" -*- nroff -*-
.ds g \" empty
.ds G \" empty
.\" Like TP, but if specified indent is more than half
.\" the current line-length - indent, use the default indent.
.de Tp
.ie \\n(.$=0:((0\\$1)*2u>(\\n(.lu-\\n(.iu)) .TP
.el .TP "\\$1"
..
.TH MKINITRD 8 "27 March 2010" "Slackware Version 13.1"
.SH NAME
mkinitrd \- create or rebuilt an initrd (initial ramdisk) using initramfs (simple cpio+gzip).
.SH SYNOPSIS
.B mkinitrd
[
.B \-F
]
[
.B \-c
]
[
.B \-f filesystem
]
[
.B \-h hibernate_partition
]
[
.B \-k kernel_version1:kernel_version2:kernel_version3...
]
[
.B \-m module1:module2:module3...
]
[
.B \-o output_file
]
[
.B \-r root_device
]
[
.B \-s source_tree
]
[
.B \-u
]
[
.B \-w wait_time
]
[
.B \-C device1:device2:device3...
]
[
.B \-K luks_keyfile
]
[
.B \-P microcode_archive
]
[
.B \-B
]
[
.B \-L
]
[
.B \-R
]
[
.B \-V
]
.SH DESCRIPTION
.B mkinitrd
is used to build an initial ramdisk.  An initial ramdisk is a very small
set of files that are loaded into RAM and "mounted" (as initramfs doesn't
actually use a filesystem) as the kernel boots (before the main root
filesystem is mounted). The usual reason to use an initrd is to load kernel
kernel modules in order to access the root partition. An initrd might also
be required to unlock an encrypted device.  Usually these modules are
required to support the filesystem used by the root partition (e.g., ext4,
jfs, xfs), or perhaps the controller that the hard drive is attached to
(SCSI, RAID, etc).  Essentially, there are so many different options
available in modern Linux kernels that it isn't practical to try to ship
many different kernels to try to cover everyone's needs. It's a lot more
flexible to ship a generic kernel and a set of kernel modules for it.
.SH OPTIONS
.TP
.B \-F
Use the contents of /etc/mkinitrd.conf as options to mkinitrd (optional).
If this is used in conjunction with any other options passed on the command 
line, the command-line options will override the config file options.  
.br
See mkinitrd.conf(5) for details.
.TP
.B \-c
Clear the existing initrd tree (by default in /boot/initrd-tree/) first.
If this is not done, running mkinitrd will add additional modules to the
existing initrd.
.TP
.B \-f filesystem
Specify the filesystem to use for the root partition.  If this isn't given,
mount will usually figure it out.  This option must be used together with the
\-r option in order to be beneficial.
.TP
.B \--help
Display a help summary.
.TP
.B \-h hibernate_partition
Specify the swap partition holding the hibernation image.
.TP
.B \-k kernel version list
Use kernel modules from the specified kernel version.  mkinitrd will look
for them in /lib/modules/(kernel version). This may be a single kernel version,
or it may be a colon-delimited list of kernel versions.
.TP
.B \-l keymap
Load an alternative keyboard mapping. All supported keyboard mappings
can be found in /usr/share/mkinitrd/keymaps.tar.gz
Leave the '.bmap' out when you supply this parameter. E.g. '-l nl' will 
add support for dutch keyboard mapping to the initrd.
.TP
.B \-m module list
This is a list of colon-delimited modules to build into the initrd.
Any dependencies of requested modules will also be added to the initrd.
Additional options may be added to use when loading the kernel modules
(but in this case the entire list must be wrapped with double quotes). 
.TP
.B \-o output image
The file to write the initrd to.  (default: /boot/initrd.gz)
.TP
.BI \-r \ root_device
Specify the device to be used as the root partition.  If this isn't given, the
kernel default will be used (which is usually fine).  This option must be used
together with the \-f option in order to be beneficial.
.I root_device
can be a device node path, UUID, or LABEL. See examples for \fB\-C\fR below.
.TP
.B \-s source tree
The directory to use as the source for the initrd.  (default: /boot/initrd-tree/)
.TP
.B \-u
Include udev in the initrd (provided for backward compatibility, as udev is 
always included in the initrd).
.TP
.B \-w
The -w option specifies how long to wait in seconds before assuming that all the
drives are spun up and ready to go.
.TP
.BI \-C \ device_list
A colon (:) delimited list of luks encrypted block devices to be unlocked by
the initrd using cryptsetup.
.I device_list
may include any of: device node path, UUID, or LABEL.  All devices that must be
unlocked in order to access the root filesystem must be specified. For example:

  -C /dev/sda2:/dev/sda3
.br
  -C UUID=ec6dea40-c8d8-4590-850a-a757be60e45a
.br
  -C LABEL=darkstar

Each unlocked device will be assigned an automatically generated luks device
name of the form luks<device> where '<device>' will be the basename of the
encrypted device node path. For example:

  /dev/mapper/lukssda2

As a convenience to users, where -r specifies one of the device names listed
on the -C option it will be automatically adjusted to use the correct luks
device name. i.e.

  "-C /dev/sda2 -r /dev/sda2" and
  "-C /dev/sda2 -r /dev/mapper/lukssda2"

are equivalent.
.br
(Use with '-r' option).
.TP
.B \-K luks_keyfile
When using cryptsetup to encrypt your partition, you can use a keyfile instead
of a passphrase to unlock the LUKS volume.  The LUKSKEY variable holds the
filename of a keyfile, as well as the label (or uuid) of the partition this
file is on.  This way, you can unlock your computer automatically if you have a
USB stick with your LUKS key inserted at boot. A passphrase will still be asked
if the LUKS key can not be found.
.br
For example, if your USB thumb drive has a FAT partition with label
"TRAVELSTICK" and the actual keyfile is called "/keys/alien.luks", then
you need to set:

  -K LABEL=TRAVELSTICK:/keys/alien.luks
.TP
.B \-T device list
A colon (:) delimited list of luks encrypted block devices to be passed the
"--allow-discards" option when unlocked by the initrd using cryptsetup, e.g.

  -T /dev/sda2:/dev/sda4

This has the effect of allowing TRIM on SSD drives. Be sure your SSD supports 
this feature (correctly) before enabling it.  See fstrim(8) for more information.
.TP
.B \-P microcode_archive
This option specifies a cpio archive containing updated microcode for your CPU.
CPU manufacturers occasionally release such updates to fix bugs in the microcode
currently embedded in the CPU. The microcode archive will be prepended to the
output initrd, where the kernel will find it for early patching:

  -P /boot/intel-ucode.cpio

.TP
.B \-B
This option adds the btrfs utility to the initrd so that multi-device filesystems
will be picked up by a scan (/sbin/btrfs device scan).  This is needed if the
root filesystem is a Btrfs multi-device filesystem.
.TP
.B \-L
This option adds LVM support to the initrd, if the tools are
available on the system.
.TP
.B \-R
This option adds RAID support to the initrd, if a static mdadm binary is
available on the system.
.TP
.B \-M
This option add the files in /etc/modprobe.d/ and /lib/modprobe.d/ to the initrd.
.TP
.B \-V
Display version information and exit.
.SH EXAMPLES
A simple example:  Build an initrd for a reiserfs root partition:

  mkinitrd -c -m reiserfs

Another example:  Build an initrd image using Linux 2.6.33.1 kernel
modules for a system with an ext3 root partition on /dev/sdb3:

  mkinitrd -c -k 2.6.33.1 -m ext3 -f ext3 -r /dev/sdb3

An example of a single encrypted partition setup:
.br
As a user convenience, the value for the "-r" option may also be specified as
"/dev/sda2" in this example:

  mkinitrd -c -k 2.6.33.1 \\
           -m ext4:ehci-hcd:uhci-hcd:usbhid \\
           -f ext4 -r /dev/mapper/lukssda2 \\
           -C /dev/sda2 \\
           -l uk

Finally, A more complex example:
.br
This one is for a LVM Volume Group (rootvg) comprising of two LVM Physical
Volumes, each of which is on a LUKS encrypted partition that will need to be
unlocked before the root filesystem (/dev/rootvg/lvroot) can be accessed.

  mkinitrd -c -k 2.6.29.6 \\
           -m ext4:ehci-hcd:uhci-hcd:usbhid \\
           -f ext4 -r /dev/rootvg/lvroot \\
           -L -C /dev/sda2:/dev/sdb2 \\
           -l uk

If run without options, mkinitrd will rebuild an initrd image using
the contents of the $SOURCE_TREE directory, or, if that directory
does not exist it will be created and populated, and then mkinitrd
will exit.  These options are handy for building an initrd mostly
by hand.  After creating /boot/initrd-tree/, you can add modules and
edit files by hand, and then rerun mkinitrd to create the initrd.

Once the initrd is created, you'll need to tell your boot loader
to load it.  If you boot with LILO, you will need to add an initrd
line to /etc/lilo.conf.  Here's a section of lilo.conf that shows
how to set this up:

 # Linux bootable partition config begins
 image = /boot/vmlinuz-generic-2.6.33.1
   initrd = /boot/initrd.gz
   root = /dev/sda3
   label = Linux26331
   read-only
 # Linux bootable partition config ends

Note that the line "root = /dev/sda3" is not needed if the root device
has been configured in the initrd image.

Once you've created the initrd and editing /etc/lilo.conf, you will
need to run 'lilo' to write out the changed boot block.  The next
time you reboot the initrd should be loaded along with the kernel.

Have fun!

.SH SEE ALSO
mkinitrd.conf (5)

.SH AUTHOR
Patrick J. Volkerding <volkerdi@slackware.com>