2 .\" This man-page is Copyright (C) 1997 John S. Kallal
4 .\" Permission is granted to make and distribute verbatim copies of this
5 .\" manual provided the copyright notice and this permission notice are
6 .\" preserved on all copies.
8 .\" Permission is granted to copy and distribute modified versions of this
9 .\" manual under the conditions for verbatim copying, provided that the
10 .\" entire resulting derived work is distributed under the terms of a
11 .\" permission notice identical to this one.
13 .\" Since the Linux kernel and libraries are constantly changing, this
14 .\" manual page may be incorrect or out-of-date. The author(s) assume no
15 .\" responsibility for errors or omissions, or for damages resulting from
16 .\" the use of the information contained herein. The author(s) may not
17 .\" have taken the same level of care in the production of this manual,
18 .\" which is licensed free of charge, as they might when working
21 .\" Formatted or processed versions of this manual, if unaccompanied by
22 .\" the source, must acknowledge the copyright and author(s) of this work.
24 .\" If the you wish to distribute versions of this work under other
25 .\" conditions than the above, please contact the author(s) at the following
29 .\" email: <kallal@voicenet.com>
30 .\" mail: 518 Kerfoot Farm RD, Wilmington, DE 19803-2444, USA
31 .\" phone: (302)654-5478
33 .\" $Id: initrd.4,v 0.9 1997/11/07 05:05:32 kallal Exp kallal $
34 .TH INITRD 4 1997-11-06 "Linux" "Linux Programmer's Manual"
36 initrd \- boot loader initialized RAM disk
40 is a read-only block device.
43 is a RAM disk that is initialized (e.g., loaded) by the boot loader before
44 the kernel is started.
45 The kernel then can use the block device
47 contents for a two phased system boot-up.
49 In the first boot-up phase, the kernel starts up
50 and mounts an initial root file-system from the contents
53 (e.g. RAM disk initialized by the boot loader).
54 In the second phase, additional drivers or other modules
55 are loaded from the initial root device's contents.
56 After loading the additional modules, a new root file system
57 (i.e., the normal root file system) is mounted from a
62 .SS "Boot-up Operation"
64 .BR initrd ", the system boots as follows:"
66 The boot loader loads the kernel program and
67 .IR /dev/initrd "'s contents into memory."
70 the kernel uncompresses and copies the contents of the device
74 and then frees the memory used by
77 The kernel then read-write mounts device
79 as the initial root file system.
81 If the indicated normal root file system is also the initial root file-system
84 ) then the kernel skips to the last step for the usual boot sequence.
86 If the executable file
87 .IR /linuxrc " is present in the initial root file-system, " /linuxrc
88 is executed with UID 0.
91 must have executable permission.
94 can be any valid executable, including a shell script.)
98 is not executed or when
100 terminates, the normal root file system is mounted.
103 exits with any file-systems mounted on the initial root
104 file-system, then the behavior of the kernel is
108 section for the current kernel behavior.)
110 If the normal root file has directory
111 .IR /initrd ", device"
114 .IR / " to " /initrd "."
115 Otherwise if directory
116 .IR /initrd " does not exist device " /dev/ram0 " is unmounted."
118 .IR / " to " /initrd ", " /dev/ram0
119 is not unmounted and therefore processes can remain running from
123 does not exist on the normal root file-system
124 and any processes remain running from
125 .IR /dev/ram0 " when " /linuxrc
126 exits, the behavior of the kernel is
130 section for the current kernel behavior.)
132 The usual boot sequence (e.g. invocation of
134 is performed on the normal root file system.
139 The following boot loader options when used with
140 .BR initrd ", affect the kernel's boot-up operation:"
142 .BI initrd= "filename"
143 Specifies the file to load as the contents of
147 this is a command line option.
150 you have to use this command in the
153 .IR /etc/lilo.config .
154 The filename specified with this
155 option will typically be a gzipped file-system image.
158 This boot time option disables the two phase boot-up operation.
159 The kernel performs the usual boot sequence as if
162 With this option, any contents of
164 loaded into memory by the boot loader contents are preserved.
165 This option permits the contents of
167 to be any data and need not be limited to a file system image.
170 is read-only and can be read only one time after system startup.
172 .BI root= "device-name"
173 Specifies the device to be used as the normal root file system.
175 this is a command line option.
176 .RB "For " LILO " this is a boot time option or
177 can be used as an option line in the
178 .BR LILO " configuration file " /etc/lilo.config "."
179 The device specified by the this option must be a mountable
180 device having a suitable root file-system.
184 .SS "Changing the Normal Root File System"
186 the kernel's settings
187 (e.g. set in the kernel file with
189 or compiled into the kernel file),
190 or the boot loader option setting
191 is used for the normal root file systems.
192 For a NFS-mounted normal root file system, one has to use the
193 .BR nfs_root_name " and " nfs_root_addrs
194 boot options to give the NFS settings.
195 For more information on NFS-mounted root see the kernel documentation file
197 For more information on setting the root file system also see the
198 .BR LILO " and " LOADLIN " documentation."
200 It is also possible for the
202 executable to change the normal root device.
205 to change the normal root device,
206 .IR /proc " must be mounted."
208 .IR /proc ", " /linuxrc
209 changes the normal root device by writing into the proc files
210 .IR /proc/sys/kernel/real-root-dev ", "
211 .IR /proc/sys/kernel/nfs-root-name ", and "
212 .IR /proc/sys/kernel/nfs-root-addrs "."
213 For a physical root device, the root device is changed by having
215 write the new root file system device number into
216 .IR /proc/sys/kernel/real-root-dev "."
217 For a NFS root file system, the root device is changed by having
219 write the NFS setting into files
220 .IR /proc/sys/kernel/nfs-root-name " and "
221 .I /proc/sys/kernel/nfs-root-addrs
222 and then writing 0xff (e.g. the pseudo-NFS-device number) into file
223 .IR /proc/sys/kernel/real-root-dev "."
224 For example, the following shell command line would change
225 the normal root device to
229 echo 0x365 >/proc/sys/kernel/real-root-dev
232 For a NFS example, the following shell command lines would change the
233 normal root device to the NFS directory
235 on a local networked NFS server with IP number 193.8.232.7 for a system with
236 IP number 193.8.232.7 and named 'idefix':
239 echo /var/nfsroot >/proc/sys/kernel/nfs-root-name
240 echo 193.8.232.2:193.8.232.7::255.255.255.0:idefix \\
241 >/proc/sys/kernel/nfs-root-addrs
242 echo 255 >/proc/sys/kernel/real-root-dev
247 .I /proc/sys/kernel/real-root-dev
248 to change the root file system is obsolete.
249 See the kernel source file
250 .I Documentation/initrd.txt
255 for information on the modern method of changing the root file system.
256 .\" FIXME the manual page should describe the pivot_root mechanism.
261 The main motivation for implementing
263 was to allow for modular kernel configuration at system installation.
265 A possible system installation scenario is as follows:
267 The loader program boots from floppy or other media with a minimal kernel
269 .IR /dev/ram ", " /dev/initrd ", and the ext2 file-system) and loads "
270 .IR /dev/initrd " with a gzipped version of the initial file-system.
274 determines what is needed to (1) mount the normal root file-system
275 (i.e., device type, device drivers, file system) and (2) the
276 distribution media (e.g. CD-ROM, network, tape, ...).
278 done by asking the user, by auto-probing, or by using a hybrid
283 loads the necessary modules from the initial root file-system.
287 creates and populates the root file system.
288 (At this stage the normal root file system does not have to be a
289 completed system yet.)
292 .IR /linuxrc " sets " /proc/sys/kernel/real-root-dev,
295 the normal root file system and any other file
296 systems it has mounted, and then terminates.
298 6. The kernel then mounts the normal root file system.
300 Now that the file system is accessible and intact,
301 the boot loader can be installed.
303 The boot loader is configured to load into
305 a file system with the set of modules that was used to bring up the system.
308 can be modified, then unmounted, and finally, the image is written from
312 The system is now bootable and additional installation tasks can be
317 in the above is to re-use the configuration data during normal system operation
318 without requiring initial kernel selection, a large generic kernel or,
319 recompiling the kernel.
321 A second scenario is for installations where Linux runs on systems with
322 different hardware configurations in a single administrative network.
323 In such cases, it may be desirable to use only a small set of kernels
324 (ideally only one) and to keep the system-specific part of configuration
325 information as small as possible.
326 In this case, create a common file
327 with all needed modules.
330 file or a file executed by
334 A third scenario is more convenient recovery disks.
335 Because information like the location of the root file-system
336 partition is not needed at boot time, the system loaded from
338 can use a dialog and/or auto-detection followed by a
339 possible sanity check.
341 Last but not least, Linux distributions on CD-ROM may use
343 for easy installation from the CD-ROM.
344 The distribution can use
348 from CD-ROM without the need of any floppies.
349 The distribution could also use a
351 boot floppy and then bootstrap a bigger ram disk via
352 .IR /dev/initrd " from the CD-ROM."
359 is a read-only block device assigned
360 major number 1 and minor number 250.
365 with mode 0400 (read access by root only).
366 If the Linux system does not have
368 already created, it can be created with the following commands:
371 mknod \-m 400 /dev/initrd b 1 250
372 chown root:disk /dev/initrd
376 Also, support for both "RAM disk" and "Initial RAM disk"
378 .BR CONFIG_BLK_DEV_RAM=y " and " CONFIG_BLK_DEV_INITRD=y
379 ) support must be compiled directly into the Linux kernel to use
383 the RAM disk driver cannot be loaded as a module.
400 With the current kernel, any file systems that remain mounted when
401 .IR /dev/ram0 " is moved from " / " to " /initrd
402 continue to be accessible.
405 entries are not updated.
407 With the current kernel, if directory
408 .IR /initrd " does not exist, then "
410 will NOT be fully unmounted if
412 is used by any process or has any file-system mounted on it.
414 .IR /dev/ram0 " is NOT fully unmounted, "
417 will remain in memory.
421 should not depend on the behavior give in the above notes.
422 The behavior may change in future versions of the Linux kernel.
427 .\" The kernel code for device
429 .\" was written by Werner Almesberger <almesber@lrc.epfl.ch> and
430 .\" Hans Lermen <lermen@elserv.ffm.fgan.de>.
433 .\" was added to the baseline Linux kernel in development version 1.3.73.
441 The documentation file
443 in the kernel source package, the LILO documentation,
444 the LOADLIN documentation, the SYSLINUX documentation.