€•Õ      Œsphinx.addnodes”Œdocument”“”)”}”(Œ	rawsource”Œ ”Œchildren”]”(Œtranslations”ŒLanguagesNode”“”)”}”(hhh]”(h Œpending_xref”“”)”}”(hhh]”Œdocutils.nodes”ŒText”“”ŒChinese (Simplified)”…””}”Œparent”hsbaŒ
attributes”}”(Œids”]”Œclasses”]”Œnames”]”Œdupnames”]”Œbackrefs”]”Œ	refdomain”Œstd”Œreftype”Œdoc”Œ	reftarget”Œ6/translations/zh_CN/filesystems/ramfs-rootfs-initramfs”Œmodname”NŒ	classname”NŒrefexplicit”ˆuŒtagname”hhhubh)”}”(hhh]”hŒChinese (Traditional)”…””}”hh2sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ6/translations/zh_TW/filesystems/ramfs-rootfs-initramfs”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒItalian”…””}”hhFsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ6/translations/it_IT/filesystems/ramfs-rootfs-initramfs”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒJapanese”…””}”hhZsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ6/translations/ja_JP/filesystems/ramfs-rootfs-initramfs”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒKorean”…””}”hhnsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ6/translations/ko_KR/filesystems/ramfs-rootfs-initramfs”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒPortuguese (Brazilian)”…””}”hh‚sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ6/translations/pt_BR/filesystems/ramfs-rootfs-initramfs”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒSpanish”…””}”hh–sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ6/translations/sp_SP/filesystems/ramfs-rootfs-initramfs”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubeh}”(h]”h ]”h"]”h$]”h&]”Œcurrent_language”ŒEnglish”uh1h
hhŒ	_document”hŒsource”NŒline”NubhŒcomment”“”)”}”(hŒ SPDX-License-Identifier: GPL-2.0”h]”hŒ SPDX-License-Identifier: GPL-2.0”…””}”hh·sbah}”(h]”h ]”h"]”h$]”h&]”Œ	xml:space”Œpreserve”uh1hµhhh²hh³ŒP/var/lib/git/docbuild/linux/Documentation/filesystems/ramfs-rootfs-initramfs.rst”h´KubhŒsection”“”)”}”(hhh]”(hŒtitle”“”)”}”(hŒRamfs, rootfs and initramfs”h]”hŒRamfs, rootfs and initramfs”…””}”(hhÏh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhhÊh²hh³hÇh´KubhŒ	paragraph”“”)”}”(hŒOctober 17, 2005”h]”hŒOctober 17, 2005”…””}”(hhßh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KhhÊh²hubhŒ
field_list”“”)”}”(hhh]”hŒfield”“”)”}”(hhh]”(hŒ
field_name”“”)”}”(hŒAuthor”h]”hŒAuthor”…””}”(hhùh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h÷hhôh³hÇh´K ubhŒ
field_body”“”)”}”(hŒRob Landley <rob@landley.net>
”h]”hÞ)”}”(hŒRob Landley <rob@landley.net>”h]”(hŒRob Landley <”…””}”(hj  h²hh³Nh´NubhŒ	reference”“”)”}”(hŒrob@landley.net”h]”hŒrob@landley.net”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”Œmailto:rob@landley.net”uh1j  hj  ubhŒ>”…””}”(hj  h²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K	hj	  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j  hhôubeh}”(h]”h ]”h"]”h$]”h&]”uh1hòh³hÇh´K	hhïh²hubah}”(h]”h ]”h"]”h$]”h&]”uh1híhhÊh²hh³hÇh´K	ubhÉ)”}”(hhh]”(hÎ)”}”(hŒWhat is ramfs?”h]”hŒWhat is ramfs?”…””}”(hjF  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhjC  h²hh³hÇh´KubhÞ)”}”(hŒ¡Ramfs is a very simple filesystem that exports Linux's disk caching
mechanisms (the page cache and dentry cache) as a dynamically resizable
RAM-based filesystem.”h]”hŒ£Ramfs is a very simple filesystem that exports Linuxâ€™s disk caching
mechanisms (the page cache and dentry cache) as a dynamically resizable
RAM-based filesystem.”…””}”(hjT  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KhjC  h²hubhÞ)”}”(hX  Normally all files are cached in memory by Linux.  Pages of data read from
backing store (usually the block device the filesystem is mounted on) are kept
around in case it's needed again, but marked as clean (freeable) in case the
Virtual Memory system needs the memory for something else.  Similarly, data
written to files is marked clean as soon as it has been written to backing
store, but kept around for caching purposes until the VM reallocates the
memory.  A similar mechanism (the dentry cache) greatly speeds up access to
directories.”h]”hX!  Normally all files are cached in memory by Linux.  Pages of data read from
backing store (usually the block device the filesystem is mounted on) are kept
around in case itâ€™s needed again, but marked as clean (freeable) in case the
Virtual Memory system needs the memory for something else.  Similarly, data
written to files is marked clean as soon as it has been written to backing
store, but kept around for caching purposes until the VM reallocates the
memory.  A similar mechanism (the dentry cache) greatly speeds up access to
directories.”…””}”(hjb  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KhjC  h²hubhÞ)”}”(hX  With ramfs, there is no backing store.  Files written into ramfs allocate
dentries and page cache as usual, but there's nowhere to write them to.
This means the pages are never marked clean, so they can't be freed by the
VM when it's looking to recycle memory.”h]”hX
  With ramfs, there is no backing store.  Files written into ramfs allocate
dentries and page cache as usual, but thereâ€™s nowhere to write them to.
This means the pages are never marked clean, so they canâ€™t be freed by the
VM when itâ€™s looking to recycle memory.”…””}”(hjp  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KhjC  h²hubhÞ)”}”(hX<  The amount of code required to implement ramfs is tiny, because all the
work is done by the existing Linux caching infrastructure.  Basically,
you're mounting the disk cache as a filesystem.  Because of this, ramfs is not
an optional component removable via menuconfig, since there would be negligible
space savings.”h]”hX>  The amount of code required to implement ramfs is tiny, because all the
work is done by the existing Linux caching infrastructure.  Basically,
youâ€™re mounting the disk cache as a filesystem.  Because of this, ramfs is not
an optional component removable via menuconfig, since there would be negligible
space savings.”…””}”(hj~  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K hjC  h²hubeh}”(h]”Œwhat-is-ramfs”ah ]”h"]”Œwhat is ramfs?”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´KubhÉ)”}”(hhh]”(hÎ)”}”(hŒramfs and ramdisk:”h]”hŒramfs and ramdisk:”…””}”(hj—  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhj”  h²hh³hÇh´K'ubhÞ)”}”(hXì  The older "ram disk" mechanism created a synthetic block device out of
an area of RAM and used it as backing store for a filesystem.  This block
device was of fixed size, so the filesystem mounted on it was of fixed
size.  Using a ram disk also required unnecessarily copying memory from the
fake block device into the page cache (and copying changes back out), as well
as creating and destroying dentries.  Plus it needed a filesystem driver
(such as ext2) to format and interpret this data.”h]”hXð  The older â€œram diskâ€ mechanism created a synthetic block device out of
an area of RAM and used it as backing store for a filesystem.  This block
device was of fixed size, so the filesystem mounted on it was of fixed
size.  Using a ram disk also required unnecessarily copying memory from the
fake block device into the page cache (and copying changes back out), as well
as creating and destroying dentries.  Plus it needed a filesystem driver
(such as ext2) to format and interpret this data.”…””}”(hj¥  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K)hj”  h²hubhÞ)”}”(hXþ  Compared to ramfs, this wastes memory (and memory bus bandwidth), creates
unnecessary work for the CPU, and pollutes the CPU caches.  (There are tricks
to avoid this copying by playing with the page tables, but they're unpleasantly
complicated and turn out to be about as expensive as the copying anyway.)
More to the point, all the work ramfs is doing has to happen _anyway_,
since all file access goes through the page and dentry caches.  The RAM
disk is simply unnecessary; ramfs is internally much simpler.”h]”hX   Compared to ramfs, this wastes memory (and memory bus bandwidth), creates
unnecessary work for the CPU, and pollutes the CPU caches.  (There are tricks
to avoid this copying by playing with the page tables, but theyâ€™re unpleasantly
complicated and turn out to be about as expensive as the copying anyway.)
More to the point, all the work ramfs is doing has to happen _anyway_,
since all file access goes through the page and dentry caches.  The RAM
disk is simply unnecessary; ramfs is internally much simpler.”…””}”(hj³  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K1hj”  h²hubhÞ)”}”(hŒòAnother reason ramdisks are semi-obsolete is that the introduction of
loopback devices offered a more flexible and convenient way to create
synthetic block devices, now from files instead of from chunks of memory.
See losetup (8) for details.”h]”hŒòAnother reason ramdisks are semi-obsolete is that the introduction of
loopback devices offered a more flexible and convenient way to create
synthetic block devices, now from files instead of from chunks of memory.
See losetup (8) for details.”…””}”(hjÁ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K9hj”  h²hubeh}”(h]”Œramfs-and-ramdisk”ah ]”h"]”Œramfs and ramdisk:”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´K'ubhÉ)”}”(hhh]”(hÎ)”}”(hŒramfs and tmpfs:”h]”hŒramfs and tmpfs:”…””}”(hjÚ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhj×  h²hh³hÇh´K?ubhÞ)”}”(hXY  One downside of ramfs is you can keep writing data into it until you fill
up all memory, and the VM can't free it because the VM thinks that files
should get written to backing store (rather than swap space), but ramfs hasn't
got any backing store.  Because of this, only root (or a trusted user) should
be allowed write access to a ramfs mount.”h]”hX]  One downside of ramfs is you can keep writing data into it until you fill
up all memory, and the VM canâ€™t free it because the VM thinks that files
should get written to backing store (rather than swap space), but ramfs hasnâ€™t
got any backing store.  Because of this, only root (or a trusted user) should
be allowed write access to a ramfs mount.”…””}”(hjè  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KAhj×  h²hubhÞ)”}”(hŒêA ramfs derivative called tmpfs was created to add size limits, and the ability
to write the data to swap space.  Normal users can be allowed write access to
tmpfs mounts.  See Documentation/filesystems/tmpfs.rst for more information.”h]”hŒêA ramfs derivative called tmpfs was created to add size limits, and the ability
to write the data to swap space.  Normal users can be allowed write access to
tmpfs mounts.  See Documentation/filesystems/tmpfs.rst for more information.”…””}”(hjö  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KGhj×  h²hubeh}”(h]”Œramfs-and-tmpfs”ah ]”h"]”Œramfs and tmpfs:”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´K?ubhÉ)”}”(hhh]”(hÎ)”}”(hŒWhat is rootfs?”h]”hŒWhat is rootfs?”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhj  h²hh³hÇh´KLubhÞ)”}”(hXZ  Rootfs is a special instance of ramfs (or tmpfs, if that's enabled), which is
always present in Linux systems.  The kernel uses an immutable empty filesystem
called nullfs as the true root of the VFS hierarchy, with the mutable rootfs
(tmpfs/ramfs) mounted on top of it.  This allows pivot_root() and unmounting
of the initramfs to work normally.”h]”hX\  Rootfs is a special instance of ramfs (or tmpfs, if thatâ€™s enabled), which is
always present in Linux systems.  The kernel uses an immutable empty filesystem
called nullfs as the true root of the VFS hierarchy, with the mutable rootfs
(tmpfs/ramfs) mounted on top of it.  This allows pivot_root() and unmounting
of the initramfs to work normally.”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KNhj  h²hubhÞ)”}”(hŒ‡Most systems just mount another filesystem over rootfs and ignore it.  The
amount of space an empty instance of ramfs takes up is tiny.”h]”hŒ‡Most systems just mount another filesystem over rootfs and ignore it.  The
amount of space an empty instance of ramfs takes up is tiny.”…””}”(hj+  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KThj  h²hubhÞ)”}”(hŒ’If CONFIG_TMPFS is enabled, rootfs will use tmpfs instead of ramfs by
default.  To force ramfs, add "rootfstype=ramfs" to the kernel command
line.”h]”hŒ–If CONFIG_TMPFS is enabled, rootfs will use tmpfs instead of ramfs by
default.  To force ramfs, add â€œrootfstype=ramfsâ€ to the kernel command
line.”…””}”(hj9  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KWhj  h²hubeh}”(h]”Œwhat-is-rootfs”ah ]”h"]”Œwhat is rootfs?”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´KLubhÉ)”}”(hhh]”(hÎ)”}”(hŒWhat is initramfs?”h]”hŒWhat is initramfs?”…””}”(hjR  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhjO  h²hh³hÇh´K\ubhÞ)”}”(hXl  All 2.6 Linux kernels contain a gzipped "cpio" format archive, which is
extracted into rootfs when the kernel boots up.  After extracting, the kernel
checks to see if rootfs contains a file "init", and if so it executes it as PID
1.  If found, this init process is responsible for bringing the system the
rest of the way up, including locating and mounting the real root device (if
any).  If rootfs does not contain an init program after the embedded cpio
archive is extracted into it, the kernel will fall through to the older code
to locate and mount a root partition, then exec some variant of /sbin/init
out of that.”h]”hXt  All 2.6 Linux kernels contain a gzipped â€œcpioâ€ format archive, which is
extracted into rootfs when the kernel boots up.  After extracting, the kernel
checks to see if rootfs contains a file â€œinitâ€, and if so it executes it as PID
1.  If found, this init process is responsible for bringing the system the
rest of the way up, including locating and mounting the real root device (if
any).  If rootfs does not contain an init program after the embedded cpio
archive is extracted into it, the kernel will fall through to the older code
to locate and mount a root partition, then exec some variant of /sbin/init
out of that.”…””}”(hj`  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K^hjO  h²hubhÞ)”}”(hŒ5All this differs from the old initrd in several ways:”h]”hŒ5All this differs from the old initrd in several ways:”…””}”(hjn  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KhhjO  h²hubhŒblock_quote”“”)”}”(hXH  - The old initrd was always a separate file, while the initramfs archive is
  linked into the linux kernel image.  (The directory ``linux-*/usr`` is
  devoted to generating this archive during the build.)

- The old initrd file was a gzipped filesystem image (in some file format,
  such as ext2, that needed a driver built into the kernel), while the new
  initramfs archive is a gzipped cpio archive (like tar only simpler,
  see cpio(1) and Documentation/driver-api/early-userspace/buffer-format.rst).
  The kernel's cpio extraction code is not only extremely small, it's also
  __init text and data that can be discarded during the boot process.

- The program run by the old initrd (which was called /initrd, not /init) did
  some setup and then returned to the kernel, while the init program from
  initramfs is not expected to return to the kernel.  (If /init needs to hand
  off control it can overmount / with a new root device and exec another init
  program.  See the switch_root utility, below.)

- When switching another root device, initrd would pivot_root and then
  umount the ramdisk.  With nullfs as the true root, pivot_root() works
  normally from the initramfs.  Userspace can simply do::

    chdir(new_root);
    pivot_root(".", ".");
    umount2(".", MNT_DETACH);

  This is the preferred method for switching root filesystems.
”h]”hŒbullet_list”“”)”}”(hhh]”(hŒ	list_item”“”)”}”(hŒÇThe old initrd was always a separate file, while the initramfs archive is
linked into the linux kernel image.  (The directory ``linux-*/usr`` is
devoted to generating this archive during the build.)
”h]”hÞ)”}”(hŒÆThe old initrd was always a separate file, while the initramfs archive is
linked into the linux kernel image.  (The directory ``linux-*/usr`` is
devoted to generating this archive during the build.)”h]”(hŒ~The old initrd was always a separate file, while the initramfs archive is
linked into the linux kernel image.  (The directory ”…””}”(hj  h²hh³Nh´NubhŒliteral”“”)”}”(hŒ``linux-*/usr``”h]”hŒlinux-*/usr”…””}”(hj—  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j•  hj  ubhŒ9 is
devoted to generating this archive during the build.)”…””}”(hj  h²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Kjhj‰  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj„  ubjˆ  )”}”(hX°  The old initrd file was a gzipped filesystem image (in some file format,
such as ext2, that needed a driver built into the kernel), while the new
initramfs archive is a gzipped cpio archive (like tar only simpler,
see cpio(1) and Documentation/driver-api/early-userspace/buffer-format.rst).
The kernel's cpio extraction code is not only extremely small, it's also
__init text and data that can be discarded during the boot process.
”h]”hÞ)”}”(hX¯  The old initrd file was a gzipped filesystem image (in some file format,
such as ext2, that needed a driver built into the kernel), while the new
initramfs archive is a gzipped cpio archive (like tar only simpler,
see cpio(1) and Documentation/driver-api/early-userspace/buffer-format.rst).
The kernel's cpio extraction code is not only extremely small, it's also
__init text and data that can be discarded during the boot process.”h]”hX³  The old initrd file was a gzipped filesystem image (in some file format,
such as ext2, that needed a driver built into the kernel), while the new
initramfs archive is a gzipped cpio archive (like tar only simpler,
see cpio(1) and Documentation/driver-api/early-userspace/buffer-format.rst).
The kernelâ€™s cpio extraction code is not only extremely small, itâ€™s also
__init text and data that can be discarded during the boot process.”…””}”(hj¹  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Knhjµ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj„  ubjˆ  )”}”(hX[  The program run by the old initrd (which was called /initrd, not /init) did
some setup and then returned to the kernel, while the init program from
initramfs is not expected to return to the kernel.  (If /init needs to hand
off control it can overmount / with a new root device and exec another init
program.  See the switch_root utility, below.)
”h]”hÞ)”}”(hXZ  The program run by the old initrd (which was called /initrd, not /init) did
some setup and then returned to the kernel, while the init program from
initramfs is not expected to return to the kernel.  (If /init needs to hand
off control it can overmount / with a new root device and exec another init
program.  See the switch_root utility, below.)”h]”hXZ  The program run by the old initrd (which was called /initrd, not /init) did
some setup and then returned to the kernel, while the init program from
initramfs is not expected to return to the kernel.  (If /init needs to hand
off control it can overmount / with a new root device and exec another init
program.  See the switch_root utility, below.)”…””}”(hjÑ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KuhjÍ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj„  ubjˆ  )”}”(hXI  When switching another root device, initrd would pivot_root and then
umount the ramdisk.  With nullfs as the true root, pivot_root() works
normally from the initramfs.  Userspace can simply do::

  chdir(new_root);
  pivot_root(".", ".");
  umount2(".", MNT_DETACH);

This is the preferred method for switching root filesystems.
”h]”(hÞ)”}”(hŒÂWhen switching another root device, initrd would pivot_root and then
umount the ramdisk.  With nullfs as the true root, pivot_root() works
normally from the initramfs.  Userspace can simply do::”h]”hŒÁWhen switching another root device, initrd would pivot_root and then
umount the ramdisk.  With nullfs as the true root, pivot_root() works
normally from the initramfs.  Userspace can simply do:”…””}”(hjé  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K{hjå  ubhŒliteral_block”“”)”}”(hŒ@chdir(new_root);
pivot_root(".", ".");
umount2(".", MNT_DETACH);”h]”hŒ@chdir(new_root);
pivot_root(".", ".");
umount2(".", MNT_DETACH);”…””}”hjù  sbah}”(h]”h ]”h"]”h$]”h&]”hÅhÆuh1j÷  h³hÇh´Khjå  ubhÞ)”}”(hŒ<This is the preferred method for switching root filesystems.”h]”hŒ<This is the preferred method for switching root filesystems.”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Kƒhjå  ubeh}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj„  ubeh}”(h]”h ]”h"]”h$]”h&]”Œbullet”Œ-”uh1j‚  h³hÇh´Kjhj~  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j|  h³hÇh´KjhjO  h²hubeh}”(h]”Œwhat-is-initramfs”ah ]”h"]”Œwhat is initramfs?”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´K\ubhÉ)”}”(hhh]”(hÎ)”}”(hŒPopulating initramfs:”h]”hŒPopulating initramfs:”…””}”(hj4  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhj1  h²hh³hÇh´K†ubhÞ)”}”(hŒÃThe 2.6 kernel build process always creates a gzipped cpio format initramfs
archive and links it into the resulting kernel binary.  By default, this
archive is empty (consuming 134 bytes on x86).”h]”hŒÃThe 2.6 kernel build process always creates a gzipped cpio format initramfs
archive and links it into the resulting kernel binary.  By default, this
archive is empty (consuming 134 bytes on x86).”…””}”(hjB  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Kˆhj1  h²hubhÞ)”}”(hXŠ  The config option CONFIG_INITRAMFS_SOURCE (in General Setup in menuconfig,
and living in usr/Kconfig) can be used to specify a source for the
initramfs archive, which will automatically be incorporated into the
resulting binary.  This option can point to an existing gzipped cpio
archive, a directory containing files to be archived, or a text file
specification such as the following example::”h]”hX‰  The config option CONFIG_INITRAMFS_SOURCE (in General Setup in menuconfig,
and living in usr/Kconfig) can be used to specify a source for the
initramfs archive, which will automatically be incorporated into the
resulting binary.  This option can point to an existing gzipped cpio
archive, a directory containing files to be archived, or a text file
specification such as the following example:”…””}”(hjP  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KŒhj1  h²hubjø  )”}”(hX  dir /dev 755 0 0
nod /dev/console 644 0 0 c 5 1
nod /dev/loop0 644 0 0 b 7 0
dir /bin 755 1000 1000
slink /bin/sh busybox 777 0 0
file /bin/busybox initramfs/busybox 755 0 0
dir /proc 755 0 0
dir /sys 755 0 0
dir /mnt 755 0 0
file /init initramfs/init.sh 755 0 0”h]”hX  dir /dev 755 0 0
nod /dev/console 644 0 0 c 5 1
nod /dev/loop0 644 0 0 b 7 0
dir /bin 755 1000 1000
slink /bin/sh busybox 777 0 0
file /bin/busybox initramfs/busybox 755 0 0
dir /proc 755 0 0
dir /sys 755 0 0
dir /mnt 755 0 0
file /init initramfs/init.sh 755 0 0”…””}”hj^  sbah}”(h]”h ]”h"]”h$]”h&]”hÅhÆuh1j÷  h³hÇh´K“hj1  h²hubhÞ)”}”(hŒjRun "usr/gen_init_cpio" (after the kernel build) to get a usage message
documenting the above file format.”h]”hŒnRun â€œusr/gen_init_cpioâ€ (after the kernel build) to get a usage message
documenting the above file format.”…””}”(hjl  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Kžhj1  h²hubhÞ)”}”(hXŠ  One advantage of the configuration file is that root access is not required to
set permissions or create device nodes in the new archive.  (Note that those
two example "file" entries expect to find files named "init.sh" and "busybox" in
a directory called "initramfs", under the linux-2.6.* directory.  See
Documentation/driver-api/early-userspace/early_userspace_support.rst for more details.)”h]”hXš  One advantage of the configuration file is that root access is not required to
set permissions or create device nodes in the new archive.  (Note that those
two example â€œfileâ€ entries expect to find files named â€œinit.shâ€ and â€œbusyboxâ€ in
a directory called â€œinitramfsâ€, under the linux-2.6.* directory.  See
Documentation/driver-api/early-userspace/early_userspace_support.rst for more details.)”…””}”(hjz  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K¡hj1  h²hubhÞ)”}”(hX  The kernel does not depend on external cpio tools.  If you specify a
directory instead of a configuration file, the kernel's build infrastructure
creates a configuration file from that directory (usr/Makefile calls
usr/gen_initramfs.sh), and proceeds to package up that directory
using the config file (by feeding it to usr/gen_init_cpio, which is created
from usr/gen_init_cpio.c).  The kernel's build-time cpio creation code is
entirely self-contained, and the kernel's boot-time extractor is also
(obviously) self-contained.”h]”hX  The kernel does not depend on external cpio tools.  If you specify a
directory instead of a configuration file, the kernelâ€™s build infrastructure
creates a configuration file from that directory (usr/Makefile calls
usr/gen_initramfs.sh), and proceeds to package up that directory
using the config file (by feeding it to usr/gen_init_cpio, which is created
from usr/gen_init_cpio.c).  The kernelâ€™s build-time cpio creation code is
entirely self-contained, and the kernelâ€™s boot-time extractor is also
(obviously) self-contained.”…””}”(hjˆ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K§hj1  h²hubhÞ)”}”(hŒÁThe one thing you might need external cpio utilities installed for is creating
or extracting your own preprepared cpio files to feed to the kernel build
(instead of a config file or directory).”h]”hŒÁThe one thing you might need external cpio utilities installed for is creating
or extracting your own preprepared cpio files to feed to the kernel build
(instead of a config file or directory).”…””}”(hj–  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K°hj1  h²hubhÞ)”}”(hŒ‡The following command line can extract a cpio image (either by the above script
or by the kernel build) back into its component files::”h]”hŒ†The following command line can extract a cpio image (either by the above script
or by the kernel build) back into its component files:”…””}”(hj¤  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K´hj1  h²hubjø  )”}”(hŒAcpio -i -d -H newc -F initramfs_data.cpio --no-absolute-filenames”h]”hŒAcpio -i -d -H newc -F initramfs_data.cpio --no-absolute-filenames”…””}”hj²  sbah}”(h]”h ]”h"]”h$]”h&]”hÅhÆuh1j÷  h³hÇh´K·hj1  h²hubhÞ)”}”(hŒmThe following shell script can create a prebuilt cpio archive you can
use in place of the above config file::”h]”hŒlThe following shell script can create a prebuilt cpio archive you can
use in place of the above config file:”…””}”(hjÀ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K¹hj1  h²hubjø  )”}”(hXn  #!/bin/sh

# Copyright 2006 Rob Landley <rob@landley.net> and TimeSys Corporation.
# Licensed under GPL version 2

if [ $# -ne 2 ]
then
  echo "usage: mkinitramfs directory imagename.cpio.gz"
  exit 1
fi

if [ -d "$1" ]
then
  echo "creating $2 from $1"
  (cd "$1"; find . | cpio -o -H newc | gzip) > "$2"
else
  echo "First argument must be a directory"
  exit 1
fi”h]”hXn  #!/bin/sh

# Copyright 2006 Rob Landley <rob@landley.net> and TimeSys Corporation.
# Licensed under GPL version 2

if [ $# -ne 2 ]
then
  echo "usage: mkinitramfs directory imagename.cpio.gz"
  exit 1
fi

if [ -d "$1" ]
then
  echo "creating $2 from $1"
  (cd "$1"; find . | cpio -o -H newc | gzip) > "$2"
else
  echo "First argument must be a directory"
  exit 1
fi”…””}”hjÎ  sbah}”(h]”h ]”h"]”h$]”h&]”hÅhÆuh1j÷  h³hÇh´K¼hj1  h²hubhŒnote”“”)”}”(hXW  The cpio man page contains some bad advice that will break your initramfs
archive if you follow it.  It says "A typical way to generate the list
of filenames is with the find command; you should give find the -depth
option to minimize problems with permissions on directories that are
unwritable or not searchable."  Don't do this when creating
initramfs.cpio.gz images, it won't work.  The Linux kernel cpio extractor
won't create files in a directory that doesn't exist, so the directory
entries must go before the files that go in those directories.
The above script gets them in the right order.”h]”hÞ)”}”(hXW  The cpio man page contains some bad advice that will break your initramfs
archive if you follow it.  It says "A typical way to generate the list
of filenames is with the find command; you should give find the -depth
option to minimize problems with permissions on directories that are
unwritable or not searchable."  Don't do this when creating
initramfs.cpio.gz images, it won't work.  The Linux kernel cpio extractor
won't create files in a directory that doesn't exist, so the directory
entries must go before the files that go in those directories.
The above script gets them in the right order.”h]”hXc  The cpio man page contains some bad advice that will break your initramfs
archive if you follow it.  It says â€œA typical way to generate the list
of filenames is with the find command; you should give find the -depth
option to minimize problems with permissions on directories that are
unwritable or not searchable.â€  Donâ€™t do this when creating
initramfs.cpio.gz images, it wonâ€™t work.  The Linux kernel cpio extractor
wonâ€™t create files in a directory that doesnâ€™t exist, so the directory
entries must go before the files that go in those directories.
The above script gets them in the right order.”…””}”(hjâ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KÒhjÞ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1jÜ  hj1  h²hh³hÇh´Nubeh}”(h]”Œpopulating-initramfs”ah ]”h"]”Œpopulating initramfs:”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´K†ubhÉ)”}”(hhh]”(hÎ)”}”(hŒExternal initramfs images:”h]”hŒExternal initramfs images:”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhjþ  h²hh³hÇh´KÝubhÞ)”}”(hX  If the kernel has initrd support enabled, an external cpio.gz archive can also
be passed into a 2.6 kernel in place of an initrd.  In this case, the kernel
will autodetect the type (initramfs, not initrd) and extract the external cpio
archive into rootfs before trying to run /init.”h]”hX  If the kernel has initrd support enabled, an external cpio.gz archive can also
be passed into a 2.6 kernel in place of an initrd.  In this case, the kernel
will autodetect the type (initramfs, not initrd) and extract the external cpio
archive into rootfs before trying to run /init.”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Kßhjþ  h²hubhÞ)”}”(hX  This has the memory efficiency advantages of initramfs (no ramdisk block
device) but the separate packaging of initrd (which is nice if you have
non-GPL code you'd like to run from initramfs, without conflating it with
the GPL licensed Linux kernel binary).”h]”hX  This has the memory efficiency advantages of initramfs (no ramdisk block
device) but the separate packaging of initrd (which is nice if you have
non-GPL code youâ€™d like to run from initramfs, without conflating it with
the GPL licensed Linux kernel binary).”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Kähjþ  h²hubhÞ)”}”(hX/  It can also be used to supplement the kernel's built-in initramfs image.  The
files in the external archive will overwrite any conflicting files in
the built-in initramfs archive.  Some distributors also prefer to customize
a single kernel image with task-specific initramfs images, without recompiling.”h]”hX1  It can also be used to supplement the kernelâ€™s built-in initramfs image.  The
files in the external archive will overwrite any conflicting files in
the built-in initramfs archive.  Some distributors also prefer to customize
a single kernel image with task-specific initramfs images, without recompiling.”…””}”(hj+  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Kéhjþ  h²hubeh}”(h]”Œexternal-initramfs-images”ah ]”h"]”Œexternal initramfs images:”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´KÝubhÉ)”}”(hhh]”(hÎ)”}”(hŒContents of initramfs:”h]”hŒContents of initramfs:”…””}”(hjD  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhjA  h²hh³hÇh´KïubhÞ)”}”(hŒêAn initramfs archive is a complete self-contained root filesystem for Linux.
If you don't already understand what shared libraries, devices, and paths
you need to get a minimal root filesystem up and running, here are some
references:”h]”hŒìAn initramfs archive is a complete self-contained root filesystem for Linux.
If you donâ€™t already understand what shared libraries, devices, and paths
you need to get a minimal root filesystem up and running, here are some
references:”…””}”(hjR  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KñhjA  h²hubjƒ  )”}”(hhh]”(jˆ  )”}”(hŒ*https://www.tldp.org/HOWTO/Bootdisk-HOWTO/”h]”hÞ)”}”(hje  h]”j  )”}”(hje  h]”hŒ*https://www.tldp.org/HOWTO/Bootdisk-HOWTO/”…””}”(hjj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”je  uh1j  hjg  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Köhjc  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj`  h²hh³hÇh´Nubjˆ  )”}”(hŒAhttps://www.tldp.org/HOWTO/From-PowerUp-To-Bash-Prompt-HOWTO.html”h]”hÞ)”}”(hj†  h]”j  )”}”(hj†  h]”hŒAhttps://www.tldp.org/HOWTO/From-PowerUp-To-Bash-Prompt-HOWTO.html”…””}”(hj‹  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j†  uh1j  hjˆ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´K÷hj„  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj`  h²hh³hÇh´Nubjˆ  )”}”(hŒ1http://www.linuxfromscratch.org/lfs/view/stable/
”h]”hÞ)”}”(hŒ0http://www.linuxfromscratch.org/lfs/view/stable/”h]”j  )”}”(hj«  h]”hŒ0http://www.linuxfromscratch.org/lfs/view/stable/”…””}”(hj­  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j«  uh1j  hj©  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Køhj¥  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj`  h²hh³hÇh´Nubeh}”(h]”h ]”h"]”h$]”h&]”j!  j"  uh1j‚  h³hÇh´KöhjA  h²hubhÞ)”}”(hŒÍThe "klibc" package (https://www.kernel.org/pub/linux/libs/klibc) is
designed to be a tiny C library to statically link early userspace
code against, along with some related utilities.  It is BSD licensed.”h]”(hŒThe â€œklibcâ€ package (”…””}”(hjÍ  h²hh³Nh´Nubj  )”}”(hŒ+https://www.kernel.org/pub/linux/libs/klibc”h]”hŒ+https://www.kernel.org/pub/linux/libs/klibc”…””}”(hjÕ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j×  uh1j  hjÍ  ubhŒ) is
designed to be a tiny C library to statically link early userspace
code against, along with some related utilities.  It is BSD licensed.”…””}”(hjÍ  h²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KúhjA  h²hubhÞ)”}”(hŒÈI use uClibc (https://www.uclibc.org) and busybox (https://www.busybox.net)
myself.  These are LGPL and GPL, respectively.  (A self-contained initramfs
package is planned for the busybox 1.3 release.)”h]”(hŒI use uClibc (”…””}”(hjî  h²hh³Nh´Nubj  )”}”(hŒhttps://www.uclibc.org”h]”hŒhttps://www.uclibc.org”…””}”(hjö  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”jø  uh1j  hjî  ubhŒ) and busybox (”…””}”(hjî  h²hh³Nh´Nubj  )”}”(hŒhttps://www.busybox.net”h]”hŒhttps://www.busybox.net”…””}”(hj	  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j  uh1j  hjî  ubhŒ~)
myself.  These are LGPL and GPL, respectively.  (A self-contained initramfs
package is planned for the busybox 1.3 release.)”…””}”(hjî  h²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´KþhjA  h²hubhÞ)”}”(hX  In theory you could use glibc, but that's not well suited for small embedded
uses like this.  (A "hello world" program statically linked against glibc is
over 400k.  With uClibc it's 7k.  Also note that glibc dlopens libnss to do
name lookups, even when otherwise statically linked.)”h]”hX#  In theory you could use glibc, but thatâ€™s not well suited for small embedded
uses like this.  (A â€œhello worldâ€ program statically linked against glibc is
over 400k.  With uClibc itâ€™s 7k.  Also note that glibc dlopens libnss to do
name lookups, even when otherwise statically linked.)”…””}”(hj"  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MhjA  h²hubhÞ)”}”(hŒ´A good first step is to get initramfs to run a statically linked "hello world"
program as init, and test it under an emulator like qemu (www.qemu.org) or
User Mode Linux, like so::”h]”hŒ·A good first step is to get initramfs to run a statically linked â€œhello worldâ€
program as init, and test it under an emulator like qemu (www.qemu.org) or
User Mode Linux, like so:”…””}”(hj0  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MhjA  h²hubjø  )”}”(hX^  cat > hello.c << EOF
#include <stdio.h>
#include <unistd.h>

int main(int argc, char *argv[])
{
  printf("Hello world!\n");
  sleep(999999999);
}
EOF
gcc -static hello.c -o init
echo init | cpio -o -H newc | gzip > test.cpio.gz
# Testing external initramfs using the initrd loading mechanism.
qemu -kernel /boot/vmlinuz -initrd test.cpio.gz /dev/zero”h]”hX^  cat > hello.c << EOF
#include <stdio.h>
#include <unistd.h>

int main(int argc, char *argv[])
{
  printf("Hello world!\n");
  sleep(999999999);
}
EOF
gcc -static hello.c -o init
echo init | cpio -o -H newc | gzip > test.cpio.gz
# Testing external initramfs using the initrd loading mechanism.
qemu -kernel /boot/vmlinuz -initrd test.cpio.gz /dev/zero”…””}”hj>  sbah}”(h]”h ]”h"]”h$]”h&]”hÅhÆuh1j÷  h³hÇh´MhjA  h²hubhÞ)”}”(hŒ¢When debugging a normal root filesystem, it's nice to be able to boot with
"init=/bin/sh".  The initramfs equivalent is "rdinit=/bin/sh", and it's
just as useful.”h]”hŒ®When debugging a normal root filesystem, itâ€™s nice to be able to boot with
â€œinit=/bin/shâ€.  The initramfs equivalent is â€œrdinit=/bin/shâ€, and itâ€™s
just as useful.”…””}”(hjL  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MhjA  h²hubeh}”(h]”Œcontents-of-initramfs”ah ]”h"]”Œcontents of initramfs:”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´KïubhÉ)”}”(hhh]”(hÎ)”}”(hŒWhy cpio rather than tar?”h]”hŒWhy cpio rather than tar?”…””}”(hje  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhjb  h²hh³hÇh´MubhÞ)”}”(hŒLThis decision was made back in December, 2001.  The discussion started here:”h]”hŒLThis decision was made back in December, 2001.  The discussion started here:”…””}”(hjs  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M!hjb  h²hubjƒ  )”}”(hhh]”jˆ  )”}”(hŒ@https://lore.kernel.org/lkml/a03cke$640$1@cesium.transmeta.com/
”h]”hÞ)”}”(hŒ?https://lore.kernel.org/lkml/a03cke$640$1@cesium.transmeta.com/”h]”j  )”}”(hjŠ  h]”hŒ?https://lore.kernel.org/lkml/a03cke$640$1@cesium.transmeta.com/”…””}”(hjŒ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”jŠ  uh1j  hjˆ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M#hj„  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj  h²hh³hÇh´Nubah}”(h]”h ]”h"]”h$]”h&]”j!  j"  uh1j‚  h³hÇh´M#hjb  h²hubhÞ)”}”(hŒIAnd spawned a second thread (specifically on tar vs cpio), starting here:”h]”hŒIAnd spawned a second thread (specifically on tar vs cpio), starting here:”…””}”(hj¬  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M%hjb  h²hubjƒ  )”}”(hhh]”jˆ  )”}”(hŒ9https://lore.kernel.org/lkml/3C25A06D.7030408@zytor.com/
”h]”hÞ)”}”(hŒ8https://lore.kernel.org/lkml/3C25A06D.7030408@zytor.com/”h]”j  )”}”(hjÃ  h]”hŒ8https://lore.kernel.org/lkml/3C25A06D.7030408@zytor.com/”…””}”(hjÅ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”jÃ  uh1j  hjÁ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M'hj½  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hjº  h²hh³hÇh´Nubah}”(h]”h ]”h"]”h$]”h&]”j!  j"  uh1j‚  h³hÇh´M'hjb  h²hubhÞ)”}”(hŒ^The quick and dirty summary version (which is no substitute for reading
the above threads) is:”h]”hŒ^The quick and dirty summary version (which is no substitute for reading
the above threads) is:”…””}”(hjå  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M)hjb  h²hubhŒenumerated_list”“”)”}”(hhh]”(jˆ  )”}”(hXò  cpio is a standard.  It's decades old (from the AT&T days), and already
widely used on Linux (inside RPM, Red Hat's device driver disks).  Here's
a Linux Journal article about it from 1996:

   http://www.linuxjournal.com/article/1213

It's not as popular as tar because the traditional cpio command line tools
require _truly_hideous_ command line arguments.  But that says nothing
either way about the archive format, and there are alternative tools,
such as:

   https://linux.die.net/man/1/afio
”h]”(hÞ)”}”(hŒ½cpio is a standard.  It's decades old (from the AT&T days), and already
widely used on Linux (inside RPM, Red Hat's device driver disks).  Here's
a Linux Journal article about it from 1996:”h]”hŒÃcpio is a standard.  Itâ€™s decades old (from the AT&T days), and already
widely used on Linux (inside RPM, Red Hatâ€™s device driver disks).  Hereâ€™s
a Linux Journal article about it from 1996:”…””}”(hjü  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M,hjø  ubj}  )”}”(hŒ)http://www.linuxjournal.com/article/1213
”h]”hÞ)”}”(hŒ(http://www.linuxjournal.com/article/1213”h]”j  )”}”(hj  h]”hŒ(http://www.linuxjournal.com/article/1213”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j  uh1j  hj  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M0hj
  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j|  h³hÇh´M0hjø  ubhÞ)”}”(hŒàIt's not as popular as tar because the traditional cpio command line tools
require _truly_hideous_ command line arguments.  But that says nothing
either way about the archive format, and there are alternative tools,
such as:”h]”hŒâItâ€™s not as popular as tar because the traditional cpio command line tools
require _truly_hideous_ command line arguments.  But that says nothing
either way about the archive format, and there are alternative tools,
such as:”…””}”(hj,  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M2hjø  ubj}  )”}”(hŒ!https://linux.die.net/man/1/afio
”h]”hÞ)”}”(hŒ https://linux.die.net/man/1/afio”h]”j  )”}”(hj@  h]”hŒ https://linux.die.net/man/1/afio”…””}”(hjB  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j@  uh1j  hj>  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M7hj:  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j|  h³hÇh´M7hjø  ubeh}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hjõ  h²hh³hÇh´Nubjˆ  )”}”(hX‚  The cpio archive format chosen by the kernel is simpler and cleaner (and
thus easier to create and parse) than any of the (literally dozens of)
various tar archive formats.  The complete initramfs archive format is
explained in buffer-format.rst, created in usr/gen_init_cpio.c, and
extracted in init/initramfs.c.  All three together come to less than 26k
total of human-readable text.
”h]”hÞ)”}”(hX  The cpio archive format chosen by the kernel is simpler and cleaner (and
thus easier to create and parse) than any of the (literally dozens of)
various tar archive formats.  The complete initramfs archive format is
explained in buffer-format.rst, created in usr/gen_init_cpio.c, and
extracted in init/initramfs.c.  All three together come to less than 26k
total of human-readable text.”h]”hX  The cpio archive format chosen by the kernel is simpler and cleaner (and
thus easier to create and parse) than any of the (literally dozens of)
various tar archive formats.  The complete initramfs archive format is
explained in buffer-format.rst, created in usr/gen_init_cpio.c, and
extracted in init/initramfs.c.  All three together come to less than 26k
total of human-readable text.”…””}”(hjf  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M9hjb  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hjõ  h²hh³hÇh´Nubjˆ  )”}”(hŒ²The GNU project standardizing on tar is approximately as relevant as
Windows standardizing on zip.  Linux is not part of either, and is free
to make its own technical decisions.
”h]”hÞ)”}”(hŒ±The GNU project standardizing on tar is approximately as relevant as
Windows standardizing on zip.  Linux is not part of either, and is free
to make its own technical decisions.”h]”hŒ±The GNU project standardizing on tar is approximately as relevant as
Windows standardizing on zip.  Linux is not part of either, and is free
to make its own technical decisions.”…””}”(hj~  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M@hjz  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hjõ  h²hh³hÇh´Nubjˆ  )”}”(hŒãSince this is a kernel internal format, it could easily have been
something brand new.  The kernel provides its own tools to create and
extract this format anyway.  Using an existing standard was preferable,
but not essential.
”h]”hÞ)”}”(hŒâSince this is a kernel internal format, it could easily have been
something brand new.  The kernel provides its own tools to create and
extract this format anyway.  Using an existing standard was preferable,
but not essential.”h]”hŒâSince this is a kernel internal format, it could easily have been
something brand new.  The kernel provides its own tools to create and
extract this format anyway.  Using an existing standard was preferable,
but not essential.”…””}”(hj–  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MDhj’  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hjõ  h²hh³hÇh´Nubjˆ  )”}”(hXà  Al Viro made the decision (quote: "tar is ugly as hell and not going to be
supported on the kernel side"):

 - https://lore.kernel.org/lkml/Pine.GSO.4.21.0112222109050.21702-100000@weyl.math.psu.edu/

explained his reasoning:

 - https://lore.kernel.org/lkml/Pine.GSO.4.21.0112222240530.21702-100000@weyl.math.psu.edu/
 - https://lore.kernel.org/lkml/Pine.GSO.4.21.0112230849550.23300-100000@weyl.math.psu.edu/

and, most importantly, designed and implemented the initramfs code.
”h]”(hÞ)”}”(hŒjAl Viro made the decision (quote: "tar is ugly as hell and not going to be
supported on the kernel side"):”h]”hŒnAl Viro made the decision (quote: â€œtar is ugly as hell and not going to be
supported on the kernel sideâ€):”…””}”(hj®  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MIhjª  ubj}  )”}”(hŒ[- https://lore.kernel.org/lkml/Pine.GSO.4.21.0112222109050.21702-100000@weyl.math.psu.edu/
”h]”jƒ  )”}”(hhh]”jˆ  )”}”(hŒYhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112222109050.21702-100000@weyl.math.psu.edu/
”h]”hÞ)”}”(hŒXhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112222109050.21702-100000@weyl.math.psu.edu/”h]”j  )”}”(hjÉ  h]”hŒXhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112222109050.21702-100000@weyl.math.psu.edu/”…””}”(hjË  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”jÉ  uh1j  hjÇ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MLhjÃ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hjÀ  ubah}”(h]”h ]”h"]”h$]”h&]”j!  j"  uh1j‚  h³hÇh´MLhj¼  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j|  h³hÇh´MLhjª  ubhÞ)”}”(hŒexplained his reasoning:”h]”hŒexplained his reasoning:”…””}”(hjñ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MNhjª  ubj}  )”}”(hŒ¶- https://lore.kernel.org/lkml/Pine.GSO.4.21.0112222240530.21702-100000@weyl.math.psu.edu/
- https://lore.kernel.org/lkml/Pine.GSO.4.21.0112230849550.23300-100000@weyl.math.psu.edu/
”h]”jƒ  )”}”(hhh]”(jˆ  )”}”(hŒXhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112222240530.21702-100000@weyl.math.psu.edu/”h]”hÞ)”}”(hj  h]”j  )”}”(hj  h]”hŒXhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112222240530.21702-100000@weyl.math.psu.edu/”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j  uh1j  hj
  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MPhj  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj  ubjˆ  )”}”(hŒYhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112230849550.23300-100000@weyl.math.psu.edu/
”h]”hÞ)”}”(hŒXhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112230849550.23300-100000@weyl.math.psu.edu/”h]”j  )”}”(hj-  h]”hŒXhttps://lore.kernel.org/lkml/Pine.GSO.4.21.0112230849550.23300-100000@weyl.math.psu.edu/”…””}”(hj/  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”j-  uh1j  hj+  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MQhj'  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hj  ubeh}”(h]”h ]”h"]”h$]”h&]”j!  j"  uh1j‚  h³hÇh´MPhjÿ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j|  h³hÇh´MPhjª  ubhÞ)”}”(hŒCand, most importantly, designed and implemented the initramfs code.”h]”hŒCand, most importantly, designed and implemented the initramfs code.”…””}”(hjU  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MShjª  ubeh}”(h]”h ]”h"]”h$]”h&]”uh1j‡  hjõ  h²hh³hÇh´Nubeh}”(h]”h ]”h"]”h$]”h&]”Œenumtype”Œarabic”Œprefix”hŒsuffix”Œ)”uh1jó  hjb  h²hh³hÇh´M,ubeh}”(h]”Œwhy-cpio-rather-than-tar”ah ]”h"]”Œwhy cpio rather than tar?”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´MubhÉ)”}”(hhh]”(hÎ)”}”(hŒFuture directions:”h]”hŒFuture directions:”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÍhj|  h²hh³hÇh´MVubhÞ)”}”(hXV  Today (2.6.16), initramfs is always compiled in, but not always used.  The
kernel falls back to legacy boot code that is reached only if initramfs does
not contain an /init program.  The fallback is legacy code, there to ensure a
smooth transition and allowing early boot functionality to gradually move to
"early userspace" (I.E. initramfs).”h]”hXZ  Today (2.6.16), initramfs is always compiled in, but not always used.  The
kernel falls back to legacy boot code that is reached only if initramfs does
not contain an /init program.  The fallback is legacy code, there to ensure a
smooth transition and allowing early boot functionality to gradually move to
â€œearly userspaceâ€ (I.E. initramfs).”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´MXhj|  h²hubhÞ)”}”(hX#  The move to early userspace is necessary because finding and mounting the real
root device is complex.  Root partitions can span multiple devices (raid or
separate journal).  They can be out on the network (requiring dhcp, setting a
specific MAC address, logging into a server, etc).  They can live on removable
media, with dynamically allocated major/minor numbers and persistent naming
issues requiring a full udev implementation to sort out.  They can be
compressed, encrypted, copy-on-write, loopback mounted, strangely partitioned,
and so on.”h]”hX#  The move to early userspace is necessary because finding and mounting the real
root device is complex.  Root partitions can span multiple devices (raid or
separate journal).  They can be out on the network (requiring dhcp, setting a
specific MAC address, logging into a server, etc).  They can live on removable
media, with dynamically allocated major/minor numbers and persistent naming
issues requiring a full udev implementation to sort out.  They can be
compressed, encrypted, copy-on-write, loopback mounted, strangely partitioned,
and so on.”…””}”(hj›  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´M^hj|  h²hubhÞ)”}”(hŒÀThis kind of complexity (which inevitably includes policy) is rightly handled
in userspace.  Both klibc and busybox/uClibc are working on simple initramfs
packages to drop into a kernel build.”h]”hŒÀThis kind of complexity (which inevitably includes policy) is rightly handled
in userspace.  Both klibc and busybox/uClibc are working on simple initramfs
packages to drop into a kernel build.”…””}”(hj©  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Mghj|  h²hubhÞ)”}”(hŒôThe klibc package has now been accepted into Andrew Morton's 2.6.17-mm tree.
The kernel's current early boot code (partition detection, etc) will probably
be migrated into a default initramfs, automatically created and used by the
kernel build.”h]”hŒøThe klibc package has now been accepted into Andrew Mortonâ€™s 2.6.17-mm tree.
The kernelâ€™s current early boot code (partition detection, etc) will probably
be migrated into a default initramfs, automatically created and used by the
kernel build.”…””}”(hj·  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hÝh³hÇh´Mkhj|  h²hubeh}”(h]”Œfuture-directions”ah ]”h"]”Œfuture directions:”ah$]”h&]”uh1hÈhhÊh²hh³hÇh´MVubeh}”(h]”Œramfs-rootfs-and-initramfs”ah ]”h"]”Œramfs, rootfs and initramfs”ah$]”h&]”uh1hÈhhh²hh³hÇh´Kubeh}”(h]”h ]”h"]”h$]”h&]”Œsource”hÇuh1hŒcurrent_source”NŒcurrent_line”NŒsettings”Œdocutils.frontend”ŒValues”“”)”}”(hÍNŒ	generator”NŒ	datestamp”NŒsource_link”NŒ
source_url”NŒtoc_backlinks”Œentry”Œfootnote_backlinks”KŒsectnum_xform”KŒstrip_comments”NŒstrip_elements_with_classes”NŒstrip_classes”NŒreport_level”KŒ
halt_level”KŒexit_status_level”KŒdebug”NŒwarning_stream”NŒ	traceback”ˆŒinput_encoding”Œ	utf-8-sig”Œinput_encoding_error_handler”Œstrict”Œoutput_encoding”Œutf-8”Œoutput_encoding_error_handler”jø  Œerror_encoding”Œutf-8”Œerror_encoding_error_handler”Œbackslashreplace”Œlanguage_code”Œen”Œrecord_dependencies”NŒconfig”NŒ	id_prefix”hŒauto_id_prefix”Œid”Œdump_settings”NŒdump_internals”NŒdump_transforms”NŒdump_pseudo_xml”NŒexpose_internals”NŒstrict_visitor”NŒ_disable_config”NŒ_source”hÇŒ_destination”NŒ_config_files”]”Œ7/var/lib/git/docbuild/linux/Documentation/docutils.conf”aŒfile_insertion_enabled”ˆŒraw_enabled”KŒline_length_limit”M'Œpep_references”NŒpep_base_url”Œhttps://peps.python.org/”Œpep_file_url_template”Œpep-%04d”Œrfc_references”NŒrfc_base_url”Œ&https://datatracker.ietf.org/doc/html/”Œ	tab_width”KŒtrim_footnote_reference_space”‰Œsyntax_highlight”Œlong”Œsmart_quotes”ˆŒsmartquotes_locales”]”Œcharacter_level_inline_markup”‰Œdoctitle_xform”‰Œdocinfo_xform”KŒsectsubtitle_xform”‰Œimage_loading”Œlink”Œembed_stylesheet”‰Œcloak_email_addresses”ˆŒsection_self_link”‰Œenv”NubŒreporter”NŒindirect_targets”]”Œsubstitution_defs”}”Œsubstitution_names”}”Œrefnames”}”Œrefids”}”Œnameids”}”(jÒ  jÏ  j‘  jŽ  jÔ  jÑ  j	  j  jL  jI  j.  j+  jû  jø  j>  j;  j_  j\  jy  jv  jÊ  jÇ  uŒ	nametypes”}”(jÒ  ‰j‘  ‰jÔ  ‰j	  ‰jL  ‰j.  ‰jû  ‰j>  ‰j_  ‰jy  ‰jÊ  ‰uh}”(jÏ  hÊjŽ  jC  jÑ  j”  j  j×  jI  j  j+  jO  jø  j1  j;  jþ  j\  jA  jv  jb  jÇ  j|  uŒfootnote_refs”}”Œcitation_refs”}”Œautofootnotes”]”Œautofootnote_refs”]”Œsymbol_footnotes”]”Œsymbol_footnote_refs”]”Œ	footnotes”]”Œ	citations”]”Œautofootnote_start”KŒsymbol_footnote_start”K Œ
id_counter”Œcollections”ŒCounter”“”}”…”R”Œparse_messages”]”Œtransform_messages”]”Œtransformer”NŒinclude_log”]”Œ
decoration”Nh²hub.