This patch adds default module loading and uses it to load the default
block elevator.  During boot, it's called right after initramfs or
initrd is made available and right before control is passed to
userland.  This ensures that as long as the modules are available in
the usual places in initramfs, initrd or the root filesystem, the
default modules are loaded as soon as possible.

This will replace the on-demand elevator module loading from elevator
init path.

v2: Fixed build breakage when !CONFIG_BLOCK.  Reported by kbuild test
    robot.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Alex Riesen <raa.lkml@gmail.com>
Cc: Fengguang We <fengguang.wu@intel.com>

The init/mount.o source files produce a number of sparse warnings of the
type:

warning: incorrect type in argument 1 (different address spaces)
   expected char [noderef] <asn:1>*dev_name
   got char *name

This is due to the syscalls expecting some of the arguments to be user
pointers but they are being passed as kernel pointers.  This is harmless
but adds a lot of noise to a sparse build.

To limit the noise just disable the sparse checking in the relevant source
files, but still display a warning so that the user knows this has been
done.

Since the sparse checking has been disabled we can also remove the __user
__force casts that are scattered thru the source.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

When calling syscall service routines in kernel, some of arguments should
be user pointers but were missing __user markup on string literals.  Add
it.  Removes some sparse warnings.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Cc: Phillip Lougher <phillip@lougher.demon.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The size of a built-in initramfs is calculated in init/initramfs.c by
"__initramfs_end - __initramfs_start".  Those symbols are defined in the
linker script include/asm-generic/vmlinux.lds.h:

#define INIT_RAM_FS                                                     \
        . = ALIGN(PAGE_SIZE);                                           \
        VMLINUX_SYMBOL(__initramfs_start) = .;                          \
        *(.init.ramfs)                                                  \
        VMLINUX_SYMBOL(__initramfs_end) = .;

If the initramfs file has an odd number of bytes, the "__initramfs_end"
symbol points to an odd address, for example, the symbols in the
System.map might look like:

    0000000000572000 T __initramfs_start
    00000000005bcd05 T __initramfs_end	  <-- odd address

At least on s390 this causes a problem:

Certain s390 instructions, especially instructions for loading addresses
(larl) or branch addresses must be on even addresses.  The compiler loads
the symbol addresses with the "larl" instruction.  This instruction sets
the last bit to 0 and, therefore, for odd size files, the calculated size
is one byte less than it should be:

    0000000000540a9c <populate_rootfs>:
      540a9c:     eb cf f0 78 00 24       stmg    %r12,%r15,120(%r15),
      540aa2:     c0 10 00 01 8a af       larl    %r1,572000 <__initramfs_start>
      540aa8:     c0 c0 00 03 e1 2e       larl    %r12,5bcd04 <initramfs_end>
                                                  (Instead of  5bcd05)
      ...
      540abe:     1b c1                   sr      %r12,%r1

To fix the problem, this patch introduces the global variable
__initramfs_size, which is calculated in the "usr/initramfs_data.S" file.
The populate_rootfs() function can then use the start marker of the
.init.ramfs section and the value of __initramfs_size for loading the
initramfs.  Because the start marker and size is sufficient, the
__initramfs_end symbol is no longer needed and is removed.
Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NHendrik Brueckner <brueckner@linux.vnet.ibm.com>
Reviewed-by: NWANG Cong <xiyou.wangcong@gmail.com>
Acked-by: NMichal Marek <mmarek@suse.cz>
Acked-by: N"H. Peter Anvin" <hpa@zytor.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

The unpack routine fails to handle the decompress_method() returning
unrecognised decompressor (compress_name == NULL).  This results in the
routine looping eventually oopsing on an out of bounds memory access.

Note this bug is usually hidden, only triggering on trailing junk after
one or more correct compressed blocks.  The case of the compressed archive
being complete junk is (by accident?) caught by the if (state != Reset)
check because state is initialised to Start, but not updated due to the
decompressor not having been called.  Obviously if the junk is trailing a
correctly decompressed buffer, state == Reset from the previous call to
the decompressor.
Signed-off-by: NPhillip Lougher <phillip@lougher.demon.co.uk>
Reported-by: NAaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The symbol 'count' is a local global variable in this file.  The function
clean_rootfs() should use a different symbol name to prevent "symbol
shadows an earlier one" noise.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The decompressors return error by calling a supplied error function, and/or
by returning an error return value.  The initramfs code, however, fails to
check the exit code returned by the decompressor, and only checks the error
status set by calling the error function.

This patch adds a return code check and calls the error function.
Signed-off-by: NPhillip Lougher <phillip@lougher.demon.co.uk>
LKML-Reference: <4b26b1ef.0+ZWxT6886olqcSc%phillip@lougher.demon.co.uk>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

With the removal of duplicate unpack_to_rootfs() (commit
df52092f) the messages displayed do not
actually correspond to what the kernel is doing.  In addition, depending
if ramdisks are supported or not, the messages are not at all the same.

So keep the messages more in sync with what is really doing the kernel,
and only display a second message in case of failure.  This also ensure
that the printk message cannot be split by other printk's.
Signed-off-by: NEric Piel <eric.piel@tremplin-utc.net>
Acked-by: NH. Peter Anvin <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Change cb6ff208 ("NOMMU: Support XIP on
initramfs") seems to have broken booting from initramfs with /sbin/init
being a hardlink.

It seems like the logic required for XIP on nommu, i.e.  ftruncate to
reported cpio header file size (body_len) is broken for hardlinks, which
have a reported size of 0, and the truncate thus nukes the contents of the
file (in my case busybox), making boot impossible and ending with runaway
loop modprobe binfmt-0000 - and of course 0000 is not a valid binary
format.

My fix is to only call ftruncate if size is non-zero which fixes things
for me, but I'm not certain whether this will break XIP for those files on
nommu systems, although I would guess not.
Signed-off-by: NRandy Robertson <rmrobert@vmware.com>
Acked-by: NDavid Howells <dhowells@redhat.com>
Acked-by: NPaul Mundt <lethal@linux-sh.org>
Acked-by: NH. Peter Anvin <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

init/initramfs.c:520: warning: 'clean_rootfs' defined but not used
Signed-off-by: NNikanth Karthikesan <knikanth@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

initramfs uses printk without a linefeed, then does some work, then uses
printk to finish the message off.  However if some other code does a
printk in between, then the messages get mixed together.  Better for each
message to be an independent line...

Example of problem that this fixes:

    checking if image is initramfs...<7>Switched to high resolution mode on CPU 1
    Switched to high resolution mode on CPU 0
    it is
Signed-off-by: NSimon Kitching <skitching@apache.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

we check if initrd is initramfs first and then do the real unpack. The check
isn't required, we can directly do unpack.  If the initrd isn't an
initramfs, we can remove the garbage.  In my laptop, this saves 0.1s boot
time.

This patch penalizes non-initramfs initrd case, but nowadays, initramfs is
the most widely used method for initrds.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Acked-by: NArjan van de Ven <arjan@infradead.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: More consistent behaviour, avoid policy in the kernel

Upgrade/downgrade initrd/initramfs decompression failure from
inconsistently a panic or a KERN_ALERT message to a KERN_EMERG event.
It is, however, possible do design a system which can recover from
this (using the kernel builtin code and/or the internal initramfs),
which means this is policy, not a technical necessity.

A good way to handle this would be to have a panic-level=X option, to
force a panic on a printk above a certain level.  That is a separate
patch, however.
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Instead of failing to identify a compressed image with a decompressor
that we don't have compiled in, identify it and fail with a
comprehensible panic message.
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Impact: build fix

flush_buffer() is used unconditionally:

  init/initramfs.c:456: error: 'flush_buffer' undeclared (first use in this function)
  init/initramfs.c:456: error: (Each undeclared identifier is reported only once
  init/initramfs.c:456: error: for each function it appears in.)

So remove the decompressor #ifdefs from around it.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Centralize the compression format detection to a common routine in the
lib directory, and use it for both initramfs and initrd.
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Support XIP on files unpacked from the initramfs image on NOMMU systems.  This
simply requires the length of the file to be preset so that the ramfs fs can
attempt to garner sufficient contiguous storage to store the file (NOMMU mmap
can only map contiguous RAM).

All the other bits to do XIP on initramfs files are present:

 (1) ramfs's truncate attempts to allocate a contiguous run of pages when a
     file is truncated upwards from nothing.

 (2) ramfs sets BDI on its files to indicate direct mapping is possible, and
     that its files can be mapped for read, write and exec.

 (3) NOMMU mmap() will use the above bits to determine that it can do XIP.
     Possibly this needs better controls, because it will _always_ try and do
     XIP.

One disadvantage of this very simplistic approach is that sufficient space
will be allocated to store the whole file, and not just the bit that would be
XIP'd.  To deal with this, though, the initramfs unpacker would have to be
able to parse the file contents.
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Acked-by: NPaul Mundt <lethal@linux-sh.org>

Impact: Resolves build failures in some configurations

Makes it possible to disable CONFIG_RD_GZIP . In that case, the
built-in initramfs will be compressed by whatever compressor is
available (bzip2 or lzma) or left uncompressed if none is available.

It also removes a couple of warnings which occur when no ramdisk
compression at all is chosen.

It also restores the select ZLIB_INFLATE in drivers/block/Kconfig
which somehow came missing. This is needed to activate compilation of
the stuff in zlib_deflate.
Signed-off-by: NAlain Knaff <alain@knaff.lu>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Impact: New code for initramfs decompression, new features

This is the second part of the bzip2/lzma patch

The bzip patch is based on an idea by Christian Ludwig, includes support for
compressing the kernel with bzip2 or lzma rather than gzip. Both
compressors give smaller sizes than gzip.  Lzma's decompresses faster
than bzip2.

It also supports ramdisks and initramfs' compressed using these two
compressors.

The functionality has been successfully used for a couple of years by
the udpcast project

This version applies to "tip" kernel 2.6.28

This part contains:
- support for new compressions (bzip2 and lzma) in initramfs and
old-style ramdisk
- config dialog for kernel compression (but new kernel compressions
not yet supported)
Signed-off-by: NAlain Knaff <alain@knaff.lu>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

When unpacking the cpio into the initramfs, mtimes are not preserved by
default.  This patch adds an INITRAMFS_PRESERVE_MTIME option that allows
mtimes stored in the cpio image to be used when constructing the
initramfs.

For embedded applications that run exclusively out of the initramfs, this
is invaluable:

When building embedded application initramfs images, its nice to know when
the files were actually created during the build process - that makes it
easier to see what files were modified when so we can compare the files
that are being used on the image with the files used during the build
process.  This might help (for example) to determine if the target system
has all the updated files you expect to see w/o having to check MD5s etc.

In our environment, the whole system runs off the initramfs partition, and
seeing the modified times of the shared libraries (for example) helps us
find bugs that may have been introduced by the build system incorrectly
propogating outdated shared libraries into the image.

Similarly, many of the initializion/configuration files in /etc might be
dynamically built by the build system, and knowing when they were modified
helps us sanity check whether the target system has the "latest" files
etc.

Finally, we might use last modified times to determine whether a hot fix
should be applied or not to the running ramfs.
Signed-off-by: NNye Liu <nyet@nyet.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Inflate requires some dynamic memory allocation very early in the boot
process and this is provided with a set of four functions:
malloc/free/gzip_mark/gzip_release.

The old inflate code used a mark/release strategy rather than implement
free.  This new version instead keeps a count on the number of outstanding
allocations and when it hits zero, it resets the malloc arena.

This allows removing all the mark and release implementations and unifying
all the malloc/free implementations.

The architecture-dependent code must define two addresses:
 - free_mem_ptr, the address of the beginning of the area in which
   allocations should be made
 - free_mem_end_ptr, the address of the end of the area in which
   allocations should be made. If set to 0, then no check is made on
   the number of allocations, it just grows as much as needed

The architecture-dependent code can also provide an arch_decomp_wdog()
function call.  This function will be called several times during the
decompression process, and allow to notify the watchdog that the system is
still running.  If an architecture provides such a call, then it must
define ARCH_HAS_DECOMP_WDOG so that the generic inflate code calls
arch_decomp_wdog().

Work initially done by Matt Mackall, updated to a recent version of the
kernel and improved by me.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: NThomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Mikael Starvik <mikael.starvik@axis.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Acked-by: NPaul Mundt <lethal@linux-sh.org>
Acked-by: NYoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Instead of using the malloc() and free() wrappers needed by the
lib/inflate.c code for allocations, simply use kmalloc() and kfree() in the
initramfs code.  This is needed for a further lib/inflate.c-related cleanup
patch that will remove the malloc() and free() functions.

Take that opportunity to remove the useless kmalloc() return value
cast.

Based on work done by Matt Mackall.
Signed-off-by: NThomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: NMatt Mackall <mpm@selenic.com>
Cc: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This essentially reverts commit 71fc47a9
("ACPI: basic initramfs DSDT override support"), because the code simply
isn't ready.

It did ugly things to the init sequence to populate the rootfs image
early, but that just ended up showing other problems with the whole
approach.  The fact is, the VFS layer simply isn't initialized this
early, and the relevant ACPI code should either run much later, or this
shouldn't be done at all.

For 2.6.25, we'll just pick the latter option.  We can revisit this
concept later if necessary.

Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Tilman Schmidt <tilman@imap.cc>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Renninger <trenn@suse.de>
Cc: Eric Piel <eric.piel@tremplin-utc.net>
Cc: Len Brown <len.brown@intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Markus Gaugusch <dsdt@gaugusch.at>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The basics of DSDT from initramfs. In case this option is selected,
populate_rootfs() is called a bit earlier to have the initramfs content
available during ACPI initialization.

This is a very similar path to the one available at
http://gaugusch.at/kernel.shtml but with some update in the
documentation, default set to No and the change of populate_rootfs() the
"Jeff Mahony way" (which avoids reading the initramfs twice).
Signed-off-by: NThomas Renninger <trenn@suse.de>
Signed-off-by: NEric Piel <eric.piel@tremplin-utc.net>
Signed-off-by: NLen Brown <len.brown@intel.com>

Given that init/Makefile includes initramfs.c in the build only if
CONFIG_BLK_DEV_INITRD is defined, there seems to be no point checking for
it yet again.
Signed-off-by: NRobert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add retain_initrd option to control freeing of initrd memory after
extraction.  By default, free memory as previously.

The first boot will need to hold a copy of the in memory fs for the second
boot.  This image can be large (much larger than the kernel), hence we can
save time when the memory loader is slow.  Also, it reduces the memory
footprint while extracting the first boot since you don't need another copy
of the fs.
Signed-off-by: NMichael Neuling <mikey@neuling.org>
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We should not initialize rootfs before all the core initializers have
run.  So do it as a separate stage just before starting the regular
driver initializers.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

It turns out that the "-c" option of cpio is highly unportable even between
distros let alone unix variants, and may actually make the wrong type of
cpio archive.  I just wasted quite some time on this, and the kernel can
detect this and warn about it (it's __init memory so it gets thrown away
and thus there is no runtime overhead)
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch ensures that initramfs overwrites work correctly, even when dealing
with device nodes of different types.  Furthermore, when replacing a file
which already exists, we must make very certain that we truncate the existing
file.
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
Cc: Michael Neuling <mikey@neuling.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Copy the filenames of hardlinks when inserting them into the hash, since
the "name" pointer may point to scratch space (name_buf).  Not doing so
results in corruption if the scratch space is later overwritten: the wrong
file may be hardlinked, or, if the scratch space contains garbage, the link
will fail and a 0-byte file will be created instead.
Signed-off-by: NMark Huang <mlhuang@cs.princeton.edu>
Acked-by: NAl Viro <viro@zeniv.linux.org.uk>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I noticed that after boot with an initrd in 2.6.16 the rootfs had:

--w-r-xr-T    1 root     root      6241141 Jan  1  1970 initrd.image

Which is caused by a small typo:
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initramfs initrd images do not need a ramdisk device, so remove this
restriction in Kconfig.  BLK_DEV_RAM=n saves about 13k on i386.  Also
without ramdisk device there's no need for "dry run", so initramfs unpacks
much faster.

People using cramfs, squashfs, or gzipped ext2/minix initrd images are
probably smart enough not to turn off ramdisk support by accident.

Cc: Al Viro <viro@ftp.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

It is possible that the reserved crashkernel region can be overlapped with
initrd since the bootloader sets the initrd location.  When the initrd
region is freed, the second kernel memory will not be contiguous.  The
Kexec_load can cause an oops since there is no contiguous memory to write
the second kernel or this memory could be used in the first kernel itself
and may not be part of the dump.  For example, on powerpc, the initrd is
located at 36MB and the crashkernel starts at 32MB.  The kexec_load caused
panic since writing into non-allocated memory (after 36MB).  We could see
the similar issue even on other archs.

One possibility is to move the initrd outside of crashkernel region.  But,
the initrd region will be freed anyway before the system is up.  This patch
fixes this issue and frees only regions that are not part of crashkernel
memory in case overlaps.
Signed-off-by: NHaren Myneni <haren@us.ibm.com>
Acked-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Besides freeing initrd memory, also clear out the now dangling pointers to
it, to make sure accidental late use attempts can be detected.
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!