Whamcloud - gitweb
tools/e2fsprogs.git
12 years agoe2fsck: allow deleting or zeroing shared blocks
Andreas Dilger [Thu, 12 Apr 2012 23:32:53 +0000 (17:32 -0600)]
e2fsck: allow deleting or zeroing shared blocks

E2fsck fixes files that are found to be sharing blocks by cloning
the shared blocks and giving each file a private copy in pass 1D.

Allowing all files claiming the shared blocks to have copies can
inadvertantly bypass access restrictions.  Deleting all the files,
zeroing the cloned blocks, or placing the files in the /lost+found
directory after cloning may be preferable in some secure environments.

The following patches implement config file and command line options
in e2fsck that allow pass 1D behavior to be tuned according to site
policy.  It adds two extended options and config file counterparts.
On the command line:

 -E clone=dup|zero

    Select the block cloning method.  "dup" is old behavior,
    and is the default.  "zero" is a new method that substitutes
    zero-filled blocks for the shared blocks in all the files
    that claim them.

 -E shared=preserve|lost+found|delete

    Select the disposition of files containing shared blocks.
    "preserve" is the old behavior which remains the default.
    "lost+found" causes files to be unlinked after cloning so
    they will be reconnected to /lost+found in pass 3.
    "delete" skips cloning entirely and simply deletes the files.

In the config file:
  [options]
      clone=dup|zero
      shared=preserve|lost+found|delete

Signed-off-by: Jim Garlick <garlick@llnl.gov>
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agoe2fsck: parse config file before command-line opts
Andreas Dilger [Thu, 12 Apr 2012 23:24:55 +0000 (17:24 -0600)]
e2fsck: parse config file before command-line opts

The patch changes the order that the config file and command line
are parsed so that command line has precedence.  It also parses
the -E option for every occurrence, otherwise the -E option is
not cumulative.

Signed-off-by: Jim Garlick <garlick@llnl.gov>
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agotests: invalid value of in-inode EA offset
Andreas Dilger [Thu, 12 Apr 2012 22:57:40 +0000 (16:57 -0600)]
tests: invalid value of in-inode EA offset

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agotests: extent pointing to non-existent block
Andreas Dilger [Fri, 25 May 2012 07:01:28 +0000 (01:01 -0600)]
tests: extent pointing to non-existent block

Signed-off-by: Girish Shilamkar <girish@clusterfs.com>
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agoe2fsck: workaround for old extents tests
Andreas Dilger [Thu, 12 Apr 2012 22:15:07 +0000 (16:15 -0600)]
e2fsck: workaround for old extents tests

The e2fsck_ext2fs_extent_get() part of this patch is a workaround
to handle problems with old Lustre extents patches that didn't
clear the ee_start_hi or ei_leaf_hi fields.

That has been fixed for long time and could be removed as soon
as the f_extent_* tests are fixed to clear these _hi fields.
Otherwise the extents are all marked as corrupt and it ruins those
tests value.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agotests: verify > 65000 subdirectories
Andreas Dilger [Thu, 12 Apr 2012 22:02:12 +0000 (16:02 -0600)]
tests: verify > 65000 subdirectories

Add test case to verify nlink handling of large directories.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agoTT-177 build: add .spec file for SLES11 packaging
Andreas Dilger [Fri, 13 Apr 2012 08:23:12 +0000 (02:23 -0600)]
TT-177 build: add .spec file for SLES11 packaging

Include the upstream SLES11 .spec file to ensure the packages we
build match the upstream packages.  Any later patches that change
the packaging should patch the .spec file appropriately.

Add in the SLES-specific patches, excluding the replacement de.po
file, since the original SLES11 de.po file is only against 1.41.4,
and is missing a large number of changes to the translated messages
related to 64-bit format specifiers.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I783d58bd78d7c4c66cc85ec5557ae1aaf64016ba

12 years agobuild: add RHEL6 .spec file for packaging
Andreas Dilger [Fri, 13 Apr 2012 08:19:19 +0000 (02:19 -0600)]
build: add RHEL6 .spec file for packaging

Include the upstream RHEL6 .spec file to ensure the packages we
build match the upstream packages.  Any later patches that change
the packaging should patch the .spec file appropriately.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agobuild: update e2fsprogs.spec for distro builds
Andreas Dilger [Thu, 12 Apr 2012 21:39:04 +0000 (15:39 -0600)]
build: update e2fsprogs.spec for distro builds

Add the distro version to the RPM release number, so that it the
RPM names do not conflict.

Allow the RPM built from upstream to replace the split packages
provided by the distros.  At some point in the future it may be
desirable to also split the RPM built by this spec file, but this
is complicated by the fact that SLES and RHEL have different splits.

Signed-off-by: Girish Shilamkar <girish.shilamkar@sun.com>
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agofilefrag: Lustre changes to filefrag FIEMAP handling
Andreas Dilger [Thu, 12 Apr 2012 21:31:35 +0000 (15:31 -0600)]
filefrag: Lustre changes to filefrag FIEMAP handling

Add support for multiple-device filesystems by defining a new
fe_device field in the fiemap_extent structure.  This allows
printing the filesystem-relative or linux block device number
associated with each extent of a file.  If a single filesystem
extent is mirrored to multiple block devices, the fe_device
field can be used to disambiguate the multiple copies.

If the "-l" (device-logical) option is given to filefrag, then
all extents for a particular device of a file are returned
before returning extents for the next device.  This makes it
easier to see if extent allocation within a single device is
contiguous, instead of returning all of the blocks of a file
interleaved in file-logical-offset order.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agoblkid: fix ZFS device detection
Andreas Dilger [Thu, 12 Apr 2012 21:26:49 +0000 (15:26 -0600)]
blkid: fix ZFS device detection

Fix the ZFS device detection by looking at multiple uberblocks to
see if any are present, rather than looking for the ZFS boot block
which is not always present.

There may be up to 128 uberblocks, but the first 4 are not written
to disk on a newly-formatted filesystem so check several of them at
different offsets within the uberblock array.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agoe2fsck: improve in-inode xattr checks
Andreas Dilger [Thu, 12 Apr 2012 21:23:47 +0000 (15:23 -0600)]
e2fsck: improve in-inode xattr checks

Add check for in-inode xattr to make sure that it is not referencing
an offset that is beyond the end of the inode.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agobuild: fix build warnings for MacOS
Andreas Dilger [Fri, 27 Apr 2012 05:22:45 +0000 (23:22 -0600)]
build: fix build warnings for MacOS

The BLKFLSBUF and FDFLUSH ioctls are Linux specific, and do not
really have anything to do with __GNUC__ (which is also used on
OS/X and Solaris).  Only print these warnings on Linux systems.

O_DIRECT is not defined in the OS/X headers.  Since this is only a
performance enhancement, and not required for correct operation,
just ignore it if undefined.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agotests: make generated test scripts read-only
Andreas Dilger [Wed, 23 May 2012 21:05:21 +0000 (15:05 -0600)]
tests: make generated test scripts read-only

Make generated test scripts read-only, to avoid errors by developers
editing the generated test scripts and then having them accidentally
clobbered when "make" is run again.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agobuild: update version for Lustre build
Andreas Dilger [Thu, 12 Apr 2012 20:00:07 +0000 (14:00 -0600)]
build: update version for Lustre build

Add Whamcloud-specific build version to distinguish packages from
upstream packages.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
12 years agoblkid: remove no-op statement which caused a clang warning
Theodore Ts'o [Wed, 13 Mar 2013 18:34:08 +0000 (14:34 -0400)]
blkid: remove no-op statement which caused a clang warning

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: fix command failures for extent_inode commands that take arguments
Theodore Ts'o [Wed, 13 Mar 2013 18:19:16 +0000 (14:19 -0400)]
debugfs: fix command failures for extent_inode commands that take arguments

The extent_inode commands split_node, replace_node, and insert_node
take arguments which resulted in confusing error messages after
succeeding.  Fix this.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoe2fsck: fix build failure with --enable-jbd-debug
Theodore Ts'o [Wed, 13 Mar 2013 18:02:53 +0000 (14:02 -0400)]
e2fsck: fix build failure with --enable-jbd-debug

Commit e3507739e4185 introduced a build failure if e2fsprogs is
configured with --enable-jbd-debug.  Fix this.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agolibext2fs: Provide prototype for ext2fs_symlink()
Jan Kara [Mon, 25 Feb 2013 05:55:07 +0000 (05:55 +0000)]
libext2fs: Provide prototype for ext2fs_symlink()

New function ext2fs_symlink() doesn't have a prototype in ext2fs.h and
thus debugfs compilation gives warning:

debugfs.c:2219:2: warning: implicit declaration of function 'ext2fs_symlink'

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agoe2p: Fix 's' handling in parse_num_blocks2()
Jan Kara [Mon, 25 Feb 2013 05:55:05 +0000 (05:55 +0000)]
e2p: Fix 's' handling in parse_num_blocks2()

parse_num_blocks2() wrongly did:
num << 1;
when log_block_size < 0. That is obviously wrong as such statement has
no effect (and the compiler properly warns about it). Callers expect
returned value to be in bytes when log_block_size < 0 so fix the
statement accordingly.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agochattr: allow clearing the extent flag
Theodore Ts'o [Sat, 23 Feb 2013 03:23:05 +0000 (22:23 -0500)]
chattr: allow clearing the extent flag

In order to support kernels which support conversion of extent-mapped
files to direct/indirect mapped files, remove the sanity check which
prevented clearing the extent flag in chattr.  Kernels which don't
support this will simply give an Operation Not Supported error.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: add sanity check to make sure we never shift 64 bits right
Theodore Ts'o [Mon, 28 Jan 2013 14:02:23 +0000 (09:02 -0500)]
debugfs: add sanity check to make sure we never shift 64 bits right

In the tables which are used to parse the fields for the set_fields
command, there should never be a entry which has a size set to 8
bytes, and two pointers defined.  Not only would it result in
undefined behavior in the compiled code, it doesn't make any sense and
is definitely a bug.

Reported-by: Xi Wang <xi.wang@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agopo: update de.po (from translationproject.org)
Philipp Thomas [Mon, 28 Jan 2013 03:44:56 +0000 (22:44 -0500)]
po: update de.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agoFix warnings about functions not returning a value
Philipp Thomas [Mon, 28 Jan 2013 03:41:44 +0000 (22:41 -0500)]
Fix warnings about functions not returning a value

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agobuild: quiet build warnings for "gcc -Wall"
Andreas Dilger [Mon, 28 Jan 2013 03:29:01 +0000 (22:29 -0500)]
build: quiet build warnings for "gcc -Wall"

Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agocontrib: add missing '-p' to fallocate's usage message
Zheng Liu [Fri, 25 Jan 2013 03:44:19 +0000 (11:44 +0800)]
contrib: add missing '-p' to fallocate's usage message

Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agolibext2fs: fix ext2fs_llseek on i386
Phillip Susi [Thu, 24 Jan 2013 16:21:56 +0000 (11:21 -0500)]
libext2fs: fix ext2fs_llseek on i386

ext2fs_llseek() was using lseek instead of lseek64.  The
only time it would use lseek64 is if passed an offset that
overflowed 32 bits.  This works for SEEK_SET, but not
SEEK_CUR, which can apply a small offset to move the file
pointer past the 32 bit limit.

The code has been changed to instead try lseek64 first, and
fall back to lseek if that fails.  It also was doing a
runtime check of the size of off_t.  This has been moved to
compile time.

This fixes a problem which would cause e2image when built for
x86-32 to bomb out when used with large file systems.

Signed-off-by: Phillip Susi <psusi@ubuntu.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoUpdate Release Notes, Changelogs, version.h, for final 1.42.7 release v1.42.7
Theodore Ts'o [Tue, 22 Jan 2013 03:03:18 +0000 (22:03 -0500)]
Update Release Notes, Changelogs, version.h, for final 1.42.7 release

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agomke2fs, tune2fs, resize2fs: add warning messages for bigalloc and quota
Theodore Ts'o [Tue, 22 Jan 2013 00:07:38 +0000 (19:07 -0500)]
mke2fs, tune2fs, resize2fs: add warning messages for bigalloc and quota

The bigalloc and quota features have some known issues, so issue
warnings in case users try to use them.

More information can be found here:
https://ext4.wiki.kernel.org/index.php/Bigalloc
https://ext4.wiki.kernel.org/index.php/Quota

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agocontrib: fix namespace leakage in spd_readdir
Theodore Ts'o [Mon, 21 Jan 2013 22:19:50 +0000 (17:19 -0500)]
contrib: fix namespace leakage in spd_readdir

Declare the internal symbols alloc_dirstruct() and cache_dirstruct()
as static so they don't leak out into the global namespace.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agocontrib: add thread locking and readdir64_r support to spd_readdir
Theodore Ts'o [Mon, 21 Jan 2013 22:15:25 +0000 (17:15 -0500)]
contrib: add thread locking and readdir64_r support to spd_readdir

This is part of a series of improvements from a 2008 version of
spd_readdir.c that somehow didn't make it into the version which we
checked into e2fsprogs git tree.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agocontrib: add safe_getenv() support to spd_readdir
Theodore Ts'o [Mon, 21 Jan 2013 21:35:16 +0000 (16:35 -0500)]
contrib: add safe_getenv() support to spd_readdir

This is part of a series of improvements from a 2008 version of
spd_readdir.c that somehow didn't make it into the version which we
checked into e2fsprogs git tree.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: move a cluster at a time with bigalloc file systems
Theodore Ts'o [Sun, 20 Jan 2013 05:25:00 +0000 (00:25 -0500)]
resize2fs: move a cluster at a time with bigalloc file systems

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: correctly account for clusters when calculating summary stats
Theodore Ts'o [Sun, 20 Jan 2013 04:41:36 +0000 (23:41 -0500)]
resize2fs: correctly account for clusters when calculating summary stats

Fixes resize2fs so it correctly calculates the number of free clusters
in each block group for file systems with the bigalloc feature
enabled.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agolibext2fs: teach the ext2fs_*_block_bitmap_range2() about clusters
Theodore Ts'o [Sun, 20 Jan 2013 04:21:11 +0000 (23:21 -0500)]
libext2fs: teach the ext2fs_*_block_bitmap_range2() about clusters

The ext2fs_{mark,unmark,test}_block_bitmap2() functions understand
about clusters, and will take block numbers and convert them to
clusters before checking the bitmap.  The
ext2fs_*_block_bitmap_range2() functions did not do this, which made
them inconsistent.  Fortunately, nothing has depended on this
incorrect behavior, and in fact most of the usage of these functions
have only recently been added, and only for optimizations that were
only enabled for non-bigalloc file systems.

So this is a change in previously exported functions, but (a) it
doesn't change the behavior at all for non-bigalloc file systems, and
(b) the change is more likely to fix bugs for bigalloc file systems.
For example, this change fixes a problem with resize2fs and bigalloc
file systems.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoUpdate Release Notes, Changelogs, version.h, etc. for 1.42.7 release
Theodore Ts'o [Wed, 16 Jan 2013 18:50:12 +0000 (13:50 -0500)]
Update Release Notes, Changelogs, version.h, etc. for 1.42.7 release

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: check in test-resize script
Theodore Ts'o [Wed, 16 Jan 2013 18:43:36 +0000 (13:43 -0500)]
resize2fs: check in test-resize script

This should be made into a more formal, automated test case, but for
now, save this as script since it's useful for validating resize2fs's
handling of very large file systems.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoUpdate config.{guess,sub} to the latest version
Theodore Ts'o [Wed, 16 Jan 2013 18:10:54 +0000 (13:10 -0500)]
Update config.{guess,sub} to the latest version

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agopo: update vi.po (from translationproject.org)
Trần Ngọc Quân [Wed, 16 Jan 2013 04:30:36 +0000 (23:30 -0500)]
po: update vi.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agopo: update sv.po (from translationproject.org)
Göran Uddeborg [Wed, 16 Jan 2013 04:30:36 +0000 (23:30 -0500)]
po: update sv.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agopo: update pl.po (from translationproject.org)
Jakub Bogusz [Wed, 16 Jan 2013 04:30:36 +0000 (23:30 -0500)]
po: update pl.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agopo: update nl.po (from translationproject.org)
Benno Schulenberg [Wed, 16 Jan 2013 04:30:36 +0000 (23:30 -0500)]
po: update nl.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agopo: update fr.po (from translationproject.org)
Samuel Thibault [Wed, 16 Jan 2013 04:30:35 +0000 (23:30 -0500)]
po: update fr.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agopo: update de.po (from translationproject.org)
Philipp Thomas [Wed, 16 Jan 2013 04:30:35 +0000 (23:30 -0500)]
po: update de.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agopo: update cs.po (from translationproject.org)
Petr Pisar [Wed, 16 Jan 2013 04:30:35 +0000 (23:30 -0500)]
po: update cs.po (from translationproject.org)

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agoUpdate misc/Makefile.in using "make depend"
Theodore Ts'o [Wed, 16 Jan 2013 04:28:17 +0000 (23:28 -0500)]
Update misc/Makefile.in using "make depend"

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoe2image: add -a switch to include all data
Phillip Susi [Tue, 15 Jan 2013 20:31:23 +0000 (15:31 -0500)]
e2image: add -a switch to include all data

Normally the raw and QCOW2 images only contain fs metadata.
Add a new switch ( -a ) to include all data.  This makes it
possible to use e2image to clone a whole filesystem.

Signed-off-by: Phillip Susi <psusi@ubuntu.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agotests: create test for debugfs creating special files
Theodore Ts'o [Tue, 15 Jan 2013 19:52:30 +0000 (14:52 -0500)]
tests: create test for debugfs creating special files

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: fix mknod command so that it updates the block group statistics
Theodore Ts'o [Tue, 15 Jan 2013 19:50:02 +0000 (14:50 -0500)]
debugfs: fix mknod command so that it updates the block group statistics

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: add symlink command
Darren Hart [Fri, 4 Jan 2013 20:00:59 +0000 (12:00 -0800)]
debugfs: add symlink command

Add support for symbolic links using a new symlink command.  Modeled
after the do_mkdir() command.

Testing demonstrates both fastlinks and slowlinks work correctly.
Very long target paths fail as the command parsing appears to truncate
the input to somewhere around 256 bytes.

Signed-off-by: Darren Hart <dvhart@infradead.org>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: Andreas Dilger <adilger@dilger.ca>
12 years agolibext2fs: add the ext2fs_symlink() function
Darren Hart [Fri, 4 Jan 2013 20:00:58 +0000 (12:00 -0800)]
libext2fs: add the ext2fs_symlink() function

Creating symlinks is a complex affair when accounting for slowlinks.

Create a new function, ext2fs_symlink(), modeled after ext2fs_mkdir().
Like ext2fs_mkdir(), ext2fs_symlink() takes on the task of allocating a
new inode and block (for slowlinks), setting up sane default values in
the inode, copying the target path to either the inode (for fastlinks)
or to the first block (for slowlinks), and accounting for the inode and
block stats.  Disallow link targets longer than blocksize as the Linux
kernel prevents this.

It does not attempt to expand the parent directory, instead returning
EXT2_ET_DIR_NO_SPACE and leaving it to the caller to expand just as
ext2fs_mkdir() does.  Ideally, I think both of these functions should
make a single attempt to expand the directory.

[ Fixed a few bugs discovered when creating a test case for ext2fs_symlink() ]

Signed-off-by: Darren Hart <dvhart@infradead.org>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: Andreas Dilger <adilger@dilger.ca>
12 years agolibext2fs: add error codes from 1.43.x development branch
Theodore Ts'o [Wed, 16 Jan 2013 19:07:25 +0000 (14:07 -0500)]
libext2fs: add error codes from 1.43.x development branch

To maintain the error codes numbering, we need to pull in the changes
from the 1.43.x development branch for the libext2's error table.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agomke2fs: document bigalloc and cluster-size
Zheng Liu [Sun, 13 Jan 2013 09:08:15 +0000 (17:08 +0800)]
mke2fs: document bigalloc and cluster-size

Bigalloc feature has been used for a long time, but the documentation
in mke2fs is still missing.  So add it.

Addresses-Debian-Bug: #669730

Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agolibext2fs: avoid 32-bit overflow in ext2fs_initialize with a 512M cluster size
Theodore Ts'o [Tue, 15 Jan 2013 00:29:54 +0000 (19:29 -0500)]
libext2fs: avoid 32-bit overflow in ext2fs_initialize with a 512M cluster size

If the user attemps to create a 512MB cluster, we need to adjust the
defaults to avoid a 32-bit overflow of s_blocks_per_group.  Also check
to make sure that the caller of ext2fs_initialize() has not given a
value of s_clusters_per_group that would result in an overflow of
s_blocks_per_group.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
12 years agomke2fs: teach mke2fs to understand -b 4k and -C 256M
Theodore Ts'o [Tue, 15 Jan 2013 00:03:11 +0000 (19:03 -0500)]
mke2fs: teach mke2fs to understand -b 4k and -C 256M

The -b and -C options now use parse_num_blocks2() instead of strtol,
so that users can specify -C 256M instead of the much less convenient
-C 268435456.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
12 years agolibe2p: teach parse_num_blocks2() to return bytes if log_block_size < 0
Theodore Ts'o [Tue, 15 Jan 2013 00:01:25 +0000 (19:01 -0500)]
libe2p: teach parse_num_blocks2() to return bytes if log_block_size < 0

Previously the behavior of parse_num_block2 was undefined if
log_block_size was less than zero.  It will now return a number in
units of bytes.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
12 years agomke2fs: the -g option will now specify the clusters per block group
Theodore Ts'o [Mon, 14 Jan 2013 22:30:23 +0000 (17:30 -0500)]
mke2fs: the -g option will now specify the clusters per block group

If bigalloc is enabled, then -g will specify the clusters per block
group.  (If bigalloc is not enabled, then a cluster == a block, so the
meaning of -g is not changed.)

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
12 years agomke2fs: enforce that the cluster size must be greater that the block size
Theodore Ts'o [Mon, 14 Jan 2013 22:28:00 +0000 (17:28 -0500)]
mke2fs: enforce that the cluster size must be greater that the block size

In addition, do not allow a cluster size of 1024, since that will be
interpreted by ext2fs_initialize() as requesting the default cluster
size.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
Reported-by: Zheng Liu <wenqing.lz@taobao.com>
12 years agomke2fs: require the bigalloc feature explicity if the cluster size is set
Zheng Liu [Sun, 13 Jan 2013 09:08:13 +0000 (17:08 +0800)]
mke2fs: require the bigalloc feature explicity if the cluster size is set

When cluster-size is specified without the bigalloc feature, mke2fs
will ignore this argument silently.  But user might think bigalloc
feature has been enabled unless they use the dumpe2fs command.  So now
we ask user to set bigalloc feature explicity when cluster-size is
enabled.  This can make sure that users understand what they are doing
because bigalloc might impact the performance for some workloads.

Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: fix gcc-wall complaints
Theodore Ts'o [Mon, 14 Jan 2013 19:41:46 +0000 (14:41 -0500)]
debugfs: fix gcc-wall complaints

Fix the missing function prototypes from the recently added new
debugfs commands, plus some signed vs unsigned comparison complaints.

Also change the abbreviation of the block_dump command from "bp" to
the more appropriate "bp".

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: add debugging code to test the old online resizing interfaces
Theodore Ts'o [Sat, 12 Jan 2013 22:33:44 +0000 (17:33 -0500)]
resize2fs: add debugging code to test the old online resizing interfaces

The old online resize ioctl interfaces are still present in the
kernel, and we want to make it easy to test both the kernel code for
those older interfaces, and resize2fs's use of those interfaces in a
relatively easy manner.

To do this, resize2fs will now check the environment variable
RESIZE2FS_KERNEL_VERSION.  If the version given is less than 3.3, then
do not try using the new resizing ioctl, but instead use the resizing
ioctls that were used before Linux version 3.3.

If the version given is less than 3.7, then emulate sanity checks
which get done to protect against the fact that the new resizing ioctl
prior to 3.7 did not handle meta_bg resizing.  (This was previously
tested via the presence of the RESIZE2FS_NO_META_BG_RESIZE environment
variable.  But the new environment variable, RESIZE2FS_KERNEL_VERISON,
is more general.)

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
12 years agodebugfs: document zap_block & block_dump
Eric Sandeen [Fri, 11 Jan 2013 07:40:25 +0000 (07:40 +0000)]
debugfs: document zap_block & block_dump

What little docs there were had a cut & paste error.
We can do better. :)

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agoconfig: silence printf format warnings
Eric Whitney [Fri, 4 Jan 2013 10:27:12 +0000 (10:27 +0000)]
config: silence printf format warnings

The printfs in the asm_types.c code contained within parse-types.sh
expect sizeof to return an int.  Fix this for architectures where this
isn't true (x86_64, etc.) so we don't see warning messages while
running the configure script.

Signed-off-by: Eric Whitney <enwlinux@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agolibext2fs: ext2fs_open2() should not set ret_fs after a MMP failure
Theodore Ts'o [Wed, 9 Jan 2013 01:47:11 +0000 (20:47 -0500)]
libext2fs: ext2fs_open2() should not set ret_fs after a MMP failure

The addition of MMP code was added in the wrong place, so ret_fs could
get set (and EXT2_FLAG_NOFREE_ON_ERROR was cleared as well, which
could confuse e2fsck which depends on this flag being cleared if
ext2fs_open2() succeeded.)

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoe2fsck: avoid memory corruption on ext2fs_open2 failure
Nickolai Zeldovich [Tue, 8 Jan 2013 20:45:31 +0000 (15:45 -0500)]
e2fsck: avoid memory corruption on ext2fs_open2 failure

In try_open_fs(), if ext2fs_open2() returns an error, do not try to
access the struct ext2_filesys.  The previous check 'if (ret_fs)' was
always true, but even 'if (*ret_fs)' might be incorrect in some cases,
so check 'retval==0' instead.

Signed-off-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoe2fsck: do not crash on long log file names
Nickolai Zeldovich [Tue, 8 Jan 2013 20:31:18 +0000 (15:31 -0500)]
e2fsck: do not crash on long log file names

Previously e2fsck would corrupt memory if the log file name was longer
than 100 bytes (e.g., a long log_filename value in e2fsck.conf or a
pattern that expands out to more than 100 bytes).  This was due to
incorrectly calling realloc() in append_string() on the struct string
instead of the malloc'ed char* buffer, among other problems.  This
patch fixes the call to realloc() and also ensures that the buffer is
grown by sufficiently many bytes (not just by 2x).

Signed-off-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoe2fsck: show size requested when memory allocation fails
Eric Sandeen [Mon, 7 Jan 2013 21:45:54 +0000 (15:45 -0600)]
e2fsck: show size requested when memory allocation fails

"e2fsck: Can't allocate dx_block info array"
is only so helpful - it'd be nice to know how much it tried to allocate.

In particular, since I think malloc(0) can return NULL,
it'd be nice to know if maybe we passed in an uninitialized (or
0-initialized) size.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agomke2fs: check extents feature when bigalloc feature is enabled
Zheng Liu [Sun, 6 Jan 2013 12:25:17 +0000 (20:25 +0800)]
mke2fs: check extents feature when bigalloc feature is enabled

When bigalloc feature is enabled in mkfs, extents feature also needs
to be enabled.  But now when bigalloc feature is enabled without
extents feature, users will not get any warning messages until they
try to mount this file system.

Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoFix 32-bit overflow problems: dgrp_t * s_blocks_per_group
Theodore Ts'o [Thu, 3 Jan 2013 18:42:38 +0000 (13:42 -0500)]
Fix 32-bit overflow problems: dgrp_t * s_blocks_per_group

There are a number of places where we multiply a dgrp_t with
s_blocks_per_group expecting that we will get a blk64_t.  This
requires a cast, or using the convenience function
ext2fs_group_first_block2().

This audit was suggested by Eric Sandeen.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
12 years agoresize2fs: use [un]mark_block_range bitmap functions to reduce CPU usage
Theodore Ts'o [Thu, 3 Jan 2013 13:35:25 +0000 (08:35 -0500)]
resize2fs: use [un]mark_block_range bitmap functions to reduce CPU usage

Use ext2fs_[un]mark_block_range2() functions to reduce the CPU
overhead of resizing large file systems by 45%, primarily by
reducing the time spent in fix_uninit_block_bitmaps().

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: add resource tracking as a debug option
Theodore Ts'o [Thu, 3 Jan 2013 03:43:36 +0000 (22:43 -0500)]
resize2fs: add resource tracking as a debug option

Add a new debug flag which prints how much time is consumed by the
various parts of resize2fs's processing.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: fix 32-bit overflow when calculating the number of free blocks
Theodore Ts'o [Thu, 3 Jan 2013 14:01:22 +0000 (09:01 -0500)]
resize2fs: fix 32-bit overflow when calculating the number of free blocks

This caused the free blocks count in the superblock to be incorrect
after resizing a 64-bit file system if the number of free blocks
overflowed a 32-bit value.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: fix 32-bit overflow issue which can corrupt 64-bit file systems
Theodore Ts'o [Thu, 3 Jan 2013 04:31:04 +0000 (23:31 -0500)]
resize2fs: fix 32-bit overflow issue which can corrupt 64-bit file systems

Fix a 32-bit overflow bug caused by a missing blk64_t cast which can
cause the block bitmap to get corrupted when doing an off-line resize
of a 64-bit file system.

This problem can be reproduced as follows:

rm -f foo.img; touch foo.img
truncate -s 8T foo.img
mke2fs -F -t ext4 -O 64bit foo.img
e2fsck -f foo.img
truncate -s 21T foo.img
resize2fs foo.img
e2fsck -fy foo.img

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoClean up texinfo files
Theodore Ts'o [Wed, 2 Jan 2013 15:06:09 +0000 (10:06 -0500)]
Clean up texinfo files

Fix up the com_err.texinfo file so it will produce a valid printed
output, by cleaning up some errors in the texinfo file, and updating
texinfo.tex to be consistent with the version in the doc subdirectory.

Also add rules so we can generate pdf and ps files from
com_err.texinfo and libext2fs.texinfo.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoUpdate changelogs for 1.42.7 release
Theodore Ts'o [Tue, 1 Jan 2013 22:17:14 +0000 (17:17 -0500)]
Update changelogs for 1.42.7 release

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: fixup the hard-coded buffer length in dump_file
Zheng Liu [Tue, 1 Jan 2013 12:30:14 +0000 (20:30 +0800)]
debugfs: fixup the hard-coded buffer length in dump_file

Allocate the block buffer in dump_file() instead of assuming that the
block size is no more than 8k.

CC: George Spelvin <linux@horizon.com>
Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebain: update e2fslibs.symbols with newly exported functions
Theodore Ts'o [Tue, 1 Jan 2013 19:57:06 +0000 (14:57 -0500)]
debain: update e2fslibs.symbols with newly exported functions

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoFix gcc -Wall nits
Theodore Ts'o [Tue, 1 Jan 2013 18:28:27 +0000 (13:28 -0500)]
Fix gcc -Wall nits

This fixes the last set of gcc -Wall complaints.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agofilefrag: improvements to filefrag FIEMAP handling
Andreas Dilger [Thu, 22 Nov 2012 15:06:12 +0000 (15:06 +0000)]
filefrag: improvements to filefrag FIEMAP handling

Update the filefrag program to allow displaying the extents in
some different formats.  Try and stay within 80 columns.
* add -k option to print extents in kB-sized units (like df -k)
* add -b {blocksize} to print extents in blocksize units
* add -e option to print extent format, even when FIBMAP is used
* add -X option to print extents in hexadecimal format

Internally, the FIBMAP handling code has been moved into its own
function like FIEMAP, so that the code is more modular.  Extent
offsets are now handled in bytes instead of in blocks, to allow
printing extents with arbitrary block sizes.  The extent header
printing also moved into its own function so that it can be shared
between the FIEMAP and FIBMAP handling routines, since it got more
complex with the different output options.

Only print error about FIBMAP being root-only a single time.
Print the filesystem type if it changes between specified files.
Add fsync() for FIBMAP if "-s" is given.

Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agolibquota: quiet log_err() bad format warnings
Andreas Dilger [Thu, 29 Nov 2012 12:47:54 +0000 (05:47 -0700)]
libquota: quiet log_err() bad format warnings

The macro for log_err() was written so that it needed to always
have an argument, but GCC was unhappy to have an argument when
none was specified in the format string.  Use the CPP "##" to
eat the preceeding comma if no argument is specified.

Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agobuild: quiet some "gcc -Wall" compiler warnings
Andreas Dilger [Thu, 29 Nov 2012 12:47:53 +0000 (05:47 -0700)]
build: quiet some "gcc -Wall" compiler warnings

Quiet a number of simple compiler warnings:
- pointers not initialized by ext2fs_get_mem()
- return without value in non-void function
- dereferencing type-punned pointers
- unused variables

Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agomisc: cleanup unused variables on MacOS
Andreas Dilger [Thu, 29 Nov 2012 12:47:52 +0000 (05:47 -0700)]
misc: cleanup unused variables on MacOS

Clean up unused variables found by GCC on MacOS.

Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agotests: don't consider "make check" a compiler error
Andreas Dilger [Thu, 29 Nov 2012 12:47:51 +0000 (05:47 -0700)]
tests: don't consider "make check" a compiler error

In a number of places, the output format from "make check" is
incorrectly interpreted as compiler warning output (triggered by
the presence of colons and parenthesis in the output).  Convert
these lines to similar output that does not trigger false build
warnings.

In the case of the tst_uuid.c program, the "ctime()" output was
difficult to change, but in fact it is better to actually compare
the time-based UUID against wallclock time instead of just printing
the formatted time as a string, so this test is improved.

Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
12 years agoresize2fs: create optimized flex_bg block groups
Theodore Ts'o [Sat, 29 Dec 2012 08:43:55 +0000 (03:43 -0500)]
resize2fs: create optimized flex_bg block groups

Now that we are reserving all of the bg-specific metadata before we
try to allocate the metadata for the new block groups, we don't have
to temporarily disable the flex_bg feature flag while we allocate the
new metadata blocks --- this allows the newly created block groups to
have a much more optimized layout, instead of fragmenting the inode
table and block/inode bitmaps in sepraate block groups.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: allow resizing flex_bg && !resize_inode file systems
Theodore Ts'o [Sat, 29 Dec 2012 08:38:34 +0000 (03:38 -0500)]
resize2fs: allow resizing flex_bg && !resize_inode file systems

With the bug fixes from the last two commits, resize2fs can now fully
support off-line resizing of file systems with flex_bg even if the
resize_inode feature is not present; so we no longer need to disallow
this combination.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: handle bg descriptors which overlap with other bg's metadata
Theodore Ts'o [Sat, 29 Dec 2012 08:25:10 +0000 (03:25 -0500)]
resize2fs: handle bg descriptors which overlap with other bg's metadata

With flex_bg file systems, bg-specific metadata (i.e., bitmaps and the
inode table blocks) can be located in another block group.  Hence,
when we grow the number of block group descriptors, we need to check
if we need to relocate metadata blocks not just for the block group
where the bgd blocks are located, but in all block groups.

This change fixes the following test case:

rm -f foo.img; touch foo.img
truncate -s 32G foo.img
mke2fs -F -t ext4 -E resize=12582912 foo.img
e2fsck -f foo.img
truncate -s 256G foo.img
./resize2fs foo.img
e2fsck -fy foo.img

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: reserve all metadata blocks for flex_bg file systems
Theodore Ts'o [Sat, 29 Dec 2012 05:53:16 +0000 (00:53 -0500)]
resize2fs: reserve all metadata blocks for flex_bg file systems

For flex_bg file systems, if we need to relocate an allocation bitmap
or inode table, we need to make sure that all metadata blocks have
been reserved, lest we end up overwriting a metadata block belonging
to a different block group.

This change fixes the following test case:

rm -f foo.img; touch foo.img
truncate -s 32G foo.img
mke2fs -F -t ext4 -E resize=12582912 foo.img
e2fsck -f foo.img
truncate -s 64G foo.img
./resize2fs foo.img
e2fsck -fy foo.img

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agoresize2fs: reserve fs metadata blocks first in blocks_to_move()
Theodore Ts'o [Sat, 29 Dec 2012 00:54:19 +0000 (19:54 -0500)]
resize2fs: reserve fs metadata blocks first in blocks_to_move()

This is the first commit to add support for off-line resizing using
flex_bg without the assist of using the resize_inode to reserve gdt
blocks.  This functionality has been broken up into separate commits
which are hopefully obviously correct to make them easier to review
for correctness.

In this first step, we break up the for loop at the end of
blocks_to_move() so that we first mark all of the metadata blocks
which don't need to be moved in the reserve_blocks bitmap, and then
try to allocate the metadata blocks are new or which need to moved
second.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agomke2fs: fix crash when parsing "-E resize=NNN" with "-O 64bit"
Theodore Ts'o [Fri, 28 Dec 2012 23:26:12 +0000 (18:26 -0500)]
mke2fs: fix crash when parsing "-E resize=NNN" with "-O 64bit"

If the 64-bit file system feature is enabled, then mke2fs would crash
due to a divide-by-zero error caused by s_desc_size not being
initialized yet.

Reported-by: George Spelvin <linux@horizon.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebian: use more specific library file names in *.files
Filipe Brandenburger [Sat, 15 Dec 2012 04:56:55 +0000 (20:56 -0800)]
debian: use more specific library file names in *.files

This patch specifies libraries using a more specific glob that will pick
only the lib*.so.<version> file and will not match the lib*.so symlink
and the lib*.a archive/static library.

Signed-off-by: Filipe Brandenburger <filbranden@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebian: look for quota.pc and libquota.a from root of install tree
Filipe Brandenburger [Sat, 15 Dec 2012 04:54:13 +0000 (20:54 -0800)]
debian: look for quota.pc and libquota.a from root of install tree

The rules makefile was already using `find' in order to cope with
multi-arch directories under /usr/lib. This patch changes it to look for
those files from the root of the install tree. This allows for
installing to libdirs of /usr/lib64 or /lib or /lib64. There are no
other files with the same names in the package so it's not a problem to
find from the root of the tree.

Signed-off-by: Filipe Brandenburger <filbranden@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebian: add make variable to pass extra parameters to configure
Filipe Brandenburger [Sat, 15 Dec 2012 04:50:28 +0000 (20:50 -0800)]
debian: add make variable to pass extra parameters to configure

Setting EXTRA_CONF_FLAGS in rules.custom will pass the extra arguments
to calls of ./configure when building e2fsprogs. This can be used, for
instance, to pass a --libdir argument or similar to the configure
script.

Signed-off-by: Filipe Brandenburger <filbranden@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebian: add make variable to prevent building e2fsck.static package
Filipe Brandenburger [Sat, 15 Dec 2012 04:47:03 +0000 (20:47 -0800)]
debian: add make variable to prevent building e2fsck.static package

Setting BUILD_E2FSCK_STATIC=no in rules.custom will prevent the
debian/rules makefile from building a statically-linked e2fsck and
from creating a deb package for it.

Signed-off-by: Filipe Brandenburger <filbranden@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: add the commands "zap_block" and "block_dump"
Theodore Ts'o [Mon, 24 Dec 2012 05:22:10 +0000 (00:22 -0500)]
debugfs: add the commands "zap_block" and "block_dump"

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agolibext2fs: fix spelling typo in texinfo docs
Darren Hart [Sun, 23 Dec 2012 03:47:24 +0000 (22:47 -0500)]
libext2fs: fix spelling typo in texinfo docs

Signed-off-by: Darren Hart <dvhart@infradead.org>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: Andreas Dilger <adilger@dilger.ca>
12 years agoconfigure.in: require at least autoconf 2.54
Theodore Ts'o [Sun, 23 Dec 2012 03:24:45 +0000 (22:24 -0500)]
configure.in: require at least autoconf 2.54

AC_PROG_EGREP requires autoconf 2.54, so bump up the AC_PREREQ
accordingly.

Reported-by: g.esp@free.fr
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agolibext2fs: link the test programs with static libraries
Theodore Ts'o [Sun, 23 Dec 2012 03:15:22 +0000 (22:15 -0500)]
libext2fs: link the test programs with static libraries

Force the use of the static libraries when linking the test program so
that "make check" works when the shared libraries have not been
installed, and so that we test against the version of the libraries in
the source tree.

Reported-by: g.esp@free.fr
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: add a makefile rule to build debugfs.static
Theodore Ts'o [Sun, 23 Dec 2012 02:24:38 +0000 (21:24 -0500)]
debugfs: add a makefile rule to build debugfs.static

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agodebugfs: add the ability to manipulate the extent tree directly
Theodore Ts'o [Sat, 22 Dec 2012 23:50:33 +0000 (18:50 -0500)]
debugfs: add the ability to manipulate the extent tree directly

This commit adds the functionality which had previously only been in
the tst_extents command to debugfs.  The debugfs command extent_open
will open extent tree of a particular inode, and enables a series of
commands which will allow the user to interact with the extent tree
directly.  Once the extent tree is closed via extent_open(), these
additional commands will be disabled again.

This commit exports two new functions from lib/ext2fs/extent.c which
had previously been statically defined: ext2fs_extent_node_split() and
ext2fs_extent_goto2().

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
12 years agotests: add test of an incorrect interior node in an extent tree
Theodore Ts'o [Fri, 21 Dec 2012 02:48:08 +0000 (21:48 -0500)]
tests: add test of an incorrect interior node in an extent tree

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>