This patch turns on the BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF superblock flag
when creating a new file system in mkfs, enabling extended inode refs.
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
Add a function, process_inode_extref() to be called from process_one_leaf()
when an item type of BTRFS_INODE_EXTREF_KEY is encountered.
Similarly to process_inode_ref(), process_inode_extref() walks an extref and
adds an inode_backref structure for each reference found within.
I modified fsck's inode_backref to get a type field (ref_type) which helps
us internally track the exact type of backrefs found. Of course this field
could be overwritten in case of disk corruption (duplicate refs) but
duplicate refs themselves are tracked by btrfsck so that should not be an
issue as btrfsck is written today.
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
This patch syncs the extended inode ref definitions from kernels ctree.h and
adds support in btrfs-debug-tree for visualizing the state of extended refs
on disk.
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
In cmd_send_start(), there is a check to make sure dump_fd is not a tty
before parsing command options. So if we use the option "-f file",
it doesn't work for the dump_fd has not been created. So fix it.
Signed-off-by: Cheng Yang <chenyang.fnst@cn.fujitsu.com>
Parallel build may fail due to late creation of version.h, fix the rule name
that does not match the filename.
Signed-off-by: David Sterba <dsterba@suse.cz>
Reviewed-by: Eri Sandeen <sandeen@redhat.com>
Commit 605e806166 broke the
mkfs.btrfs -r option, because it calls make_btrfs
without ever setting dev_block_count, in the -r case,
so we tell it to make a filesystem of size 0.
Then we wander into ENOSPC land and segfault.
As a quick one-line-fix, just set the dev_block_count
to the size of the destination image file.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
This is the user mode part of the device replace patch series.
The command group "btrfs replace" is added with three commands:
- btrfs replace start srcdev|srcdevid targetdev [-Bfr] mount_point
- btrfs replace status mount_point [-1]
- btrfs replace cancel mount_point
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
"btrfs device stats" is used to retrieve and print the device stats.
"btrfs device stats -z" is used to atomically retrieve, reset and
print the stats.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
Two convenient utility functions that have so far been local to scrub are
moved to utils.c.
They will be used in the device stats code in a following commit.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
The definition of the function open_file_or_dir() is moved from common.c
to utils.c in order to be able to share some common code between scrub
and the device stats in the following step. That common code uses
open_file_or_dir(). Since open_file_or_dir() makes use of the function
dirfd(3), the required XOPEN version was raised from 6 to 7.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Original-Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
gcc optimizes out the memcpy calls at -O2 and -Os.
Replacing memcpy with memmove does't work - gcc treats memmove
the same way it treats memcpy.
This patch brings in {get|put}_unaligned_le{16|32|64} (using the
packed struct method), and uses them in the failing get/set calls.
On architectures where unaligned accesses are cheap, these unaligned
macros should be optimized out by the compiler.
Signed-off-by: Ben Peddell <klightspeed@killerwolves.net>
A quieter build makes warnings more obvious.
This could probably be improved, but just to see
if people like this or if they hate it. :)
make V=1 overrides it and gets you the full
glory again.
[CC] ctree.o
[CC] disk-io.o
[CC] radix-tree.o
[CC] extent-tree.o
...
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
This patch kills a check in mkfs's label stuff which doesn't allow labels that
have /'s in them. This causes problems for Anaconda which try to label volumes
with their mountpoints. Thanks,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Bug:
-------------
btrfs subvolume list / -a
ID 258 gen 4226 top level 384 path media/smbshare
::
btrfs subvolume list /home -a
ID 258 gen 4226 top level 5 path <FS_TREE>/__active/media/smbshare4.snap
In the first command's output, this path is printed correctly, however
in the second output it has "4.snap" appended, similar to the names of
the snapshots I made 22 hours ago.
------------
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reported-by: Brendan Hide <brendan@swiftspirit.co.za>
We use find_good_parent() to look for a suit snapshot in the clone source
snapshots as the parent, not the source subvolume of the snapshot which
is about to be sent. fix it
Reviewed-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
Signed-off-by: Cheng Yang <chenyang.fnst@cn.fujitsu.com>
Just a small program to print the fields of a super block.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: Goffredo Baroncelli <kreijack@inwind.it>
This command will be used by things like dracut that wish to know very
simply if all of the devices have been added to the kernel cache yet for the
device to be fully mounted. This keeps initrd's from constantly having to
try to mount the file system until it succeeds every time a device is added
to the system. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
Patch rebased because of changes in mkfs.c but otherwise the same
as created by Josef Bacik
SSD's do not gain anything by having metadata DUP turned on. The underlying
file system that is a part of all SSD's could easily map duplicate metadat
blocks into the same erase block which effectively eliminates the benefit of
duplicating the metadata on disk. So detect if we are formatting a single
SSD drive and if we are do not use DUP. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
Rawhide is getting cranky with posix compliance, and a few
things have stopped building.
getpagesize() is now only available -with- __USE_XOPEN_EXTENDED
or __USE_BSD, and NOT __USE_XOPEN2K.
_GNU_SOURCE must define __USE_XOPEN2K because getpagesize()
has gone away for mkfs. I gave up and used sysconf.
Also, something used to pull in stat that no longer does, so
things like S_ISREG weren't getting defined.
The following fixes things for me.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
man pages for btrfs-progs are compressed by gzip by default. In Makefile the
variable GZIP is use, this evaluates to 'gzip gzip' on my system. From man
gzip:
> The environment variable GZIP can hold a set of default options for gzip.
> These options are interpreted first and can be overwritten by explicit
> command line parameters.
So using any other variable name fixes this. Patch is attached.
Signed-off-by: Christian Hesse <list@eworm.de>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
We missed to add the default subvolume, because it has no ROOT_BACKREF_ITEM.
This made get_parent always fail for direct decendants of the default
subvolume, resulting in lots of full streams where incremental streams were
requested.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
Reviewed-by: Alexander Block <ablock84@googlemail.com>
Unfortunately, the command line options for btrfs send were misnamed. The
-i option should not be used to give "clone sources", we'll be using -c
instead.
Compatibily note: -i option was broken anyway, which makes it less critical
renaming it. For potential users of the old option style, we emit a fatal
warning if the -i option is used.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
Likely not strictly needed but I noticed valgrind complaining about
uninitialised memory in the ioctl call.
Signed-off-by: Arvin Schnell <aschnell@suse.de>
Comparing qgroupid is not good way to check the relationship of two groups,
the right way is to compare the real level numbers.
Signed-off-by: Wang Shilong <wangsl-fnst@cn.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
1. parse_qgroupid() is implemented twice, clean up the reduplicate code.
2. atoi() can not detect errors, so use strtoull() instead of it.
Signed-off-by: Wang Shilong <wangsl-fnst@cn.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
We can use this command in two ways.
1. btrfs qgroup limit size qgroupid path
2. btrfs qgroup limit size path
Before applying this patch, we differentiate them by check the parsing result
of the second argument. It is not so good because it may make some mistakes,
For example:
btrfs qgroup limit 1M 123456
^ It is a subvolume name.
In fact, we can differentiate them just by the number of arguments, so fix it
by this way.
Signed-off-by: Wang Shilong <wangsl-fnst@cn.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
Fix the usage string to specify the correct '-f' option for input file,
not '-i'.
Signed-off-by: Ulrik Sverdrup <ulrik.sverdrup@gmail.com>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
The LOOP_GET_STATUS ioctl truncates filenames to 64 characters. We should get
the backing file for a given loop device from /sys/. This is how losetup does it
as well.
Signed-off-by: Nirbheek Chauhan <nirbheek.chauhan@collabora.co.uk>
Signed-off-by: Gene Czarcinski <gene@czarc.net>
Tested-By: Hector Oron <hector.oron@collabora.co.uk>
Scrub can be invoked to scrub only a single device of a (mounted) filesystem.
The code determines whether the given path is a mountpoint of a filesystem
by issueing a btrfs-specific ioctl to it. Only in case of EINVAL it assumed
it may be a device, all other errnos just caused it fail, but some devices
(correctly) return ENOTTY. This patch adds this to the error check.
Signed-off-by: Arne Jansen <sensille@gmx.net>
Ignore the error ENXIO (device don't exists) and ENOMEDIUM (
No medium found -> like a cd tray empty) in the function
btrfs_scan_one_dir.
This avoids spurios errors due to an empty CD or a block device node
without a device (which is frequent in a static /dev).
Signed-off-by: Goffredo Baroncelli <kreijack@inwind.it>
Let mkfs accept '-a' option and not complain. When a partition has non-zero
value in the fs_passno filed in /etc/fstab, the fsck is run but fails and boot
stops. As fsck does not break things currently, it's safe to ignore the option
and let the boot proceed.
Reference: https://bugzilla.novell.com/show_bug.cgi?id=655906
Signed-off-by: David Sterba <dsterba@suse.cz>
There, 'char' is unsigned, so once assigned '-1' from getopt, it gets
the value 255. Then, it compared to '-1' gives false.
Signed-off-by: Lluis Batlle i Rossell <viric@viric.name>
The mount status is checked against the arg <device> (argv[optind]),
not the <directory>. Correct the info printed on <device> mounted.
Signed-off-by: Wang Sheng-Hui <shhuiw@gmail.com>
This uses uintptr_t to cast pointers to u64 ioctl arguments to silence
some 32bit build warnings:
cmds-inspect.c: In function ‘__ino_to_path_fd’:
cmds-inspect.c:47:15: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
cmds-inspect.c: In function ‘cmd_logical_resolve’:
cmds-inspect.c:171:15: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
Signed-off-by: Zach Brown <zab@redhat.com>
Add new suffixes in parse_size() function. New suffixes are: T as
terabyte, P as petabyte, E as exabyte. Note these units are
multiply of 2 .
Signed-off-by: Goffredo Baroncelli <kreijack@inwind.it>
Replace the function atoll with strtoull(); Check that the suffix for the
parse_size() input is of only one character.
Signed-off-by: Goffredo Baroncelli <kreijack@inwind.it>
