The list-chunk command is deemed to be reasonably complete so make it
visible in the default build. The output can be tweaked later.
Issue: #559
Signed-off-by: David Sterba <dsterba@suse.com>
The long options now allow to pass the unit mode in the usual way, drop
the local variable for raw byte values.
Signed-off-by: David Sterba <dsterba@suse.com>
The string escaping functionality is more generic and can be used in
other commands (e.g. in dump-tree). Move it to the string utils.
Signed-off-by: David Sterba <dsterba@suse.com>
tree-stats currently displays only some global trees and fs-tree 5. Add
support to show the stats of a specified tree.
Issue: #268
Signed-off-by: Chung-Chiang Cheng <cccheng@synology.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This makes btrfs scrub status --si show the Rate in metric units as well.
Before:
Total to scrub: 877.65GB
Rate: 609.22MiB/s
After:
Total to scrub: 877.65GB
Rate: 638.81MB/s
Pull-request: #832
Author: Ivan Kozik <ivan@ludios.org>
Signed-off-by: David Sterba <dsterba@suse.com>
Enhance the sorting capabilities of 'inspect list-chunks' to allow
multiple keys. Drop the gaps, this works only for pstart and it's hard
to make it work with arbitrary sort keys.
Usage is printed by default, assuming this is an interesting info and
even if it slows down the output (due to extra lookups) it's more
convenient to print it rather than not.
The options related to usage and empty were removed.
Output changes:
- rename Number to PNumber, meaning physical number on the device
- print Devid, device number, can be also sort key
Examples:
btrfs inspect list-chunks /mnt
btrfs inspect list-chunks --sort length,usage
btrfs inspect list-chunks --sort lstart
Depending on the sort key order, the output can be wild, for that the
PNumber and LNumber give some hint where the chunks lie in their space.
Example output:
$ sudo ./btrfs inspect list-chunks --sort length,usage /
Devid PNumber Type/profile PStart Length PEnd LNumber LStart Usage%
----- ------- ----------------- --------- --------- --------- ------- --------- ------
1 7 Data/single 1.52GiB 16.00MiB 1.54GiB 69 191.68GiB 86.04
1 3 System/DUP 117.00MiB 32.00MiB 149.00MiB 40 140.17GiB 0.05
1 2 System/DUP 85.00MiB 32.00MiB 117.00MiB 39 140.17GiB 0.05
1 15 Data/single 8.04GiB 64.00MiB 8.10GiB 61 188.60GiB 94.46
1 1 Data/single 1.00MiB 84.00MiB 85.00MiB 68 191.60GiB 74.24
1 5 Metadata/DUP 341.00MiB 192.00MiB 533.00MiB 60 188.41GiB 82.58
1 4 Metadata/DUP 149.00MiB 192.00MiB 341.00MiB 59 188.41GiB 82.58
1 20 Metadata/DUP 9.29GiB 256.00MiB 9.54GiB 38 139.92GiB 57.76
1 19 Metadata/DUP 9.04GiB 256.00MiB 9.29GiB 37 139.92GiB 57.76
1 22 Metadata/DUP 9.79GiB 256.00MiB 10.04GiB 25 113.15GiB 57.93
1 21 Metadata/DUP 9.54GiB 256.00MiB 9.79GiB 24 113.15GiB 57.93
1 46 Metadata/DUP 29.29GiB 256.00MiB 29.54GiB 43 142.71GiB 62.38
Signed-off-by: David Sterba <dsterba@suse.com>
There's more information available in sysfs
(/sys/fs/btrfs/FSID/allocation) that we can print in 'fi df'. This is
still meant for debugging or deeper analysis of the filesystem, the
values need to be correctly interpreted with respect to the profiles,
persistence and other conditonal features.
The extended output is not printed by default and for now is behind the
verbosity options:
$ btrfs -vv fi df /mnt
Data, single: total=47.06GiB, used=25.32GiB
System, DUP: total=32.00MiB, used=16.00KiB
Metadata, DUP: total=1.44GiB, used=961.20MiB
GlobalReserve, single: total=125.62MiB, used=0.00B
Data:
bg_reclaim_threshold 0%
bytes_may_use 8.00KiB
bytes_pinned 0.00B
bytes_readonly 64.00KiB
bytes_reserved 0.00B
bytes_used 25.32GiB
bytes_zone_unusable 0.00B
chunk_size 10.00GiB
disk_total 47.06GiB
disk_used 25.32GiB
total_bytes 47.06GiB
Metadata:
bg_reclaim_threshold 0%
bytes_may_use 126.62MiB
bytes_pinned 0.00B
bytes_readonly 0.00B
bytes_reserved 0.00B
bytes_used 961.20MiB
bytes_zone_unusable 0.00B
chunk_size 256.00MiB
disk_total 2.88GiB
disk_used 1.88GiB
total_bytes 1.44GiB
System:
bg_reclaim_threshold 0%
bytes_may_use 0.00B
bytes_pinned 0.00B
bytes_readonly 0.00B
bytes_reserved 0.00B
bytes_used 16.00KiB
bytes_zone_unusable 0.00B
chunk_size 32.00MiB
disk_total 64.00MiB
disk_used 32.00KiB
total_bytes 32.00MiB
Signed-off-by: David Sterba <dsterba@suse.com>
Commit 4db925911c ("btrfs-progs: use strncpy_null everywhere") did
not properly convert the subvolume name copying to strncpy_null() and
trimmed the last character.
Issue: #829
Signed-off-by: David Sterba <dsterba@suse.com>
The separator of key=value is only one or more space character, the
'encoded_write' also uses ',' which is inconsistent with the rest.
Signed-off-by: David Sterba <dsterba@suse.com>
The xattr names are user strings but still can potentially contain
special characters (as reported). There doesn't seem to be a restriction
on the name defined.
The xattr values care length-encoded byte arrays so escaping needs be
done.
The clone source is a path and by mistake lacked the encoding.
Issue: #818
Signed-off-by: David Sterba <dsterba@suse.com>
Use the safe version of strncpy that makes sure the string is
terminated.
To be noted:
- the conversion in scrub path handling was skipped
- sizes of device paths in some ioctl related structures is
BTRFS_DEVICE_PATH_NAME_MAX + 1
Recently gcc 13.3 started to detect problems with our use of strncpy
potentially lacking the null terminator, warnings like:
cmds/inspect.c: In function ‘cmd_inspect_logical_resolve’:
cmds/inspect.c:294:33: warning: ‘__builtin_strncpy’ specified bound 4096 equals destination size [-Wstringop-truncation]
294 | strncpy(mount_path, mounted, PATH_MAX);
| ^
Signed-off-by: David Sterba <dsterba@suse.com>
Now that there's only __strncpy_null we can drop the underscore and move
it to string-utils as it's a generic string function rather than
something for paths.
Signed-off-by: David Sterba <dsterba@suse.com>
The macro strncpy_null uses sizeof the first argument for the length,
but there are no checks and this works only for buffers with static
length, i.e. not pointers. This is error prone. Use the open coded
variant that makes the sizeof visible.
Signed-off-by: David Sterba <dsterba@suse.com>
Use unaligned access helper for code that potentially or actually
accesses data that come from on-disk structures. This is for image or
chunk restore. This may pessimize some cases but is in general safer on
strict alignment architectures and has no effect on other architectures.
Related issue #770.
Signed-off-by: David Sterba <dsterba@suse.com>
Recent patches updated stale qgroup handling, using 'unlinked' and
'dropped' where we otherwise use 'deleted' and 'cleaned'.
Signed-off-by: David Sterba <dsterba@suse.com>
Currently `btrfs qgroup show` command shows any 0 level qgroup without a
root backref as `<stale>`, which is not correct.
There are several more cases:
- Under deletion
The subvolume is not yet full dropped, but unlinked.
In that case we would not have a root backref item, but the qgroup is
not stale.
- Squota space holder
This is for squota mode, that a fully dropped subvolume still have
extents accounting on the already-gone subvolume.
In this case it's not stale either, and future accounting relies on
it.
This patch would add above special cases, and add an extra `SPECIAL
PATHS` section to explain all the cases, including `<stale>`.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The current stale qgroup deletion doesn't handle the following cases at
all:
- It doesn't detect stale qgroups correctly
The current check is using the root backref, which means unlinked but
not yet fully dropped subvolumes would mark its corresponding qgroups
stale.
This is incorrect. The real stale check should be based on the root
item, not root backref.
- Squota non-empty but stale qgroups
Such qgroups can not and should not be deleted, as future accounting
still require them.
- Full accounting mode, stale qgroups but not empty
Since qgroup numbers are inconsistent already, it's common to have
such stale qgroups with non-zero numbers.
Now it's dependent on the kernel to determine whether such qgroup can
be deleted.
Address the above problems:
- Do root_item based detection
So that btrfs_qgroup::stale would properly indicate if there is a
subvolume root item for the qgroup.
- Do not attempt to delete squota stale but non-empty qgroups
- Attempt to delete stale but non-empty qgroups for full accounting mode
And deletion failure would not count as an error.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This allows the users to identify if the running qgroup mode and whether
the numbers are already inconsistent.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since qgroup numbers are only updated at transaction commit time, it's
better to do a sync before reading the quota tree, to reduce the chance
of uncommitted qgroup changes.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This reverts commit 9da773aa46.
There are several problems related to the --delete-qgroup option:
- Currently kernel doesn't allow to delete non-empty qgroups
- A qgroup can only be empty after fully dropped and a transaction is
committed
The tool doesn't take either factor into consideration
- Things like drop_subtree_threshold or other operations can mark qgroup
inconsistent and skip accounting
This can mean the target qgroup will never be empty until next rescan
On the other hand, even we do it the proper way, it would hugely delay
the command (wait until the subvolume to be cleaned).
Furthermore, even if the waiting is handled properly,
drop_subtree_threshold can still prevent us deleting the qgroup (qgroup
numbers are inconsistent, and accounting is skipped completely).
So the qgroup cleanup needs kernel to make it work properly.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Use the objectid, type, offset natural order as it's more readable and
we're used to read keys like that.
Signed-off-by: David Sterba <dsterba@suse.com>
What basename(3) does with the argument depends on _GNU_SOURCE and
inclusion of libgen.h. This is problematic on Musl (1.2.5) as reported.
We want the GNU semantics that does not modify the argument. Common way
to make it portable is to add own helper. This is now implemented in
path_basename() that does not use the libc provided basename but preserves
the semantics. The path_dirname() is just for parity, otherwise same as
dirname().
Sources:
- https://bugs.gentoo.org/926288
- https://git.musl-libc.org/cgit/musl/commit/?id=725e17ed6dff4d0cd22487bb64470881e86a92e7
Issue: #778
Signed-off-by: David Sterba <dsterba@suse.com>
Reported by 'gcc -fanalyzer':
cmds/inspect.c:1193:1: warning: leak of ‘ctx.stats’ [CWE-401] [-Wanalyzer-malloc-leak]
There are mixed returns and gotos for error handling and the returns
miss freeing of the ctx.stats. Unify all paths to the single label that
frees the buffers and rename it.
Signed-off-by: David Sterba <dsterba@suse.com>
Reported by 'gcc -fanalyzer':
cmds/scrub.c:1150:25: warning: use of possibly-NULL ‘path’ where non-null expected [CWE-690] [-Wanalyzer-possible-null-argument]
Initialization of the datafile path is done from a static string but the
strdup() call is not handled. Store the path directly to the buffer,
it's later modified by mkdir_p().
Signed-off-by: David Sterba <dsterba@suse.com>
Reported by 'gcc -fanalyzer':
cmds/subvolume.c:1078:39: warning: use of possibly-NULL ‘name’ where non-null expected [CWE-690] [-Wanalyzer-possible-null-argument]
The failure name duplication is not handled and can potentially lead to
a NULL dereference later. Handle the error properly and return template
error message.
Signed-off-by: David Sterba <dsterba@suse.com>
Reported by 'gcc -fanalyzer':
cmds/replace.c:357:17: warning: double ‘close’ of file descriptor ‘fdmnt’ [CWE-1341] [-Wanalyzer-fd-double-close]
The first close is done right before going to the label
'leave_with_error' but the variable is not reset to -1 so in the exit
block close() is called again.
Signed-off-by: David Sterba <dsterba@suse.com>
Use a local copy of the search header for proper aligned access instead
of the unaligned helpers, move the definitions to the closest scope.
Signed-off-by: David Sterba <dsterba@suse.com>
Use tree search ioctl wrappers for code that is considered internal, ie.
leaving out libbtrfs (legacy), libbtrfsutil (needs own API for that).
Conversion is mostly direct of what the API provides.
Signed-off-by: David Sterba <dsterba@suse.com>
Bit shifts should be done on unsigned type as a matter of good practice
to avoid any problems with bit overflowing to the sign bit.
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
With the latest kernel patch to reject invalid qgroupids in
btrfs_qgroup_inherit structure, "btrfs subvolume create" or "btrfs
subvolume snapshot" can lead to the following output:
# mkfs.btrfs -O quota -f $dev
# mount $dev $mnt
# btrfs subvolume create -i 2/0 $mnt/subv1
Create subvolume '/mnt/btrfs/subv1'
ERROR: cannot create subvolume: No such file or directory
The "btrfs subvolume" command output the first line, seemingly to
indicate a successful subvolume creation, then followed by an error
message.
This can be a little confusing on whether if the subvolume is created or
not.
[FIX]
Fix the output by only outputting the regular line if the ioctl
succeeded.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Remove btrfs_qgroup_inherit_add_copy() and the command line interface.
This was designed to add a pair of source/destination qgroups into
btrfs_qgroup_inherit structure, so that rfer/excl numbers would be
copied from the source qgroup into the destination one.
This behavior has been intentionally hidden since 2016, as such copy will
cause qgroup inconsistent immediately and a rescan would reset whatever
numbers copied anyway.
Now we're going to reject the copy behavior from kernel, there is no
need to keep those hidden (and already disabled for "subvolume create")
case.
Remove btrfs_qgroup_inherit_add_copy() call, and cleanup the
undocumented options.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Use a more descriptive name, the interface is generic so it should use
the generic term for file/directory.
Signed-off-by: David Sterba <dsterba@suse.com>
There are some cases that disable verbosity (of errors) and then print
own message. Enable the verbose error messages printed by
btrfs_open_fd2() as they are specific.
Signed-off-by: David Sterba <dsterba@suse.com>
The code in scrub predates the global verbosity options and sets its own
quiet status. This is still used only for error messages that should be
printed even with -q. Drop that or replace with bconf.verbose status
check.
Signed-off-by: David Sterba <dsterba@suse.com>
There are many places that pass false as verbosity argument and then
print an error message, or don't print any message in error cases.
Use btrfs_open_file_or_dir_fd() that will be verbose in case of an error
with the same semantics.
Signed-off-by: David Sterba <dsterba@suse.com>
It's commonly used elsewhere in the code to return the -errno values if
possible, do that for the open helpers too.
Signed-off-by: David Sterba <dsterba@suse.com>
For historical reasons the helpers [btrfs_]open_dir... return also
the 'DIR *dirstream' value when a directory is opened.
However this is never used. So avoid calling diropen() and return
only the fd.
Replace the last reference to btrfs_open_file_or_dir3() with
btrfs_open_fd2().
Signed-off-by: Goffredo Baroncelli <kreijack@libero.it>
Signed-off-by: David Sterba <dsterba@suse.com>
For historical reasons the helpers [btrfs_]open_dir... return also
the 'DIR *dirstream' value when a directory is opened.
However this is never used. So avoid calling diropen() and return
only the fd.
Replace open_file_or_dir() with btrfs_open_fd2() removing any reference
to the unused/useless dirstream variables. btrfs_open_fd2() is required
to avoid spurious error messages.
Signed-off-by: Goffredo Baroncelli <kreijack@libero.it>
Signed-off-by: David Sterba <dsterba@suse.com>
For historical reasons the helpers [btrfs_]open_dir... return also
the 'DIR *dirstream' value when a directory is opened.
However this is never used. So avoid calling diropen() and return
only the fd.
Replace btrfs_open_file_or_dir() with btrfs_open_file_or_dir_fd()
removing any references to the unused/useless dirstream variables.
Signed-off-by: Goffredo Baroncelli <kreijack@libero.it>
Signed-off-by: David Sterba <dsterba@suse.com>
For historical reasons the helpers [btrfs_]open_dir... return also
the 'DIR *dirstream' value when a directory is opened.
However this is never used. So avoid calling diropen() and return
only the fd.
Replace open_file_or_dir3() with btrfs_open_fd2() removing any reference
to the unused/useless dirstream variables. btrfs_open_fd2() is needed
because sometime the callers need to set the RDONLY/RDWRITE mode, and to
avoid spurious messages.
Signed-off-by: Goffredo Baroncelli <kreijack@libero.it>
Signed-off-by: David Sterba <dsterba@suse.com>
