The 0/subvolid qgroups are not automatically deleted when the subvolume
is deleted, for historical reasons. There's a command to clean up all
such stale qgroups (btrfs qgroup clean-stale) but this should be also
possible with the subvolume deletion.
With the options we can switch the default to delete the qgroup by
default eventually, if somebody depends on the not deleting behaviour
the negation option can be used.
Issue: #366
Signed-off-by: David Sterba <dsterba@suse.com>
Compiling with clang on aarch64 leads to an error when detecting the
SIMD instruction support. Gcc ignores the arch/feature mismatch.
Conditionally detect the -m flags only on x86_64.
Issue: #712
Signed-off-by: David Sterba <dsterba@suse.com>
The enqueue option should let the user know that the expected operation
hasn't started yet and that it's waiting for another one. Although the
exclusive operations can take long, the two reason should be
distinguished.
Signed-off-by: David Sterba <dsterba@suse.com>
Reported on IRC, that it's unexpected that passing several devices on
command line for 'btrfs device delete' still uses some of the devices
during deletion. The expectation was that they'd be removed at once (and
thus not used for the intermediate chunk relocation).
As it works now, the ioctl removes only one device. As a workaround, add
a timeout (like we have for the full balance and others) when there are
more devices passed on the command line. This can be skipped by the
--force parameter.
Issue: #708
Signed-off-by: David Sterba <dsterba@suse.com>
strtoull may return the boundary values, if the callers could expect
that and verify it then the errno must be reset before the call.
Signed-off-by: David Sterba <dsterba@suse.com>
There are several APIs that have tests, add a single build target for
convenience and enable that in the CI.
Signed-off-by: David Sterba <dsterba@suse.com>
https://www.openssl.org/ Is a well known cryptography library and since
freshly released version 3.2 it also supports variable digest size of
blake2b, so we can now add it among the crypto providers.
Configure with --with-crypto=openssl.
Signed-off-by: David Sterba <dsterba@suse.com>
https://botan.randombit.net/ Botan is a cryptography library with C
bindings and provides what we need (sha256 and blake2b), among many
others. Add it to the list of crypto backends if somebody wants to use
it.
Currently the version 2.19 is the latest one. Botan3 3.2.0 exists but
does not seem to be widely available in distros yet.
Configure with --with-crypto=botan.
Signed-off-by: David Sterba <dsterba@suse.com>
The test case would verify the following behaviors:
- Partial failure
Should return 1, but the remaining valid destinations would still be
created.
- All success
That's as usual.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This patch would make "btrfs subvolume create" to accept multiple
arguments, just like "mkdir".
The existing options like "-i <qgroupid>" and "-p" would all be applied
to all subvolume(s).
If one destination failed, the command would return 1, while still retry
the remaining destinations.
Issue: #695
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The function strdup() can return NULL if the system is out of memory,
thus we need to hanle the rare but possible -ENOMEM case.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Previous fix for char devices and properties opens the path in non
blocking mode but this still triggers the watchdog, as reported. Add a
workaround to properties to completely skip opening the path and just
stat() it.
Issue: #699
Signed-off-by: David Sterba <dsterba@suse.com>
The `btrfs subvolume list -o` command intentionally doesn't
recurse subvolumes, so make it clear in the documentation.
Pull-reques: #709
Signed-off-by: Matteo Croce <teknoraver@meta.com>
Signed-off-by: David Sterba <dsterba@suse.com>
GCC 14 introduces a new -Walloc-size included in -Wextra which gives:
```
common/utils.c:983:15: warning: allocation of insufficient size ‘1’ for type ‘struct config_param’ with size ‘32’ [-Walloc-size]
cmds/qgroup.c:1644:13: warning: allocation of insufficient size ‘1’ for type ‘struct btrfs_qgroup_inherit’ with size ‘72’ [-Walloc-size]
```
The calloc prototype is:
```
void *calloc(size_t nmemb, size_t size);
```
So, just swap the number of members and size arguments to match the prototype, as
we're initialising 1 struct of size `sizeof(struct ...)`. GCC then sees we're not
doing anything wrong.
Pull-request: #707
Signed-off-by: Sam James <sam@gentoo.org>
Signed-off-by: David Sterba <dsterba@suse.com>
With the kernel commit 070bb0011ccf ("btrfs: sysfs: show if ACL
support has been compiled in") we can now check if ACL is compiled
without requiring a btrfs device. Retain older method for older
kernels.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Without the change `BTRFS_IOC_SCAN_DEV` aliased with `BTRFS_IOC_FORGET_DEV`.
It's a regression introduced in fcd9142b6 "btrfs-progs: docs: formatting,
fixups, updates".
It manifests as a sudden device disappearance when device is scanned:
machine # [ 4.095032] Btrfs loaded, crc32c=crc32c-intel, zoned=no, fsverity=no
machine # ERROR: device scan failed on '/dev/vdb': No such file or directory
machine # ERROR: device scan failed on '/dev/vdc': No such file or directory
(finished: must succeed: mkfs.btrfs -d raid0 /dev/vdb /dev/vdc, in 10.31 seconds)
Issue: #704
Pull-request: #706
Reported-by: Atemu <atemu.main@gmail.com>
Bug: https://github.com/NixOS/nixpkgs/issues/265668
Author: Sergei Trofimovich <slyich@gmail.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The zone size may not be parsed correctly in 'nullb create' due to
copy&paste error from the 'size'. This is already fixed upstream.
Signed-off-by: David Sterba <dsterba@suse.com>
In zoned mode print zone count for each device, the zone size must be
the same so it's sufficient to print it in the summary.
$ mkfs.btrfs -O zoned /dev/nullb[0-3]
...
Zoned device: yes
Zone size: 16.00MiB
...
Devices:
ID SIZE ZONES PATH
1 512.00MiB 32 /dev/nullb0
2 256.00MiB 16 /dev/nullb1
3 1.00GiB 64 /dev/nullb2
4 2.00GiB 128 /dev/nullb3
Issue: #693
Signed-off-by: David Sterba <dsterba@suse.com>
There's a report that reading properties from a sound device the system
is stuck and then gets rebooted by watchdog. Reading from fifo files
gets stuck as well, although this would not trigger the watchdog.
The reason is that open() on fifo files is blocking until the other end
of the pipe is opened. For device nodes it's driver specific, most
device nodes fail right away:
$ btrfs prop get /dev/tty
ERROR: object is not a btrfs object: /dev/tty
In case of the sound device the consequences were fatal. We can fix that
by opening the path on non-blocking mode. This is only for reading the
fsid, the fd is closed right after the ioctl so the non-blocking mode
does not affect other operation.
The blocking mode must be used for block devices as e.g. loop devices
may not be finalized when the open() call returns and get_fsid fails.
The known problematic devices are character and fifos.
Issue: #699
Signed-off-by: David Sterba <dsterba@suse.com>
We can run more tests in parallel, the overall CI run time will be
shorter (from 12m to 5m) at the cost of building the same sources.
Also build documentation only once.
Signed-off-by: David Sterba <dsterba@suse.com>
Since a few days the CI started to fail randomly when there were loop
devices used in the tests. The mount fails because some device is
reported to be missing:
$ losetup --show --find
/dev/loop3
...
$ mkfs ...
ERROR: device scan failed on '/dev/loop3': No such file or directory
...
$ mount
mount: /home/runner/work/btrfs-progs/btrfs-progs/tests/mnt: wrong fs
type, bad option, bad superblock on /dev/loop3, missing codepage or
helper program, or other error.
$ dmesg
...
BTRFS error (device loop0): devid 3 uuid 11d9c345-9527-433e-a024-7102659fa0ee is missing
BTRFS error (device loop0): failed to read the system array: -2
BTRFS error (device loop0): open_ctree failed
This was reproducible in the "cli" tests, but also happened on a local
machine.
To fix that wait for all loop devices before mount, the command
'btrfs device ready' should block until that. The convenience helper
does that, for any standalone 'mount' used with loop devices this must
be done manually.
Signed-off-by: David Sterba <dsterba@suse.com>
The tests are run in one job, we can make it parallel by grouping
related tests together.
- musl test, using docker, does not have to wait for build
- simple tests, library, ioctl, api, ...
- the rest are functional tests
Signed-off-by: David Sterba <dsterba@suse.com>
"that are do not match -> that do not match"
[ci skip]
Pull-request: #700
Author: Martin Ligabue <martinligabue@gmail.com>
Signed-off-by: David Sterba <dsterba@suse.com>
- update Status page
- new features in 6.7
- more ioctls
- CSS fix to wrap long lines in tables
[ci skip]
Signed-off-by: David Sterba <dsterba@suse.com>
The function almost always returns 0 even for errors as the ret value is
not used in the final return. This was attempted to be fixed in
55438f3930 ("btrfs-progs: resize: return error value from
check_resize_args()") but this broke 'resize cancel' when devid 1 was
missing and was later reverted as 4286eb552e ("Revert "btrfs-progs:
resize: return error value from check_resize_args()"").
The devid fallback has been fixed so the proper return value can be
returned now.
Issue: #539
Signed-off-by: David Sterba <dsterba@suse.com>
The implicit devid is 1 but when it does not exist then the command
'btrfs fi resize max /path' fails and requires the user to specify the
number (and finding it elsewhere, e.g. in 'btrfs fi us -T' output).
This is a usability bug, we can verify if devid 1 exists and use the
lowest devid as a fallback. This does what user would expected, though
there's still a warning. Kernel has the hardcoded devid 1 when none is
specified, with this fix in user space the kernel does not need to be
changed (or could behave the same eventually).
Example use:
$ btrfs fi us -T .
Data Metadata System
Id Path single single single Unallocated Total Slack
-- ---------- --------- --------- -------- ----------- ------- -----
4 /dev/loop3 - - - 4.00GiB 4.00GiB -
-- ---------- --------- --------- -------- ----------- ------- -----
Total 416.00MiB 256.00MiB 64.00MiB 4.00GiB 4.00GiB 0.00B
Used 0.00B 128.00KiB 16.00KiB
$ btrfs fi resize max .
WARNING: no devid specified means devid 1 which does not exist, using
lowest devid 4 as a fallback
Resize device id 4 (/dev/loop3) from 4.00GiB to max
Issue: #470
Signed-off-by: David Sterba <dsterba@suse.com>
Add a convenience option to processing the range in smaller steps than
the whole file, where a flush is done after each steps. This could be
potentially used to measure progress with 'btrfs -vv fi defrag'.
Issue: #616
Signed-off-by: David Sterba <dsterba@suse.com>
The test names 011 and 012 are the same but 012 is actually testing
recursion of the 'btrfs fi du' command.
Signed-off-by: David Sterba <dsterba@suse.com>
The option "--clear-space-cache" is not really that suitable for "btrfs
check" group, as there are some concerns:
- Allowing transid mismatch
- No leaf item checks
Thoe behaviour are inherited from the default open ctree flags for
"btrfs check", which can be unsafe if the end user just wants to clear
the cache.
- Unclear if the cache clearing would happen along with repair
Thankfully the clearing of space cache is done without any repair
Thus there is a proposal to move space cache removal to rescue group,
and this patch would do that exactly.
However this would lead to some behavior changes:
- Transid mismatch would be treated as error
- Leaf items size/offset would still be checked
If we hit any above error, we should just abort without doing any
write.
These change would increase the safety of the space cache removal, thus
I believe it's worthy to introduce such behavior change.
Since we're here, also add a small explanation on why we need this
dedicated tool to clear space cache (especially for v1 cache).
Issue: #698
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
