Age | Commit message (Collapse) | Author |
|
We don't want users hitting this accidentally, and if needed for some
strange reason (it shouldn't be), it can still be specified via -o.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
C header filename in `c_src/cmd_fusemount.c` hasn't been updated after renaming
`libbcachefs/fs-common.h`. Updating the filename to fix build failure.
Signed-off-by: Integral <integral@archlinuxcn.org>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
|
|
option to bch2_parse_mount_opts()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Migrate now works: there is an inconsequential free space inconsistency
after marking the new superblock (in migrate_superblock), which we're
hacking around with a fsck.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
bucket size now takes into account system memory and amount of memory
required to fsck: on very large filesystems it will automatically scale
up bucket size to ensure we can fsck.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
More work to regularize device options, with common command line
parsing - like other options.
This fixes an assortment of little nits: buggyness with the discard
option, device state (and others) can now be specied at device add time,
and is a decent cleanup.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
-a is used when called from system to mean "automatic invocation, run
check only if you think you need it".
Since we have self healing and online fsck, we never need run fsck
automatically at boot time.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
New tool for scanning for backup superblocks, to recover when the
primary superblock + sb layout have been overwritten.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Fix reading accounting from old kernel versions that had it in little
endian.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Implement 'bcachefs data scrub', frontend for
BCH_IOCTL_DATA.BCH_DATA_OP_scrub.
Takes a path to a device, mountpoint, or filesystem uuid. Can be run on
a specific device by passing a device, or if run on a filesystem scrubs
all devices in parallel.
Metadata only scrubbing is supported via -m.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
filesystems
|
|
filesystems
Currently, 'bcachefs set-fs-option' requires all devices to be specified
when setting options on an offline multidevice filesystem - it can't yet
scan for the rest of the filesystem members.
We do need to fix this, but the scan code is written in Rust, so that
either means a lot of ugly FFI code or rewriting cmd_set_option() in
Rust - neither of which are happening today.
So as an interim hack, add a --dev-idx option: this can be specified
(multiple times if desired) for indicating which devices the device
specific options should be applied to.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Fix set-passphrase on --no_passphrase FS
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Fix TODO "Show bcachefs version" in fusemount.
Signed-off-by: Integral <integral@archlinuxcn.org>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
This can prevent data loss as libblkid < 2.40.1 can fail
to recognize existing bcachefs filesystems.
Reported on #bcache.
Signed-off-by: Gabriel de Perthuis <g2p.code@gmail.com>
|
|
Currently, when BCACHEFS_FUSE=1 is set, bcachefs-tools failed to build.
Replace "bch2_trans_do()" macro with "bch2_trans_commit_do()" in
cmd_fusemount.c to fix build failure.
Signed-off-by: Integral <integral@archlinuxcn.org>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
This was added previously, but not added to the getopt string.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
bch_btree_ptr_v2.sectors_written validate
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
fstests generic/648 now passes
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
When static linking, it's not possible to have two functions with the
same name in different libraries:
= note: /nix/store/s2cjhni3s6shh0n35ay1hpj8d85k44qk-x86_64-unknown-linux-musl-binutils-2.43.1/bin/x86_64-unknown-linux-musl-ld: /nix/store/zwid05lzfy46k32xljb2alh0gq7ys58a-util-linux-minimal-static-x86_64-unknown-linux-musl-2.39.4-lib/lib/libblkid.a(libcommon_la-crc32c.o): in function `crc32c':
(.text+0x0): multiple definition of `crc32c'; ./libbcachefs.a(tools-util.o):/build/source/c_src/tools-util.c:484: first defined here
collect2: error: ld returned 1 exit status
Work around this by avoiding a known name collision with a function in
libblkid.
Closes: https://github.com/koverstreet/bcachefs-tools/issues/311
Signed-off-by: Alyssa Ross <hi@alyssa.is>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
This allows us to filter out e.g. all btree_node_write_work transacitons
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
When FUSE startup encounters an error after opening the filesystem--for
example because of a bad mountpoint--it exited uncleanly, which is a
bit unfriendly.
Signed-off-by: Thomas Bertschinger <tahbertschinger@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
for big endian builds
|
|
This also fixes a bug where the set-passphrase
command failed to derive the key for disks initially
formatted with --encrypted and --no_passphrase, due to
bch_sb_crypt_init not configuring the KDF params if
the passphrase wasn't specified during formatting.
Signed-off-by: Mae Kasza <git@badat.dev>
Tested-by: Jérôme Poulin <jeromepoulin@gmail.com>
|
|
no need to make it harder to dump than it needs to be
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
|
|
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|
|
set-option -> set-fs-option
setattr -> set-file-option
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
|