]> www.infradead.org Git - users/hch/xfsprogs.git/log
users/hch/xfsprogs.git
9 months agoxfs_db: make fsmap query the realtime reverse mapping tree
Darrick J. Wong [Thu, 15 Aug 2024 18:57:37 +0000 (11:57 -0700)]
xfs_db: make fsmap query the realtime reverse mapping tree

Extend the 'fsmap' debugger command to support querying the realtime
rmap btree via a new -r argument.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: copy the realtime rmap btree
Darrick J. Wong [Thu, 15 Aug 2024 18:57:37 +0000 (11:57 -0700)]
xfs_db: copy the realtime rmap btree

Copy the realtime rmapbt when we're metadumping the filesystem.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: support the realtime rmapbt
Darrick J. Wong [Thu, 15 Aug 2024 18:57:37 +0000 (11:57 -0700)]
xfs_db: support the realtime rmapbt

Wire up various parts of xfs_db for realtime rmap support so that we can
dump the btree contents.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: display the realtime rmap btree contents
Darrick J. Wong [Thu, 15 Aug 2024 18:57:37 +0000 (11:57 -0700)]
xfs_db: display the realtime rmap btree contents

Implement all the code we need to dump rtrmapbt contents, starting
from the inode root.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: don't abort when bmapping on a non-extents/bmbt fork
Darrick J. Wong [Thu, 15 Aug 2024 18:57:36 +0000 (11:57 -0700)]
xfs_db: don't abort when bmapping on a non-extents/bmbt fork

We're going to introduce new fork formats, so let's fix the problem that
xfs_db's bmap command aborts when the fork format isn't one of the
existing ones.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoman: document userspace API changes due to rt rmap
Darrick J. Wong [Tue, 1 Oct 2024 22:52:55 +0000 (15:52 -0700)]
man: document userspace API changes due to rt rmap

Update documentation to describe userspace ABI changes made for realtime
rmap support.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agolibfrog: enable scrubbing of the realtime rmap
Darrick J. Wong [Tue, 1 Oct 2024 22:49:12 +0000 (15:49 -0700)]
libfrog: enable scrubbing of the realtime rmap

Add a new entry so that we can scrub the rtrmapbt and its metadata
directory tree path too.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: create a shadow rmap btree during realtime rmap repair
Darrick J. Wong [Mon, 23 Sep 2024 20:42:18 +0000 (13:42 -0700)]
xfs: create a shadow rmap btree during realtime rmap repair

Create an in-memory btree of rmap records instead of an array.  This
enables us to do live record collection instead of freezing the fs.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: online repair of the realtime rmap btree
Darrick J. Wong [Thu, 15 Aug 2024 18:57:36 +0000 (11:57 -0700)]
xfs: online repair of the realtime rmap btree

Source kernel commit: f813af307d62d4c4d620a358bbd406f89ffdeca2

Repair the realtime rmap btree while mounted.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
9 months agoxfs: online repair of realtime bitmaps for a realtime group
Darrick J. Wong [Mon, 23 Sep 2024 20:42:16 +0000 (13:42 -0700)]
xfs: online repair of realtime bitmaps for a realtime group

For a given rt group, regenerate the bitmap contents from the group's
realtime rmap btree.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: scrub the metadir path of rt rmap btree files
Darrick J. Wong [Thu, 15 Aug 2024 18:49:04 +0000 (11:49 -0700)]
xfs: scrub the metadir path of rt rmap btree files

Add a new XFS_SCRUB_METAPATH subtype so that we can scrub the metadata
directory tree path to the rmap btree file for each rt group.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: scrub the realtime rmapbt
Darrick J. Wong [Mon, 23 Sep 2024 20:42:09 +0000 (13:42 -0700)]
xfs: scrub the realtime rmapbt

Check the realtime reverse mapping btree against the rtbitmap, and
modify the rtbitmap scrub to check against the rtrmapbt.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: report realtime rmap btree corruption errors to the health system
Darrick J. Wong [Thu, 15 Aug 2024 18:48:58 +0000 (11:48 -0700)]
xfs: report realtime rmap btree corruption errors to the health system

Whenever we encounter corrupt realtime rmap btree blocks, we should
report that to the health monitoring system for later reporting.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: create routine to allocate and initialize a realtime rmap btree inode
Darrick J. Wong [Mon, 23 Sep 2024 20:42:04 +0000 (13:42 -0700)]
xfs: create routine to allocate and initialize a realtime rmap btree inode

Create a library routine to allocate and initialize an empty realtime
rmapbt inode.  We'll use this for mkfs and repair.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: wire up rmap map and unmap to the realtime rmapbt
Darrick J. Wong [Mon, 23 Sep 2024 20:42:03 +0000 (13:42 -0700)]
xfs: wire up rmap map and unmap to the realtime rmapbt

Connect the map and unmap reverse-mapping operations to the realtime
rmapbt via the deferred operation callbacks.  This enables us to
perform rmap operations against the correct btree.

[Contains a minor bugfix from hch]

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: allow inodes with zero extents but nonzero nblocks
Darrick J. Wong [Mon, 23 Sep 2024 20:42:02 +0000 (13:42 -0700)]
xfs: allow inodes with zero extents but nonzero nblocks

Metadata inodes that store btrees will have zero extents and a nonzero
nblocks.  Adjust the inode verifier so that this combination is not
flagged.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: wire up a new inode fork type for the realtime rmap
Darrick J. Wong [Thu, 15 Aug 2024 18:48:53 +0000 (11:48 -0700)]
xfs: wire up a new inode fork type for the realtime rmap

Plumb in the pieces we need to embed the root of the realtime rmap
btree in an inode's data fork, complete with new fork type and
on-disk interpretation functions.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: add metadata reservations for realtime rmap btrees
Darrick J. Wong [Mon, 23 Sep 2024 20:42:01 +0000 (13:42 -0700)]
xfs: add metadata reservations for realtime rmap btrees

Reserve some free blocks so that we will always have enough free blocks
in the data volume to handle expansion of the realtime rmap btree.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: add realtime reverse map inode to metadata directory
Darrick J. Wong [Mon, 23 Sep 2024 20:42:00 +0000 (13:42 -0700)]
xfs: add realtime reverse map inode to metadata directory

Add a metadir path to select the realtime rmap btree inode and load
it at mount time.  The rtrmapbt inode will have a unique extent format
code, which means that we also have to update the inode validation and
flush routines to look for it.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: add a realtime flag to the rmap update log redo items
Darrick J. Wong [Mon, 23 Sep 2024 20:41:57 +0000 (13:41 -0700)]
xfs: add a realtime flag to the rmap update log redo items

Extend the rmap update (RUI) log items with a new realtime flag that
indicates that the updates apply against the realtime rmapbt.  We'll
wire up the actual rmap code later.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: prepare rmap functions to deal with rtrmapbt
Darrick J. Wong [Mon, 23 Sep 2024 20:41:57 +0000 (13:41 -0700)]
xfs: prepare rmap functions to deal with rtrmapbt

Prepare the high-level rmap functions to deal with the new realtime
rmapbt and its slightly different conventions.  Provide the ability
to talk to either rmapbt or rtrmapbt formats from the same high
level code.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: add realtime rmap btree operations
Darrick J. Wong [Thu, 15 Aug 2024 18:56:38 +0000 (11:56 -0700)]
xfs: add realtime rmap btree operations

Implement the generic btree operations needed to manipulate rtrmap
btree blocks. This is different from the regular rmapbt in that we
allocate space from the filesystem at large, and are neither
constrained to the free space nor any particular AG.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: realtime rmap btree transaction reservations
Darrick J. Wong [Thu, 15 Aug 2024 18:56:38 +0000 (11:56 -0700)]
xfs: realtime rmap btree transaction reservations

Source kernel commit: 2b08b631d6ad701ba6dda366fde4ae19cb66774a

Make sure that there's enough log reservation to handle mapping
and unmapping realtime extents.  We have to reserve enough space
to handle a split in the rtrmapbt to add the record and a second
split in the regular rmapbt to record the rtrmapbt split.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
9 months agoxfs: introduce realtime rmap btree ondisk definitions
Darrick J. Wong [Wed, 9 Oct 2024 22:30:51 +0000 (15:30 -0700)]
xfs: introduce realtime rmap btree ondisk definitions

Add the ondisk structure definitions for realtime rmap btrees. The
realtime rmap btree will be rooted from a hidden inode so it needs to
have a separate btree block magic and pointer format.

Next, add everything needed to read, write and manipulate rmap btree
blocks. This prepares the way for connecting the btree operations
implementation.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: simplify the xfs_rmap_{alloc,free}_extent calling conventions
Darrick J. Wong [Thu, 15 Aug 2024 18:56:37 +0000 (11:56 -0700)]
xfs: simplify the xfs_rmap_{alloc,free}_extent calling conventions

Simplify the calling conventions by allowing callers to pass a fsbno
(xfs_fsblock_t) directly into these functions, since we're just going to
set it in a struct anyway.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: allow inode-based btrees to reserve space in the data device
Darrick J. Wong [Tue, 1 Oct 2024 23:00:09 +0000 (16:00 -0700)]
xfs: allow inode-based btrees to reserve space in the data device

Create a new space reservation scheme so that btree metadata for the
realtime volume can reserve space in the data device to avoid space
underruns.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: update btree keys correctly when _insrec splits an inode root block
Darrick J. Wong [Thu, 15 Aug 2024 18:56:36 +0000 (11:56 -0700)]
xfs: update btree keys correctly when _insrec splits an inode root block

In commit 2c813ad66a72, I partially fixed a bug wherein xfs_btree_insrec
would erroneously try to update the parent's key for a block that had
been split if we decided to insert the new record into the new block.
The solution was to detect this situation and update the in-core key
value that we pass up to the caller so that the caller will (eventually)
add the new block to the parent level of the tree with the correct key.

However, I missed a subtlety about the way inode-rooted btrees work.  If
the full block was a maximally sized inode root block, we'll solve that
fullness by moving the root block's records to a new block, resizing the
root block, and updating the root to point to the new block.  We don't
pass a pointer to the new block to the caller because that work has
already been done.  The new record will /always/ land in the new block,
so in this case we need to use xfs_btree_update_keys to update the keys.

This bug can theoretically manifest itself in the very rare case that we
split a bmbt root block and the new record lands in the very first slot
of the new block, though I've never managed to trigger it in practice.
However, it is very easy to reproduce by running generic/522 with the
realtime rmapbt patchset if rtinherit=1.

Fixes: 2c813ad66a72 ("xfs: support btrees with overlapping intervals for keys")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: support storing records in the inode core root
Darrick J. Wong [Thu, 15 Aug 2024 18:48:39 +0000 (11:48 -0700)]
xfs: support storing records in the inode core root

Add the necessary flags and code so that we can support storing leaf
records in the inode root block of a btree.  This hasn't been necessary
before, but the realtime rmapbt will need to be able to do this.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: hoist the node iroot update code out of xfs_btree_kill_iroot
Darrick J. Wong [Thu, 15 Aug 2024 18:48:38 +0000 (11:48 -0700)]
xfs: hoist the node iroot update code out of xfs_btree_kill_iroot

In preparation for allowing records in an inode btree root, hoist the
code that copies keyptrs from an existing node child into the root block
to a separate function.  Remove some unnecessary conditionals and clean
up a few function calls in the new function.  Note that this change
reorders the ->free_block call with respect to the change in bc_nlevels
to make it easier to support inode root leaf blocks in the next patch.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: hoist the node iroot update code out of xfs_btree_new_iroot
Darrick J. Wong [Thu, 15 Aug 2024 18:48:38 +0000 (11:48 -0700)]
xfs: hoist the node iroot update code out of xfs_btree_new_iroot

In preparation for allowing records in an inode btree root, hoist the
code that copies keyptrs from an existing node root into a child block
to a separate function.  Note that the new function explicitly computes
the keys of the new child block and stores that in the root block; while
the bmap btree could rely on leaving the key alone, realtime rmap needs
to set the new high key.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: tidy up xfs_bmap_broot_realloc a bit
Darrick J. Wong [Fri, 30 Aug 2024 17:47:34 +0000 (10:47 -0700)]
xfs: tidy up xfs_bmap_broot_realloc a bit

Hoist out the code that migrates broot pointers during a resize
operation to avoid code duplication and streamline the caller.  Also
use the correct bmbt pointer type for the sizeof operation.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: make xfs_iroot_realloc a bmap btree function
Darrick J. Wong [Fri, 30 Aug 2024 00:42:23 +0000 (17:42 -0700)]
xfs: make xfs_iroot_realloc a bmap btree function

Move the inode fork btree root reallocation function part of the btree
ops because it's now mostly bmbt-specific code.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: make xfs_iroot_realloc take the new numrecs instead of deltas
Darrick J. Wong [Fri, 30 Aug 2024 03:08:34 +0000 (20:08 -0700)]
xfs: make xfs_iroot_realloc take the new numrecs instead of deltas

Change the calling signature of xfs_iroot_realloc to take the ifork and
the new number of records in the btree block, not a diff against the
current number.  This will make the callsites easier to understand.

Note that this function is misnamed because it is very specific to the
single type of inode-rooted btree supported.  This will be addressed in
a subsequent patch.

Return the new btree root to reduce the amount of code clutter.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: refactor the inode fork memory allocation functions
Darrick J. Wong [Thu, 15 Aug 2024 18:48:30 +0000 (11:48 -0700)]
xfs: refactor the inode fork memory allocation functions

Hoist the code that allocates, frees, and reallocates if_broot into a
single xfs_iroot_krealloc function.  Eventually we're going to push
xfs_iroot_realloc into the btree ops structure to handle multiple
inode-rooted btrees, but first let's separate out the bits that should
stay in xfs_inode_fork.c.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: tidy up xfs_iroot_realloc
Darrick J. Wong [Fri, 30 Aug 2024 00:17:48 +0000 (17:17 -0700)]
xfs: tidy up xfs_iroot_realloc

Tidy up this function a bit before we start refactoring the memory
handling and move the function to the bmbt code.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: enable metadata directory feature
Darrick J. Wong [Wed, 7 Aug 2024 22:54:28 +0000 (15:54 -0700)]
xfs: enable metadata directory feature

Enable the metadata directory feature.  With this feature, all metadata
inodes are placed in the metadata directory, and the only inumbers in
the superblock are the roots of the two directory trees.

The RT device is now sharded into a number of rtgroups, where 0 rtgroups
mean that no RT extents are supported, and the traditional XFS stub RT
bitmap and summary inodes don't exist.  A single rtgroup gives roughly
identical behavior to the traditional RT setup, but now with checksummed
and self identifying free space metadata.

For quota, the quota options are read from the superblock unless
explicitly overridden via mount options.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
9 months agomkfs: enable rt quota options
Darrick J. Wong [Thu, 22 Aug 2024 19:25:06 +0000 (12:25 -0700)]
mkfs: enable rt quota options

Now that the kernel supports quota and realtime devices, allow people to
format filesystems with permanent quota options.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_quota: report warning limits for realtime space quotas
Darrick J. Wong [Thu, 15 Aug 2024 18:58:33 +0000 (11:58 -0700)]
xfs_quota: report warning limits for realtime space quotas

Report the number of warnings that a user will get for exceeding the
soft limit of a realtime volume.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agomkfs: add quota flags when setting up filesystem
Darrick J. Wong [Thu, 22 Aug 2024 16:43:53 +0000 (09:43 -0700)]
mkfs: add quota flags when setting up filesystem

If we're creating a metadir filesystem, the quota accounting and
enforcement flags persist until the sysadmin changes them.  Add a means
to specify those qflags at format time.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: try not to trash qflags on metadir filesystems
Darrick J. Wong [Thu, 22 Aug 2024 16:43:35 +0000 (09:43 -0700)]
xfs_repair: try not to trash qflags on metadir filesystems

Try to preserve the accounting and enforcement quota flags when
repairing filesystems.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: support quota inodes in the metadata directory
Darrick J. Wong [Thu, 22 Aug 2024 16:43:34 +0000 (09:43 -0700)]
xfs_repair: support quota inodes in the metadata directory

Handle quota inodes on metadir filesystems.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: hoist the secondary sb qflags handling
Darrick J. Wong [Thu, 22 Aug 2024 16:43:34 +0000 (09:43 -0700)]
xfs_repair: hoist the secondary sb qflags handling

Hoist all the secondary superblock qflags and quota inode modification
code into a separate function so that we can disable it in the next
patch.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: refactor quota inumber handling
Darrick J. Wong [Thu, 22 Aug 2024 16:43:34 +0000 (09:43 -0700)]
xfs_repair: refactor quota inumber handling

In preparation for putting quota files in the metadata directory tree,
refactor repair's quota inumber handling to use its own variables
instead of the xfs_mount's.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: support metadir quotas
Darrick J. Wong [Thu, 22 Aug 2024 16:43:34 +0000 (09:43 -0700)]
xfs_db: support metadir quotas

Support finding the quota files in the metadata directory.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agolibfrog: scrub quota file metapaths
Darrick J. Wong [Thu, 22 Aug 2024 16:43:33 +0000 (09:43 -0700)]
libfrog: scrub quota file metapaths

Support scrubbing quota file metadir paths.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: scrub quota file metapaths
Darrick J. Wong [Thu, 22 Aug 2024 16:00:01 +0000 (09:00 -0700)]
xfs: scrub quota file metapaths

Enable online fsck for quota file metadata directory paths.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: use metadir for quota inodes
Darrick J. Wong [Thu, 22 Aug 2024 16:00:00 +0000 (09:00 -0700)]
xfs: use metadir for quota inodes

Store the quota inodes in the /quota metadata directory if metadir is
enabled.  This enables us to stop using the sb_[ugp]uotino fields in the
superblock.  From this point on, all metadata files will be children of
the metadata directory tree root.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
9 months agomkfs: format realtime groups
Darrick J. Wong [Thu, 15 Aug 2024 18:56:33 +0000 (11:56 -0700)]
mkfs: format realtime groups

Create filesystems with the realtime group feature enabled.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agomkfs: add headers to realtime bitmap blocks
Darrick J. Wong [Thu, 15 Aug 2024 18:56:32 +0000 (11:56 -0700)]
mkfs: add headers to realtime bitmap blocks

When the rtgroups feature is enabled, format rtbitmap blocks with the
appropriate block headers.  libxfs takes care of the actual writing for
us, so all we have to do is ensure that the bitmap is the correct size.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_scrub: use histograms to speed up phase 8 on the realtime volume
Darrick J. Wong [Thu, 15 Aug 2024 18:56:32 +0000 (11:56 -0700)]
xfs_scrub: use histograms to speed up phase 8 on the realtime volume

Use the same statistical methods that we use on the data volume to
compute the minimum threshold size for fstrims on the realtime volume.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_scrub: trim realtime volumes too
Darrick J. Wong [Thu, 15 Aug 2024 18:56:32 +0000 (11:56 -0700)]
xfs_scrub: trim realtime volumes too

On the kernel side, the XFS realtime groups patchset added support for
FITRIM of the realtime volume.  This support doesn't actually require
there to be any realtime groups, so teach scrub to run through the whole
region.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_scrub: call GETFSMAP for each rt group in parallel
Darrick J. Wong [Thu, 15 Aug 2024 18:56:32 +0000 (11:56 -0700)]
xfs_scrub: call GETFSMAP for each rt group in parallel

If realtime groups are enabled, we should take advantage of the sharding
to speed up the spacemap scans.  Do so by issuing per-rtgroup GETFSMAP
calls.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_scrub: check rtgroup metadata directory connections
Darrick J. Wong [Thu, 15 Aug 2024 18:56:31 +0000 (11:56 -0700)]
xfs_scrub: check rtgroup metadata directory connections

Run the rtgroup metapath scrubber during phase 5 to ensure that any
rtgroup metadata files are still connected to the metadir tree after
we've pruned any bad links.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_scrub: scrub realtime allocation group metadata
Darrick J. Wong [Thu, 15 Aug 2024 18:56:31 +0000 (11:56 -0700)]
xfs_scrub: scrub realtime allocation group metadata

Scan realtime group metadata as part of phase 2, just like we do for AG
metadata.  For pre-rtgroup filesystems, pretend that this is a "rtgroup
0" scrub request because the kernel expects that.  Replace the old
cond_wait code with a scrub barrier because they're equivalent for two
items that cannot be scrubbed in parallel.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_spaceman: report on realtime group health
Darrick J. Wong [Thu, 15 Aug 2024 18:56:31 +0000 (11:56 -0700)]
xfs_spaceman: report on realtime group health

Add the realtime group status to the health reporting done by
xfs_spaceman.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_io: display rt group in verbose fsmap output
Darrick J. Wong [Thu, 15 Aug 2024 18:56:31 +0000 (11:56 -0700)]
xfs_io: display rt group in verbose fsmap output

Display the rt group number in the fsmap output, just like we do for
regular data files.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_io: display rt group in verbose bmap output
Darrick J. Wong [Thu, 15 Aug 2024 18:56:31 +0000 (11:56 -0700)]
xfs_io: display rt group in verbose bmap output

Display the rt group number in the bmap -v output, just like we do for
regular data files.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_io: add a command to display realtime group information
Darrick J. Wong [Thu, 15 Aug 2024 18:56:30 +0000 (11:56 -0700)]
xfs_io: add a command to display realtime group information

Add a new 'rginfo' command to xfs_io so that we can display realtime
group geometry.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_io: add a command to display allocation group information
Darrick J. Wong [Thu, 15 Aug 2024 18:56:30 +0000 (11:56 -0700)]
xfs_io: add a command to display allocation group information

Add a new 'aginfo' command to xfs_io so that we can display allocation
group geometry.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_io: support scrubbing rtgroup metadata paths
Darrick J. Wong [Thu, 15 Aug 2024 18:56:30 +0000 (11:56 -0700)]
xfs_io: support scrubbing rtgroup metadata paths

Support scrubbing the metadata directory path of an rtgroup metadata
file.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_io: support scrubbing rtgroup metadata
Darrick J. Wong [Thu, 15 Aug 2024 18:56:30 +0000 (11:56 -0700)]
xfs_io: support scrubbing rtgroup metadata

Support scrubbing all rtgroup metadata with a scrubv call.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_mdrestore: restore rt group superblocks to realtime device
Darrick J. Wong [Thu, 15 Aug 2024 18:56:29 +0000 (11:56 -0700)]
xfs_mdrestore: restore rt group superblocks to realtime device

Support restoring realtime device metadata to the realtime device, if
the dumped filesystem had one.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: report rt group and block number in the bmap command
Darrick J. Wong [Thu, 26 Sep 2024 20:39:07 +0000 (13:39 -0700)]
xfs_db: report rt group and block number in the bmap command

The bmap command does not report startblocks for realtime files
correctly.  If rtgroups are enabled, we need to use the appropriate
functions to crack the startblock into rtgroup and block numbers; if
not, then we need to report a linear address and not try to report a
group number.

Fix both of these issues.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: dump rt summary blocks
Darrick J. Wong [Thu, 15 Aug 2024 18:56:29 +0000 (11:56 -0700)]
xfs_db: dump rt summary blocks

Now that rtsummary blocks have a header, make it so that xfs_db can
analyze the structure.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: dump rt bitmap blocks
Darrick J. Wong [Thu, 15 Aug 2024 18:56:29 +0000 (11:56 -0700)]
xfs_db: dump rt bitmap blocks

Now that rtbitmap blocks have a header, make it so that xfs_db can
analyze the structure.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: metadump realtime devices
Darrick J. Wong [Thu, 15 Aug 2024 18:56:29 +0000 (11:56 -0700)]
xfs_db: metadump realtime devices

Teach the metadump device to dump the filesystem metadata of a realtime
device to the metadump file.  Currently, this is limited to the realtime
superblock.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: metadump metadir rt bitmap and summary files
Christoph Hellwig [Thu, 15 Aug 2024 18:56:28 +0000 (11:56 -0700)]
xfs_db: metadump metadir rt bitmap and summary files

Don't skip dumping the data fork for regular files that are marked as
metadata inodes.  This catches rtbitmap and summary inodes on rtgroup
enabled file systems where their inode numbers aren't recorded in the
superblock.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: enable conversion of rt space units
Darrick J. Wong [Thu, 15 Aug 2024 18:56:28 +0000 (11:56 -0700)]
xfs_db: enable conversion of rt space units

Teach the xfs_db convert function about realtime extents, blocks, and
realtime group numbers.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: support changing the label and uuid of rt superblocks
Darrick J. Wong [Thu, 15 Aug 2024 18:56:28 +0000 (11:56 -0700)]
xfs_db: support changing the label and uuid of rt superblocks

Update the label and uuid commands to change the rt superblocks along
with the filesystem superblocks.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: support dumping realtime group data and superblocks
Darrick J. Wong [Thu, 15 Aug 2024 18:56:27 +0000 (11:56 -0700)]
xfs_db: support dumping realtime group data and superblocks

Allow dumping of realtime device superblocks and the new fields in the
primary superblock that were added for rtgroups support.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_db: listify the definition of enum typnm
Darrick J. Wong [Thu, 15 Aug 2024 18:56:27 +0000 (11:56 -0700)]
xfs_db: listify the definition of enum typnm

Convert the enum definition into a list so that future patches adding
things to enum typnm don't have to reflow the entire thing.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: stop tracking duplicate RT extents with rtgroups
Christoph Hellwig [Thu, 15 Aug 2024 19:41:39 +0000 (12:41 -0700)]
xfs_repair: stop tracking duplicate RT extents with rtgroups

Nothing ever looks them up, so don't bother with tracking them by
overloading the AG numbers.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: repair rtbitmap and rtsummary block headers
Darrick J. Wong [Thu, 15 Aug 2024 18:56:27 +0000 (11:56 -0700)]
xfs_repair: repair rtbitmap and rtsummary block headers

Check and repair the new block headers attached to rtbitmap and
rtsummary blocks.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: support realtime superblocks
Darrick J. Wong [Thu, 15 Aug 2024 18:56:27 +0000 (11:56 -0700)]
xfs_repair: support realtime superblocks

Support the realtime superblock feature.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: find and clobber rtgroup bitmap and summary files
Darrick J. Wong [Thu, 15 Aug 2024 18:56:26 +0000 (11:56 -0700)]
xfs_repair: find and clobber rtgroup bitmap and summary files

On a rtgroups filesystem, if the rtgroups bitmap or summary files are
garbage, we need to clear the dinode and update the incore bitmap so
that we don't bother to check the old rt freespace metadata.

However, we regenerate the entire rt metadata directory tree during
phase 6.  If the bitmap and summary files are ok, we still want to clear
the dinode, but we can still use the incore inode to check the old
freespace contents.  Split the clear_dinode function into two pieces,
one that merely zeroes the inode, and the old clear_dinode now turns off
checking.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agorepair: use a separate bmaps array for real time groups
Christoph Hellwig [Thu, 15 Aug 2024 19:41:38 +0000 (12:41 -0700)]
repair: use a separate bmaps array for real time groups

Stop pretending RTGs are high numbered AGs and just use separate
structures instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: support realtime groups
Christoph Hellwig [Thu, 15 Aug 2024 18:56:26 +0000 (11:56 -0700)]
xfs_repair: support realtime groups

Make repair aware of multiple rtgroups.  This now uses the same code as the
AG-based data device for block usage tracking instead of the less optimal
AVL trees and bitmaps used for the traditonal RT device.

Note this is still a bit hacky at the moment by just going beyond the AG
arrays and not fully supporting the unknown state for RT allocation yet.
The next patch will clean this up.

All this should be fixable.

Large parts of the code are based on patches from Darrick J. Wong.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: add a real per-AG bitmap abstraction
Christoph Hellwig [Thu, 15 Aug 2024 19:41:38 +0000 (12:41 -0700)]
xfs_repair: add a real per-AG bitmap abstraction

Add a struct bmap that contains the btree root and the lock, and provide
helpers for loking instead of directly poking into the data structure.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: simplify rt_lock handling
Christoph Hellwig [Thu, 15 Aug 2024 19:36:16 +0000 (12:36 -0700)]
xfs_repair: simplify rt_lock handling

No need to cacheline align rt_lock if we move it next to the data
it protects.  Also reduce the critical section to just where those
data structures are accessed.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: improve rtbitmap discrepancy reporting
Darrick J. Wong [Thu, 15 Aug 2024 18:56:26 +0000 (11:56 -0700)]
xfs_repair: improve rtbitmap discrepancy reporting

Improve the reporting of discrepancies in the realtime bitmap and
summary files by creating a separate helper function that will pinpoint
the exact (word) locations of mismatches.  This will help developers to
diagnose problems with the rtgroups feature and users to figure out
exactly what's bad in a filesystem.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: refactor offsetof+sizeof to offsetofend
Darrick J. Wong [Thu, 15 Aug 2024 18:56:26 +0000 (11:56 -0700)]
xfs_repair: refactor offsetof+sizeof to offsetofend

Replace this open-coded logic with the kernel's offsetofend macro before
we start adding more in the next patch.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: refactor phase4
Christoph Hellwig [Thu, 15 Aug 2024 18:56:25 +0000 (11:56 -0700)]
xfs_repair: refactor phase4

Split out helpers to process all duplicate extents in an AG and the RT
duplicate extents.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_repair: adjust rtbitmap/rtsummary word updates to handle big endian values
Darrick J. Wong [Tue, 1 Oct 2024 23:08:35 +0000 (16:08 -0700)]
xfs_repair: adjust rtbitmap/rtsummary word updates to handle big endian values

With rtgroups, the rt bitmap and summary file words are defined to be
be32 values.  Adjust repair to handle the endianness correctly.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs_logprint: report realtime EFIs
Darrick J. Wong [Thu, 15 Aug 2024 18:56:24 +0000 (11:56 -0700)]
xfs_logprint: report realtime EFIs

Decode the EFI format just enough to report if an EFI targets the
realtime device or not.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agolibfrog: add bitmap_clear
Darrick J. Wong [Thu, 15 Aug 2024 18:56:25 +0000 (11:56 -0700)]
libfrog: add bitmap_clear

Uncomment and fix bitmap_clear so that xfs_repair can start using it.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
[hch: split from a larger patch]
Signed-off-by: Christoph Hellwig <hch@lst.de>
9 months agolibfrog: report rt groups in output
Darrick J. Wong [Thu, 15 Aug 2024 18:56:25 +0000 (11:56 -0700)]
libfrog: report rt groups in output

Report realtime group geometry.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agolibfrog: support scrubbing rtgroup metadata paths
Darrick J. Wong [Thu, 15 Aug 2024 18:56:25 +0000 (11:56 -0700)]
libfrog: support scrubbing rtgroup metadata paths

Support scrubbing the metadata paths of rtgroup metadata inodes.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agolibxfs: implement some sanity checking for enormous rgcount
Darrick J. Wong [Thu, 15 Aug 2024 18:56:25 +0000 (11:56 -0700)]
libxfs: implement some sanity checking for enormous rgcount

Similar to what we do for suspiciously large sb_agcount values, if
someone tries to get libxfs to load a filesystem with a very large
realtime group count, let's do some basic checks of the rt device to
see if it's really that large.  If the read fails, only load the first
rtgroup and warn the user.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agolibxfs: port userspace deferred log item to handle rtgroups
Darrick J. Wong [Thu, 15 Aug 2024 18:56:22 +0000 (11:56 -0700)]
libxfs: port userspace deferred log item to handle rtgroups

Make the userspace log items to handle rt groups correctly.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoman: document the rt group geometry ioctl
Darrick J. Wong [Thu, 15 Aug 2024 18:56:20 +0000 (11:56 -0700)]
man: document the rt group geometry ioctl

Document the new ioctl that retrieves realtime allocation group geometry
information.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: implement busy extent tracking for rtgroups
Darrick J. Wong [Wed, 2 Oct 2024 15:52:07 +0000 (08:52 -0700)]
xfs: implement busy extent tracking for rtgroups

For rtgroups filesystems, track newly freed (rt) space through the log
until the rt EFIs have been committed to disk.  This way we ensure that
space cannot be reused until all traces of the old owner are gone.

As a fringe benefit, we now support -o discard on the realtime device.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: move the min and max group block numbers to xfs_group
Darrick J. Wong [Tue, 8 Oct 2024 23:41:12 +0000 (16:41 -0700)]
xfs: move the min and max group block numbers to xfs_group

Move the min and max agblock numbers to the generic xfs_group structure
so that we can start building validators for extents within an rtgroup.
While we're at it, use check_add_overflow for the extent length
computation because that has much better overflow checking.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: fix minor bug in xfs_verify_agbno
Darrick J. Wong [Wed, 9 Oct 2024 00:43:56 +0000 (17:43 -0700)]
xfs: fix minor bug in xfs_verify_agbno

There's a minor bug in xfs_verify_agbno -- min_block ought to be the
first agblock number in the AG that can be used by non-static metadata.
Unfortunately, we set it to the last agblock of the static metadata.
Fortunately this works due to the <= check, but this isn't technically
correct.

Instead, change the check to < and set it to the next agblock past the
static metadata.  This hasn't been an issue up to now, but we're going
to move these things into the generic group struct, and this will cause
problems with rtgroups, where min_block can be zero for an rtgroup.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: add a xfs_rtbno_is_group_start helper
Christoph Hellwig [Thu, 3 Oct 2024 20:33:37 +0000 (13:33 -0700)]
xfs: add a xfs_rtbno_is_group_start helper

Make the boundary condition flag more clear and implement it by
a single masking operation.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: remove __xfs_rtb_to_rgbno
Christoph Hellwig [Thu, 3 Oct 2024 20:22:57 +0000 (13:22 -0700)]
xfs: remove __xfs_rtb_to_rgbno

Now that we have a generic mask field that has the right value for
legacy file systems, the caller can triviall open code the masking.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: add a xfs_fsb_to_gbno helper
Christoph Hellwig [Thu, 3 Oct 2024 20:17:07 +0000 (13:17 -0700)]
xfs: add a xfs_fsb_to_gbno helper

Just one more generic group helper to fold.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: move the group geometry into struct xfs_groups
Christoph Hellwig [Thu, 3 Oct 2024 20:12:08 +0000 (13:12 -0700)]
xfs: move the group geometry into struct xfs_groups

Add/move the blocks, blklog and blkmask fields to the generic groups
structure so that code can work with AGs and RTGs by just using the
right index into the array.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: make xfs_rtblock_t a segmented address like xfs_fsblock_t
Darrick J. Wong [Mon, 30 Sep 2024 20:49:00 +0000 (13:49 -0700)]
xfs: make xfs_rtblock_t a segmented address like xfs_fsblock_t

Now that we've finished adding allocation groups to the realtime volume,
let's make the file block mapping address (xfs_rtblock_t) a segmented
value just like we do on the data device.  This means that group number
and block number conversions can be done with shifting and masking
instead of integer division.

While in theory we could continue caching the rgno shift value in
m_rgblklog, the fact that we now always use the shift value means that
we have an opportunity to increase the redundancy of the rt geometry by
storing it in the ondisk superblock and adding more sb verifier code.
Reuse the space vacated by sb_bad_feature2 to store the rgblklog value.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: create helpers to deal with rounding xfs_filblks_t to rtx boundaries
Darrick J. Wong [Mon, 30 Sep 2024 20:47:22 +0000 (13:47 -0700)]
xfs: create helpers to deal with rounding xfs_filblks_t to rtx boundaries

We're about to segment xfs_rtblock_t addresses, so we must create
type-specific helpers to do rt extent rounding of file mapping block
lengths because the rtb helpers soon will not do the right thing there.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
9 months agoxfs: create helpers to deal with rounding xfs_fileoff_t to rtx boundaries
Darrick J. Wong [Mon, 30 Sep 2024 20:43:12 +0000 (13:43 -0700)]
xfs: create helpers to deal with rounding xfs_fileoff_t to rtx boundaries

We're about to segment xfs_rtblock_t addresses, so we must create
type-specific helpers to do rt extent rounding of file block offsets
because the rtb helpers soon will not do the right thing there.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>