-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCZ90sDQAKCRCRxhvAZXjc
ooUnAQCaXv5U0GaEwkCcW78vw/dk7jyFG5LlrGUMvZV8MBSvuAEAsaPvU2uM6ZNf
743B8zOopOeX3Nwy8UKRcHk1nO1m5AY=
=NoZe
-----END PGP SIGNATURE-----
Merge tag 'vfs-6.15-rc1.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull vfs afs updates from Christian Brauner:
"This contains the work for afs for this cycle:
- Fix an occasional hang that's only really encountered when
rmmod'ing the kafs module
- Remove the "-o autocell" mount option. This is obsolete with the
dynamic root and removing it makes the next patch slightly easier
- Change how the dynamic root mount is constructed. Currently, the
root directory is (de)populated when it is (un)mounted if there are
cells already configured and, further, pairs of automount points
have to be created/removed each time a cell is added/deleted
This is changed so that readdir on the root dir lists all the known
cell automount pairs plus the @cell symlinks and the inodes and
dentries are constructed by lookup on demand. This simplifies the
cell management code
- A few improvements to the afs_volume and afs_server tracepoints
- Pass trace info into the afs_lookup_cell() function to allow the
trace log to indicate the purpose of the lookup
- Remove the 'net' parameter from afs_unuse_cell() as it's
superfluous
- In rxrpc, allow a kernel app (such as kafs) to store a word of
information on rxrpc_peer records
- Use the information stored on the rxrpc_peer record to point to the
afs_server record. This allows the server address lookup to be done
away with
- Simplify the afs_server ref/activity accounting to make each one
self-contained and not garbage collected from the cell management
work item
- Simplify the afs_cell ref/activity accounting to make each one of
these also self-contained and not driven by a central management
work item
The current code was intended to make it such that a single timer
for the namespace and one work item per cell could do all the work
required to maintain these records. This, however, made for some
sequencing problems when cleaning up these records. Further, the
attempt to pass refs along with timers and work items made getting
it right rather tricky when the timer or work item already had a
ref attached and now a ref had to be got rid of"
* tag 'vfs-6.15-rc1.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
afs: Simplify cell record handling
afs: Fix afs_server ref accounting
afs: Use the per-peer app data provided by rxrpc
rxrpc: Allow the app to store private data on peer structs
afs: Drop the net parameter from afs_unuse_cell()
afs: Make afs_lookup_cell() take a trace note
afs: Improve server refcount/active count tracing
afs: Improve afs_volume tracing to display a debug ID
afs: Change dynroot to create contents on demand
afs: Remove the "autocell" mount option
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCZ90rNwAKCRCRxhvAZXjc
onBJAP9Z8Ywmlb5KQ1E3HvDmkwyY6yOSyZ9/CmbzrkCJ8ywYkQD/d9/xt0EP/O/q
N8YtzXArHWt7u0YbcVpy9WK3F72BdwU=
=VJgY
-----END PGP SIGNATURE-----
Merge tag 'vfs-6.15-rc1.async.dir' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull vfs async dir updates from Christian Brauner:
"This contains cleanups that fell out of the work from async directory
handling:
- Change kern_path_locked() and user_path_locked_at() to never return
a negative dentry. This simplifies the usability of these helpers
in various places
- Drop d_exact_alias() from the remaining place in NFS where it is
still used. This also allows us to drop the d_exact_alias() helper
completely
- Drop an unnecessary call to fh_update() from nfsd_create_locked()
- Change i_op->mkdir() to return a struct dentry
Change vfs_mkdir() to return a dentry provided by the filesystems
which is hashed and positive. This allows us to reduce the number
of cases where the resulting dentry is not positive to very few
cases. The code in these places becomes simpler and easier to
understand.
- Repack DENTRY_* and LOOKUP_* flags"
* tag 'vfs-6.15-rc1.async.dir' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
doc: fix inline emphasis warning
VFS: Change vfs_mkdir() to return the dentry.
nfs: change mkdir inode_operation to return alternate dentry if needed.
fuse: return correct dentry for ->mkdir
ceph: return the correct dentry on mkdir
hostfs: store inode in dentry after mkdir if possible.
Change inode_operations.mkdir to return struct dentry *
nfsd: drop fh_update() from S_IFDIR branch of nfsd_create_locked()
nfs/vfs: discard d_exact_alias()
VFS: add common error checks to lookup_one_qstr_excl()
VFS: change kern_path_locked() and user_path_locked_at() to never return negative dentry
VFS: repack LOOKUP_ bit flags.
VFS: repack DENTRY_ flags.
Simplify afs_cell record handling to avoid very occasional races that cause
module removal to hang (it waits for all cell records to be removed).
There are two things that particularly contribute to the difficulty:
firstly, the code tries to pass a ref on the cell to the cell's maintenance
work item (which gets awkward if the work item is already queued); and,
secondly, there's an overall cell manager that tries to use just one timer
for the entire cell collection (to avoid having loads of timers). However,
both of these are probably unnecessarily restrictive.
To simplify this, the following changes are made:
(1) The cell record collection manager is removed. Each cell record
manages itself individually.
(2) Each afs_cell is given a second work item (cell->destroyer) that is
queued when its refcount reaches zero. This is not done in the
context of the putting thread as it might be in an inconvenient place
to sleep.
(3) Each afs_cell is given its own timer. The timer is used to expire the
cell record after a period of unuse if not otherwise pinned and can
also be used for other maintenance tasks if necessary (of which there
are currently none as DNS refresh is triggered by filesystem
operations).
(4) The afs_cell manager work item (cell->manager) is no longer given a
ref on the cell when queued; rather, the manager must be deleted.
This does away with the need to deal with the consequences of losing a
race to queue cell->manager. Clean up of extra queuing is deferred to
the destroyer.
(5) The cell destroyer work item makes sure the cell timer is removed and
that the normal cell work is cancelled before farming the actual
destruction off to RCU.
(6) When a network namespace is destroyed or the kafs module is unloaded,
it's now a simple matter of marking the namespace as dead then just
waking up all the cell work items. They will then remove and destroy
themselves once all remaining activity counts and/or a ref counts are
dropped. This makes sure that all server records are dropped first.
(7) The cell record state set is reduced to just four states: SETTING_UP,
ACTIVE, REMOVING and DEAD. The record persists in the active state
even when it's not being used until the time comes to remove it rather
than downgrading it to an inactive state from whence it can be
restored.
This means that the cell still appears in /proc and /afs when not in
use until it switches to the REMOVING state - at which point it is
removed.
Note that the REMOVING state is included so that someone wanting to
resurrect the cell record is forced to wait whilst the cell is torn
down in that state. Once it's in the DEAD state, it has been removed
from net->cells tree and is no longer findable and can be replaced.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org
Link: https://lore.kernel.org/r/20250224234154.2014840-16-dhowells@redhat.com/ # v1
Link: https://lore.kernel.org/r/20250310094206.801057-12-dhowells@redhat.com/ # v4
The current way that afs_server refs are accounted and cleaned up sometimes
cause rmmod to hang when it is waiting for cell records to be removed. The
problem is that the cell cleanup might occasionally happen before the
server cleanup and then there's nothing that causes the cell to
garbage-collect the remaining servers as they become inactive.
Partially fix this by:
(1) Give each afs_server record its own management timer that rather than
relying on the cell manager's central timer to drive each individual
cell's maintenance work item to garbage collect servers.
This timer is set when afs_unuse_server() reduces a server's activity
count to zero and will schedule the server's destroyer work item upon
firing.
(2) Give each afs_server record its own destroyer work item that removes
the record from the cell's database, shuts down the timer, cancels any
pending work for itself, sends an RPC to the server to cancel
outstanding callbacks.
This change, in combination with the timer, obviates the need to try
and coordinate so closely between the cell record and a bunch of other
server records to try and tear everything down in a coordinated
fashion. With this, the cell record is pinned until the server RCU is
complete and namespace/module removal will wait until all the cell
records are removed.
(3) Now that incoming calls are mapped to servers (and thus cells) using
data attached to an rxrpc_peer, the UUID-to-server mapping tree is
moved from the namespace to the cell (cell->fs_servers). This means
there can no longer be duplicates therein - and that allows the
mapping tree to be simpler as there doesn't need to be a chain of
same-UUID servers that are in different cells.
(4) The lock protecting the UUID mapping tree is switched to an
rw_semaphore on the cell rather than a seqlock on the namespace as
it's now only used during mounting in contexts in which we're allowed
to sleep.
(5) When it comes time for a cell that is being removed to purge its set
of servers, it just needs to iterate over them and wake them up. Once
a server becomes inactive, its destroyer work item will observe the
state of the cell and immediately remove that record.
(6) When a server record is removed, it is marked AFS_SERVER_FL_EXPIRED to
prevent reattempts at removal. The record will be dispatched to RCU
for destruction once its refcount reaches 0.
(7) The AFS_SERVER_FL_UNCREATED/CREATING flags are used to synchronise
simultaneous creation attempts. If one attempt fails, it will abandon
the attempt and allow another to try again.
Note that the record can't just be abandoned when dead as it's bound
into a server list attached to a volume and only subject to
replacement if the server list obtained for the volume from the VLDB
changes.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org
Link: https://lore.kernel.org/r/20250224234154.2014840-15-dhowells@redhat.com/ # v1
Link: https://lore.kernel.org/r/20250310094206.801057-11-dhowells@redhat.com/ # v4
Change the AFS dynamic root to do things differently:
(1) Rather than having the creation of cell records create inodes and
dentries for cell mountpoints, create them on demand during lookup.
This simplifies cell management and locking as we no longer have to
create these objects in advance *and* on speculative lookup by the
user for a cell that isn't precreated.
(2) Rather than using the libfs dentry-based readdir (the dentries now no
longer exist until accessed from (1)), have readdir generate the
contents by reading the list of cells. The @cell symlinks get pushed
in positions 2 and 3 if rootcell has been configured.
(3) Make the @cell symlink dentries persist for the life of the superblock
or until reclaimed, but make cell mountpoints disappear immediately if
unused.
It's not perfect as someone doing an "ls -l /afs" may create a whole
bunch of dentries which will be garbage collected immediately. But
any dentry that gets automounted will be pinned by the mount, so it
shouldn't be too bad.
(4) Allocate the inode numbers for the cell mountpoints from an IDR to
prevent duplicates appearing in the event it cycles round. The number
allocated from the IDR is doubled to provide two inode numbers - one
for the normal cell name (RO) and one for the dotted cell name (RW).
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org
Link: https://lore.kernel.org/r/20250224234154.2014840-8-dhowells@redhat.com/ # v1
Link: https://lore.kernel.org/r/20250310094206.801057-4-dhowells@redhat.com/ # v4
The ->get_link() method may be entered under RCU pathwalk conditions (in
which case, the dentry pointer is NULL). This is not taken account of by
afs_atcell_get_link() and lockdep will complain when it tries to lock an
rwsem.
Fix this by marking net->ws_cell as __rcu and using RCU access macros on it
and by making afs_atcell_get_link() just return a pointer to the name in
RCU pathwalk without taking net->cells_lock or a ref on the cell as RCU
will protect the name storage (the cell is already freed via call_rcu()).
Fixes: 30bca65bbbae ("afs: Make /afs/@cell and /afs/.@cell symlinks")
Reported-by: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org
Link: https://lore.kernel.org/r/20250310094206.801057-2-dhowells@redhat.com/ # v4
Some filesystems, such as NFS, cifs, ceph, and fuse, do not have
complete control of sequencing on the actual filesystem (e.g. on a
different server) and may find that the inode created for a mkdir
request already exists in the icache and dcache by the time the mkdir
request returns. For example, if the filesystem is mounted twice the
directory could be visible on the other mount before it is on the
original mount, and a pair of name_to_handle_at(), open_by_handle_at()
calls could instantiate the directory inode with an IS_ROOT() dentry
before the first mkdir returns.
This means that the dentry passed to ->mkdir() may not be the one that
is associated with the inode after the ->mkdir() completes. Some
callers need to interact with the inode after the ->mkdir completes and
they currently need to perform a lookup in the (rare) case that the
dentry is no longer hashed.
This lookup-after-mkdir requires that the directory remains locked to
avoid races. Planned future patches to lock the dentry rather than the
directory will mean that this lookup cannot be performed atomically with
the mkdir.
To remove this barrier, this patch changes ->mkdir to return the
resulting dentry if it is different from the one passed in.
Possible returns are:
NULL - the directory was created and no other dentry was used
ERR_PTR() - an error occurred
non-NULL - this other dentry was spliced in
This patch only changes file-systems to return "ERR_PTR(err)" instead of
"err" or equivalent transformations. Subsequent patches will make
further changes to some file-systems to return a correct dentry.
Not all filesystems reliably result in a positive hashed dentry:
- NFS, cifs, hostfs will sometimes need to perform a lookup of
the name to get inode information. Races could result in this
returning something different. Note that this lookup is
non-atomic which is what we are trying to avoid. Placing the
lookup in filesystem code means it only happens when the filesystem
has no other option.
- kernfs and tracefs leave the dentry negative and the ->revalidate
operation ensures that lookup will be called to correctly populate
the dentry. This could be fixed but I don't think it is important
to any of the users of vfs_mkdir() which look at the dentry.
The recommendation to use
d_drop();d_splice_alias()
is ugly but fits with current practice. A planned future patch will
change this.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: NeilBrown <neilb@suse.de>
Link: https://lore.kernel.org/r/20250227013949.536172-2-neilb@suse.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
Give an afs_server object a ref on the afs_cell object it points to so that
the cell doesn't get deleted before the server record.
Whilst this is circular (cell -> vol -> server_list -> server -> cell), the
ref only pins the memory, not the lifetime as that's controlled by the
activity counter. When the volume's activity counter reaches 0, it
detaches from the cell and discards its server list; when a cell's activity
counter reaches 0, it discards its root volume. At that point, the
circularity is cut.
Fixes: d2ddc776a458 ("afs: Overhaul volume and server record caching and fileserver rotation")
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
Link: https://patch.msgid.link/20250218192250.296870-6-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When allocating and building an afs_server_list struct object from a VLDB
record, we look up each server address to get the server record for it -
but a server may have more than one entry in the record and we discard the
duplicate pointers. Currently, however, when we discard, we only put a
server record, not unuse it - but the lookup got as an active-user count.
The active-user count on an afs_server_list object determines its lifetime
whereas the refcount keeps the memory backing it around. Failing to reduce
the active-user counter prevents the record from being cleaned up and can
lead to multiple copied being seen - and pointing to deleted afs_cell
objects and other such things.
Fix this by switching the incorrect 'put' to an 'unuse' instead.
Without this, occasionally, a dead server record can be seen in
/proc/net/afs/servers and list corruption may be observed:
list_del corruption. prev->next should be ffff888102423e40, but was 0000000000000000. (prev=ffff88810140cd38)
Fixes: 977e5f8ed0ab ("afs: Split the usage count on struct afs_server")
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
Link: https://patch.msgid.link/20250218192250.296870-5-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Most of the filesystem methods where we care about dentry name
and parent have their stability guaranteed by the callers;
->d_revalidate() is the major exception.
It's easy enough for callers to supply stable values for
expected name and expected parent of the dentry being
validated. That kills quite a bit of boilerplate in
->d_revalidate() instances, along with a bunch of races
where they used to access ->d_name without sufficient
precautions.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQQqUNBr3gm4hGXdBJlZ7Krx/gZQ6wUCZ5gkoQAKCRBZ7Krx/gZQ
6w9FAP4nyxNNWMjE1TwuWR/DNDMYYuw/qn/miZ88B5BUM8hzqgD/W2SjRvcbSaIm
xSIYpbtKgtqNU34P1PU+dBvL8Utz2AE=
=TWY8
-----END PGP SIGNATURE-----
Merge tag 'pull-revalidate' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs d_revalidate updates from Al Viro:
"Provide stable parent and name to ->d_revalidate() instances
Most of the filesystem methods where we care about dentry name and
parent have their stability guaranteed by the callers;
->d_revalidate() is the major exception.
It's easy enough for callers to supply stable values for expected name
and expected parent of the dentry being validated. That kills quite a
bit of boilerplate in ->d_revalidate() instances, along with a bunch
of races where they used to access ->d_name without sufficient
precautions"
* tag 'pull-revalidate' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
9p: fix ->rename_sem exclusion
orangefs_d_revalidate(): use stable parent inode and name passed by caller
ocfs2_dentry_revalidate(): use stable parent inode and name passed by caller
nfs: fix ->d_revalidate() UAF on ->d_name accesses
nfs{,4}_lookup_validate(): use stable parent inode passed by caller
gfs2_drevalidate(): use stable parent inode and name passed by caller
fuse_dentry_revalidate(): use stable parent inode and name passed by caller
vfat_revalidate{,_ci}(): use stable parent inode passed by caller
exfat_d_revalidate(): use stable parent inode passed by caller
fscrypt_d_revalidate(): use stable parent inode passed by caller
ceph_d_revalidate(): propagate stable name down into request encoding
ceph_d_revalidate(): use stable parent inode passed by caller
afs_d_revalidate(): use stable name and parent inode passed by caller
Pass parent directory inode and expected name to ->d_revalidate()
generic_ci_d_compare(): use shortname_storage
ext4 fast_commit: make use of name_snapshot primitives
dissolve external_name.u into separate members
make take_dentry_name_snapshot() lockless
dcache: back inline names with a struct-wrapped array of unsigned long
make sure that DNAME_INLINE_LEN is a multiple of word size
No need to bother with boilerplate for obtaining the latter and for
the former we really should not count upon ->d_name.name remaining
stable under us.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
->d_revalidate() often needs to access dentry parent and name; that has
to be done carefully, since the locking environment varies from caller
to caller. We are not guaranteed that dentry in question will not be
moved right under us - not unless the filesystem is such that nothing
on it ever gets renamed.
It can be dealt with, but that results in boilerplate code that isn't
even needed - the callers normally have just found the dentry via dcache
lookup and want to verify that it's in the right place; they already
have the values of ->d_parent and ->d_name stable. There is a couple
of exceptions (overlayfs and, to less extent, ecryptfs), but for the
majority of calls that song and dance is not needed at all.
It's easier to make ecryptfs and overlayfs find and pass those values if
there's a ->d_revalidate() instance to be called, rather than doing that
in the instances.
This commit only changes the calling conventions; making use of supplied
values is left to followups.
NOTE: some instances need more than just the parent - things like CIFS
may need to build an entire path from filesystem root, so they need
more precautions than the usual boilerplate. This series doesn't
do anything to that need - these filesystems have to keep their locking
mechanisms (rename_lock loops, use of dentry_path_raw(), private rwsem
a-la v9fs).
One thing to keep in mind when using name is that name->name will normally
point into the pathname being resolved; the filename in question occupies
name->len bytes starting at name->name, and there is NUL somewhere after it,
but it the next byte might very well be '/' rather than '\0'. Do not
ignore name->len.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Gabriel Krisman Bertazi <gabriel@krisman.be>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCZ4pS3AAKCRCRxhvAZXjc
okSwAPkB8Ra+oTplB/yzmab5kFB0+IUSHAiBfG6TCYb45op7wgEAs4+ignZkb+Bi
PsrfV7soiTGNUYSDVKOw7LS6PJEzkgA=
=3mcq
-----END PGP SIGNATURE-----
Merge tag 'vfs-6.14-rc1.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull afs updates from Christian Brauner:
"Dynamic root improvements:
- Create an /afs/.<cell> mountpoint to match the /afs/<cell>
mountpoint when a cell is created
- Add some more checks on cell names proposed by the user to prevent
dodgy symlink bodies from being created. Also prevent rootcell from
being altered once set to simplify the locking
- Change the handling of /afs/@cell from being a dentry name
substitution at lookup time to making it a symlink to the current
cell name and also provide a /afs/.@cell symlink to point to the
dotted cell mountpoint
Fixes:
- Fix the abort code check in the fallback handling for the
YFS.RemoveFile2 RPC call
- Use call->op->server() for oridnary filesystem RPC calls that have
an operation descriptor instead of call->server()"
* tag 'vfs-6.14-rc1.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
afs: Fix the fallback handling for the YFS.RemoveFile2 RPC call
afs: Make /afs/@cell and /afs/.@cell symlinks
afs: Add rootcell checks
afs: Make /afs/.<cell> as well as /afs/<cell> mountpoints
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCZ4pRKQAKCRCRxhvAZXjc
ov2dAQCULWjTBWdF8Ro2bfNeXzWvUUnSPjoLJ9B4xlrOB9c2MAEAiwkKHkzAxUco
hCvaRJc3H2ze2wrgbIABPKB2noQVVwk=
=4ojv
-----END PGP SIGNATURE-----
Merge tag 'vfs-6.14-rc1.netfs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull vfs netfs updates from Christian Brauner:
"This contains read performance improvements and support for monolithic
single-blob objects that have to be read/written as such (e.g. AFS
directory contents). The implementation of the two parts is interwoven
as each makes the other possible.
- Read performance improvements
The read performance improvements are intended to speed up some
loss of performance detected in cifs and to a lesser extend in afs.
The problem is that we queue too many work items during the
collection of read results: each individual subrequest is collected
by its own work item, and then they have to interact with each
other when a series of subrequests don't exactly align with the
pattern of folios that are being read by the overall request.
Whilst the processing of the pages covered by individual
subrequests as they complete potentially allows folios to be woken
in parallel and with minimum delay, it can shuffle wakeups for
sequential reads out of order - and that is the most common I/O
pattern.
The final assessment and cleanup of an operation is then held up
until the last I/O completes - and for a synchronous sequential
operation, this means the bouncing around of work items just adds
latency.
Two changes have been made to make this work:
(1) All collection is now done in a single "work item" that works
progressively through the subrequests as they complete (and
also dispatches retries as necessary).
(2) For readahead and AIO, this work item be done on a workqueue
and can run in parallel with the ultimate consumer of the data;
for synchronous direct or unbuffered reads, the collection is
run in the application thread and not offloaded.
Functions such as smb2_readv_callback() then just tell netfslib
that the subrequest has terminated; netfslib does a minimal bit of
processing on the spot - stat counting and tracing mostly - and
then queues/wakes up the worker. This simplifies the logic as the
collector just walks sequentially through the subrequests as they
complete and walks through the folios, if buffered, unlocking them
as it goes. It also keeps to a minimum the amount of latency
injected into the filesystem's low-level I/O handling
The way netfs supports filesystems using the deprecated
PG_private_2 flag is changed: folios are flagged and added to a
write request as they complete and that takes care of scheduling
the writes to the cache. The originating read request can then just
unlock the pages whatever happens.
- Single-blob object support
Single-blob objects are files for which the content of the file
must be read from or written to the server in a single operation
because reading them in parts may yield inconsistent results. AFS
directories are an example of this as there exists the possibility
that the contents are generated on the fly and would differ between
reads or might change due to third party interference.
Such objects will be written to and retrieved from the cache if one
is present, though we allow/may need to propose multiple
subrequests to do so. The important part is that read from/write to
the *server* is monolithic.
Single blob reading is, for the moment, fully synchronous and does
result collection in the application thread and, also for the
moment, the API is supplied the buffer in the form of a folio_queue
chain rather than using the pagecache.
- Related afs changes
This series makes a number of changes to the kafs filesystem,
primarily in the area of directory handling:
- AFS's FetchData RPC reply processing is made partially
asynchronous which allows the netfs_io_request's outstanding
operation counter to be removed as part of reducing the
collection to a single work item.
- Directory and symlink reading are plumbed through netfslib using
the single-blob object API and are now cacheable with fscache.
This also allows the afs_read struct to be eliminated and
netfs_io_subrequest to be used directly instead.
- Directory and symlink content are now stored in a folio_queue
buffer rather than in the pagecache. This means we don't require
the RCU read lock and xarray iteration to access it, and folios
won't randomly disappear under us because the VM wants them
back.
- The vnode operation lock is changed from a mutex struct to a
private lock implementation. The problem is that the lock now
needs to be dropped in a separate thread and mutexes don't
permit that.
- When a new directory or symlink is created, we now initialise it
locally and mark it valid rather than downloading it (we know
what it's likely to look like).
- We now use the in-directory hashtable to reduce the number of
entries we need to scan when doing a lookup. The edit routines
have to maintain the hash chains.
- Cancellation (e.g. by signal) of an async call after the
rxrpc_call has been set up is now offloaded to the worker thread
as there will be a notification from rxrpc upon completion. This
avoids a double cleanup.
- A "rolling buffer" implementation is created to abstract out the
two separate folio_queue chaining implementations I had (one for
read and one for write).
- Functions are provided to create/extend a buffer in a folio_queue
chain and tear it down again.
This is used to handle AFS directories, but could also be used to
create bounce buffers for content crypto and transport crypto.
- The was_async argument is dropped from netfs_read_subreq_terminated()
Instead we wake the read collection work item by either queuing it
or waking up the app thread.
- We don't need to use BH-excluding locks when communicating between
the issuing thread and the collection thread as neither of them now
run in BH context.
- Also included are a number of new tracepoints; a split of the
netfslib write collection code to put retrying into its own file
(it gets more complicated with content encryption).
- There are also some minor fixes AFS included, including fixing the
AFS directory format struct layout, reducing some directory
over-invalidation and making afs_mkdir() translate EEXIST to
ENOTEMPY (which is not available on all systems the servers
support).
- Finally, there's a patch to try and detect entry into the folio
unlock function with no folio_queue structs in the buffer (which
isn't allowed in the cases that can get there).
This is a debugging patch, but should be minimal overhead"
* tag 'vfs-6.14-rc1.netfs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (31 commits)
netfs: Report on NULL folioq in netfs_writeback_unlock_folios()
afs: Add a tracepoint for afs_read_receive()
afs: Locally initialise the contents of a new symlink on creation
afs: Use the contained hashtable to search a directory
afs: Make afs_mkdir() locally initialise a new directory's content
netfs: Change the read result collector to only use one work item
afs: Make {Y,}FS.FetchData an asynchronous operation
afs: Fix cleanup of immediately failed async calls
afs: Eliminate afs_read
afs: Use netfslib for symlinks, allowing them to be cached
afs: Use netfslib for directories
afs: Make afs_init_request() get a key if not given a file
netfs: Add support for caching single monolithic objects such as AFS dirs
netfs: Add functions to build/clean a buffer in a folio_queue
afs: Add more tracepoints to do with tracking validity
cachefiles: Add auxiliary data trace
cachefiles: Add some subrequest tracepoints
netfs: Remove some extraneous directory invalidations
afs: Fix directory format encoding struct
afs: Fix EEXIST error returned from afs_rmdir() to be ENOTEMPTY
...
Fix a pair of bugs in the fallback handling for the YFS.RemoveFile2 RPC
call:
(1) Fix the abort code check to also look for RXGEN_OPCODE. The lack of
this masks the second bug.
(2) call->server is now not used for ordinary filesystem RPC calls that
have an operation descriptor. Fix to use call->op->server instead.
Fixes: e49c7b2f6de7 ("afs: Build an abstraction around an "operation" concept")
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/109541.1736865963@warthog.procyon.org.uk
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Make /afs/@cell a symlink in the /afs dynamic root to match what other AFS
clients do rather than doing a substitution in the dentry name. This has
the bonus of being tab-expandable also.
Further, provide a /afs/.@cell symlink to point to the dotted cell share.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20250107183454.608451-4-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Add some checks for the validity of the cell name. It's may get put into a
symlink, so preclude it containing any slashes or "..". Also disallow
starting/ending with a dot. This makes /afs/@cell/ as a symlink less of a
security risk.
Also disallow multiple setting of /proc/net/afs/rootcell for any given
network namespace. Once set, the value may not be changed. This makes it
easier to only create /afs/@cell and /afs/.@cell if there's a rootcell.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20250107183454.608451-3-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
When a cell is instantiated, automatically create an /afs/.<cell>
mountpoint to match the /afs/<cell> mountpoint to match other AFS clients.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20250107183454.608451-2-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
syzbot reported a lock held when returning to userspace[1]. This is
because if argc is less than 0 and the function returns directly, the held
inode lock is not released.
Fix this by store the error in ret and jump to done to clean up instead of
returning directly.
[dh: Modified Lizhi Xu's original patch to make it honour the error code
from afs_split_string()]
[1]
WARNING: lock held when returning to user space!
6.13.0-rc3-syzkaller-00209-g499551201b5f #0 Not tainted
------------------------------------------------
syz-executor133/5823 is leaving the kernel with locks still held!
1 lock held by syz-executor133/5823:
#0: ffff888071cffc00 (&sb->s_type->i_mutex_key#9){++++}-{4:4}, at: inode_lock include/linux/fs.h:818 [inline]
#0: ffff888071cffc00 (&sb->s_type->i_mutex_key#9){++++}-{4:4}, at: afs_proc_addr_prefs_write+0x2bb/0x14e0 fs/afs/addr_prefs.c:388
Reported-by: syzbot+76f33569875eb708e575@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=76f33569875eb708e575
Signed-off-by: Lizhi Xu <lizhi.xu@windriver.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241226012616.2348907-1-lizhi.xu@windriver.com/
Link: https://lore.kernel.org/r/529850.1736261552@warthog.procyon.org.uk
Tested-by: syzbot+76f33569875eb708e575@syzkaller.appspotmail.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
The kafs filesystem limits the maximum length of a cell to 256 bytes, but a
problem occurs if someone actually does that: kafs tries to create a
directory under /proc/net/afs/ with the name of the cell, but that fails
with a warning:
WARNING: CPU: 0 PID: 9 at fs/proc/generic.c:405
because procfs limits the maximum filename length to 255.
However, the DNS limits the maximum lookup length and, by extension, the
maximum cell name, to 255 less two (length count and trailing NUL).
Fix this by limiting the maximum acceptable cellname length to 253. This
also allows us to be sure we can create the "/afs/.<cell>/" mountpoint too.
Further, split the YFS VL record cell name maximum to be the 256 allowed by
the protocol and ignore the record retrieved by YFSVL.GetCellName if it
exceeds 253.
Fixes: c3e9f888263b ("afs: Implement client support for the YFSVL.GetCellName RPC op")
Reported-by: syzbot+7848fee1f1e5c53f912b@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/r/6776d25d.050a0220.3a8527.0048.GAE@google.com/
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/376236.1736180460@warthog.procyon.org.uk
Tested-by: syzbot+7848fee1f1e5c53f912b@syzkaller.appspotmail.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Since we know what the contents of a symlink will be when we create it on
the server, initialise its contents locally too to avoid the need to
download it.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-31-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Each directory image contains a hashtable with 128 buckets to speed up
searching. Currently, kafs does not use this, but rather iterates over all
the occupied slots in the image as it can share this with readdir.
Switch kafs to use the hashtable for lookups to reduce the latency. Care
must be taken that the hash chains are acyclic.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-30-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Initialise a new directory's content when it is created by mkdir locally
rather than downloading the content from the server as we can predict what
it's going to look like.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-29-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Change the way netfslib collects read results to do all the collection for
a particular read request using a single work item that walks along the
subrequest queue as subrequests make progress or complete, unlocking folios
progressively rather than doing the unlock in parallel as parallel requests
come in.
The code is remodelled to be more like the write-side code, though only
using a single stream. This makes it more directly comparable and thus
easier to duplicate fixes between the two sides.
This has a number of advantages:
(1) It's simpler. There doesn't need to be a complex donation mechanism
to handle mismatches between the size and alignment of subrequests and
folios. The collector unlocks folios as the subrequests covering each
complete.
(2) It should cause less scheduler overhead as there's a single work item
in play unlocking pages in parallel when a read gets split up into a
lot of subrequests instead of one per subrequest.
Whilst the parallellism is nice in theory, in practice, the vast
majority of loads are sequential reads of the whole file, so
committing a bunch of threads to unlocking folios out of order doesn't
help in those cases.
(3) It should make it easier to implement content decryption. A folio
cannot be decrypted until all the requests that contribute to it have
completed - and, again, most loads are sequential and so, most of the
time, we want to begin decryption sequentially (though it's great if
the decryption can happen in parallel).
There is a disadvantage in that we're losing the ability to decrypt and
unlock things on an as-things-arrive basis which may affect some
applications.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-28-dhowells@redhat.com
cc: Jeff Layton <jlayton@kernel.org>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Make FS.FetchData and YFS.FetchData an asynchronous operation in that the
request is queued in AF_RXRPC and then we return to the caller rather than
waiting. Processing of the returning packets is then done inline if it's a
synchronous VFS/VM call (readdir, read_folio, sync DIO, prep for write) or
offloaded to a workqueue if asynchronous VM calls (eg. readahead, async
DIO).
This reduces the chain of workqueues invoking workqueues and cuts out some
of the overhead, driving rxrpc data extraction and netfslib read collection
from a thread that's going to block to completion anyway if possible.
The ->done() call op is also split with ->immediate_cancel() handling the
cancellation on failure to begin the call and ->done() handling the rest.
This means that the AFS async FetchData code doesn't try to terminate the
netfs subrequest twice.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-26-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
If we manage to begin an async call, but fail to transmit any data on it
due to a signal, we then abort it which causes a race between the
notification of call completion from rxrpc and our attempt to cancel the
notification. The notification will be necessary, however, for async
FetchData to terminate the netfs subrequest.
However, since we get a notification from rxrpc upon completion of a call
(aborted or otherwise), we can just leave it to that.
This leads to calls not getting cleaned up, but appearing in
/proc/net/rxrpc/calls as being aborted with code 6.
Fix this by making the "error_do_abort:" case of afs_make_call() abort the
call and then abandon it to the notification handler.
Fixes: 34fa47612bfe ("afs: Fix race in async call refcounting")
Reported-by: Marc Dionne <marc.dionne@auristor.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-25-dhowells@redhat.com
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Now that directory and symlink reads go through netfslib, the afs_read
struct is mostly redundant with almost all data duplicated in the
netfs_io_request and netfs_io_subrequest structs that are also available
any time we're doing a fetch.
Eliminate afs_read by moving the one field we still need there to the
afs_call struct (we may be given a different amount of data than what we
asked for and have to track what remains of that) and using the
netfs_io_subrequest directly instead.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-24-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
In the AFS ecosystem, directories are just a special type of file that is
downloaded and parsed locally. Download is done by the same mechanism as
ordinary files and the data can be cached. There is one important semantic
restriction on directories over files: the client must download the entire
directory in one go because, for example, the server could fabricate the
contents of the blob on the fly with each download and give a different
image each time.
So that we can cache the directory download, switch AFS directory support
over to using the netfslib single-object API, thereby allowing directory
content to be stored in the local cache.
To make this work, the following changes are made:
(1) A directory's contents are now stored in a folio_queue chain attached
to the afs_vnode (inode) struct rather than its associated pagecache,
though multipage folios are still used to hold the data. The folio
queue is discarded when the directory inode is evicted.
This also helps with the phasing out of ITER_XARRAY.
(2) Various directory operations are made to use and unuse the cache
cookie.
(3) The content checking, content dumping and content iteration are now
performed with a standard iov_iter iterator over the contents of the
folio queue.
(4) Iteration and modification must be done with the vnode's validate_lock
held. In conjunction with (1), this means that the iteration can be
done without the need to lock pages or take extra refs on them, unlike
when accessing ->i_pages.
(5) Convert to using netfs_read_single() to read data.
(6) Provide a ->writepages() to call netfs_writeback_single() to save the
data to the cache according to the VM's scheduling whilst holding the
validate_lock read-locked as (4).
(7) Change local directory image editing functions:
(a) Provide a function to get a specific block by number from the
folio_queue as we can no longer use the i_pages xarray to locate
folios by index. This uses a cursor to remember the current
position as we need to iterate through the directory contents.
The block is kmapped before being returned.
(b) Make the function in (a) extend the directory by an extra folio if
we run out of space.
(c) Raise the check of the block free space counter, for those blocks
that have one, higher in the function to eliminate a call to get a
block.
(d) Remove the page unlocking and putting done during the editing
loops. This is no longer necessary as the folio_queue holds the
references and the pages are no longer in the pagecache.
(e) Mark the inode dirty and pin the cache usage till writeback at the
end of a successful edit.
(8) Don't set the large_folios flag on the inode as we do the allocation
ourselves rather than the VM doing it automatically.
(9) Mark the inode as being a single object that isn't uploaded to the
server.
(10) Enable caching on directories.
(11) Only set the upload key for writeback for regular files.
Notes:
(*) We keep the ->release_folio(), ->invalidate_folio() and
->migrate_folio() ops as we set the mapping pointer on the folio.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-22-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Jeff Layton <jlayton@kernel.org>
cc: linux-afs@lists.infradead.org
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
In a future patch, AFS directory caching will go through netfslib and this
will involve, at times, running on behalf of ->lookup(), which doesn't
provide us with a file from which we can get an authentication key.
If a file isn't provided, make afs_init_request() get a key from the
process's keyrings instead when setting up a read.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-21-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Add wrappers to set and clear the callback promise and to mark a directory
as invalidated, and add tracepoints to track these events:
(1) afs_cb_promise: Log when a callback promise is set on a vnode.
(2) afs_vnode_invalid: Log when the server's callback promise for a vnode
is no longer valid and we need to refetch the vnode metadata.
(3) afs_dir_invalid: Log when the contents of a directory are marked
invalid and requiring refetching from the server and the cache
invalidating.
and two tracepoints to record data version number management:
(4) afs_set_dv: Log when the DV is recorded on a vnode.
(5) afs_dv_mismatch: Log when the DV recorded on a vnode plus the expected
delta for the operation does not match the DV we got back from the
server.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-18-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
The AFS directory format structure, union afs_xdr_dir_block::meta, has too
many alloc counter slots declared and so pushes the hash table along and
over the data. This doesn't cause a problem at the moment because I'm
currently ignoring the hash table and only using the correct number of
alloc_ctrs in the code anyway. In future, however, I should start using
the hash table to try and speed up afs_lookup().
Fix this by using the correct constant to declare the counter array.
Fixes: 4ea219a839bf ("afs: Split the directory content defs into a header")
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-14-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
AFS servers pass back a code indicating EEXIST when they're asked to remove
a directory that is not empty rather than ENOTEMPTY because not all the
systems that an AFS server can run on have the latter error available and
AFS preexisted the addition of that error in general.
Fix afs_rmdir() to translate EEXIST to ENOTEMPTY.
Fixes: 260a980317da ("[AFS]: Add "directory write" support.")
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-13-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Don't use the standard mutex for the I/O operation lock, but rather
implement our own as the standard mutex must be released in the same thread
as locked it. This is a problem when it comes to doing async FetchData
where the lock will be dropped from the workqueue that processed the
incoming data and not from the issuing thread.
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-12-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Drop the was_async argument from netfs_read_subreq_terminated(). Almost
every caller is either in process context and passes false. Some
filesystems delegate the call to a workqueue to avoid doing the work in
their network message queue parsing thread.
The only exception is netfs_cache_read_terminated() which handles
completion in the cache - which is usually a callback from the backing
filesystem in softirq context, though it can be from process context if an
error occurred. In this case, delegate to a workqueue.
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wiVC5Cgyz6QKXFu6fTaA6h4CjexDR-OV9kL6Vo5x9v8=A@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241216204124.3752367-10-dhowells@redhat.com
cc: Jeff Layton <jlayton@kernel.org>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
syzkaller reported recursion with a loop of three calls (netfs_rreq_assess,
netfs_retry_reads and netfs_rreq_terminated) hitting the limit of the stack
during an unbuffered or direct I/O read.
There are a number of issues:
(1) There is no limit on the number of retries.
(2) A subrequest is supposed to be abandoned if it does not transfer
anything (NETFS_SREQ_NO_PROGRESS), but that isn't checked under all
circumstances.
(3) The actual root cause, which is this:
if (atomic_dec_and_test(&rreq->nr_outstanding))
netfs_rreq_terminated(rreq, ...);
When we do a retry, we bump the rreq->nr_outstanding counter to
prevent the final cleanup phase running before we've finished
dispatching the retries. The problem is if we hit 0, we have to do
the cleanup phase - but we're in the cleanup phase and end up
repeating the retry cycle, hence the recursion.
Work around the problem by limiting the number of retries. This is based
on Lizhi Xu's patch[1], and makes the following changes:
(1) Replace NETFS_SREQ_NO_PROGRESS with NETFS_SREQ_MADE_PROGRESS and make
the filesystem set it if it managed to read or write at least one byte
of data. Clear this bit before issuing a subrequest.
(2) Add a ->retry_count member to the subrequest and increment it any time
we do a retry.
(3) Remove the NETFS_SREQ_RETRYING flag as it is superfluous with
->retry_count. If the latter is non-zero, we're doing a retry.
(4) Abandon a subrequest if retry_count is non-zero and we made no
progress.
(5) Use ->retry_count in both the write-side and the read-size.
[?] Question: Should I set a hard limit on retry_count in both read and
write? Say it hits 50, we always abandon it. The problem is that
these changes only mitigate the issue. As long as it made at least one
byte of progress, the recursion is still an issue. This patch
mitigates the problem, but does not fix the underlying cause. I have
patches that will do that, but it's an intrusive fix that's currently
pending for the next merge window.
The oops generated by KASAN looks something like:
BUG: TASK stack guard page was hit at ffffc9000482ff48 (stack is ffffc90004830000..ffffc90004838000)
Oops: stack guard page: 0000 [#1] PREEMPT SMP KASAN NOPTI
...
RIP: 0010:mark_lock+0x25/0xc60 kernel/locking/lockdep.c:4686
...
mark_usage kernel/locking/lockdep.c:4646 [inline]
__lock_acquire+0x906/0x3ce0 kernel/locking/lockdep.c:5156
lock_acquire.part.0+0x11b/0x380 kernel/locking/lockdep.c:5825
local_lock_acquire include/linux/local_lock_internal.h:29 [inline]
___slab_alloc+0x123/0x1880 mm/slub.c:3695
__slab_alloc.constprop.0+0x56/0xb0 mm/slub.c:3908
__slab_alloc_node mm/slub.c:3961 [inline]
slab_alloc_node mm/slub.c:4122 [inline]
kmem_cache_alloc_noprof+0x2a7/0x2f0 mm/slub.c:4141
radix_tree_node_alloc.constprop.0+0x1e8/0x350 lib/radix-tree.c:253
idr_get_free+0x528/0xa40 lib/radix-tree.c:1506
idr_alloc_u32+0x191/0x2f0 lib/idr.c:46
idr_alloc+0xc1/0x130 lib/idr.c:87
p9_tag_alloc+0x394/0x870 net/9p/client.c:321
p9_client_prepare_req+0x19f/0x4d0 net/9p/client.c:644
p9_client_zc_rpc.constprop.0+0x105/0x880 net/9p/client.c:793
p9_client_read_once+0x443/0x820 net/9p/client.c:1570
p9_client_read+0x13f/0x1b0 net/9p/client.c:1534
v9fs_issue_read+0x115/0x310 fs/9p/vfs_addr.c:74
netfs_retry_read_subrequests fs/netfs/read_retry.c:60 [inline]
netfs_retry_reads+0x153a/0x1d00 fs/netfs/read_retry.c:232
netfs_rreq_assess+0x5d3/0x870 fs/netfs/read_collect.c:371
netfs_rreq_terminated+0xe5/0x110 fs/netfs/read_collect.c:407
netfs_retry_reads+0x155e/0x1d00 fs/netfs/read_retry.c:235
netfs_rreq_assess+0x5d3/0x870 fs/netfs/read_collect.c:371
netfs_rreq_terminated+0xe5/0x110 fs/netfs/read_collect.c:407
netfs_retry_reads+0x155e/0x1d00 fs/netfs/read_retry.c:235
netfs_rreq_assess+0x5d3/0x870 fs/netfs/read_collect.c:371
...
netfs_rreq_terminated+0xe5/0x110 fs/netfs/read_collect.c:407
netfs_retry_reads+0x155e/0x1d00 fs/netfs/read_retry.c:235
netfs_rreq_assess+0x5d3/0x870 fs/netfs/read_collect.c:371
netfs_rreq_terminated+0xe5/0x110 fs/netfs/read_collect.c:407
netfs_retry_reads+0x155e/0x1d00 fs/netfs/read_retry.c:235
netfs_rreq_assess+0x5d3/0x870 fs/netfs/read_collect.c:371
netfs_rreq_terminated+0xe5/0x110 fs/netfs/read_collect.c:407
netfs_dispatch_unbuffered_reads fs/netfs/direct_read.c:103 [inline]
netfs_unbuffered_read fs/netfs/direct_read.c:127 [inline]
netfs_unbuffered_read_iter_locked+0x12f6/0x19b0 fs/netfs/direct_read.c:221
netfs_unbuffered_read_iter+0xc5/0x100 fs/netfs/direct_read.c:256
v9fs_file_read_iter+0xbf/0x100 fs/9p/vfs_file.c:361
do_iter_readv_writev+0x614/0x7f0 fs/read_write.c:832
vfs_readv+0x4cf/0x890 fs/read_write.c:1025
do_preadv fs/read_write.c:1142 [inline]
__do_sys_preadv fs/read_write.c:1192 [inline]
__se_sys_preadv fs/read_write.c:1187 [inline]
__x64_sys_preadv+0x22d/0x310 fs/read_write.c:1187
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83
Fixes: ee4cdf7ba857 ("netfs: Speed up buffered reading")
Closes: https://syzkaller.appspot.com/bug?extid=1fc6f64c40a9d143cfb6
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20241108034020.3695718-1-lizhi.xu@windriver.com/ [1]
Link: https://lore.kernel.org/r/20241213135013.2964079-9-dhowells@redhat.com
Tested-by: syzbot+885c03ad650731743489@syzkaller.appspotmail.com
Suggested-by: Lizhi Xu <lizhi.xu@windriver.com>
cc: Dominique Martinet <asmadeus@codewreck.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: v9fs@lists.linux.dev
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
Reported-by: syzbot+885c03ad650731743489@syzkaller.appspotmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCZyTGAQAKCRCRxhvAZXjc
opd6AQCal4omyfS8FYe4VRRZ/0XHouagq99I0U0TAmKkvoKAsgD/XrdE+pSTEkPX
Pv4T9phh1cZRxcyKVu77UoYkuHJEDAg=
=Lu9R
-----END PGP SIGNATURE-----
Merge tag 'vfs-6.12-rc6.fixes' of gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs
Pull filesystem fixes from Christian Brauner:
"VFS:
- Fix copy_page_from_iter_atomic() if KMAP_LOCAL_FORCE_MAP=y is set
- Add a get_tree_bdev_flags() helper that allows to modify e.g.,
whether errors are logged into the filesystem context during
superblock creation. This is used by erofs to fix a userspace
regression where an error is currently logged when its used on a
regular file which is an new allowed mode in erofs.
netfs:
- Fix the sysfs debug path in the documentation.
- Fix iov_iter_get_pages*() for folio queues by skipping the page
extracation if we're at the end of a folio.
afs:
- Fix moving subdirectories to different parent directory.
autofs:
- Fix handling of AUTOFS_DEV_IOCTL_TIMEOUT_CMD ioctl in
validate_dev_ioctl(). The actual ioctl number, not the ioctl
command needs to be checked for autofs"
* tag 'vfs-6.12-rc6.fixes' of gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs:
iov_iter: fix copy_page_from_iter_atomic() if KMAP_LOCAL_FORCE_MAP
autofs: fix thinko in validate_dev_ioctl()
iov_iter: Fix iov_iter_get_pages*() for folio_queue
afs: Fix missing subdir edit when renamed between parent dirs
doc: correcting the debug path for cachefiles
erofs: use get_tree_bdev_flags() to avoid misleading messages
fs/super.c: introduce get_tree_bdev_flags()
When rename moves an AFS subdirectory between parent directories, the
subdir also needs a bit of editing: the ".." entry needs updating to point
to the new parent (though I don't make use of the info) and the DV needs
incrementing by 1 to reflect the change of content. The server also sends
a callback break notification on the subdirectory if we have one, but we
can take care of recovering the promise next time we access the subdir.
This can be triggered by something like:
mount -t afs %example.com:xfstest.test20 /xfstest.test/
mkdir /xfstest.test/{aaa,bbb,aaa/ccc}
touch /xfstest.test/bbb/ccc/d
mv /xfstest.test/{aaa/ccc,bbb/ccc}
touch /xfstest.test/bbb/ccc/e
When the pathwalk for the second touch hits "ccc", kafs spots that the DV
is incorrect and downloads it again (so the fix is not critical).
Fix this, if the rename target is a directory and the old and new
parents are different, by:
(1) Incrementing the DV number of the target locally.
(2) Editing the ".." entry in the target to refer to its new parent's
vnode ID and uniquifier.
Link: https://lore.kernel.org/r/3340431.1729680010@warthog.procyon.org.uk
Fixes: 63a4681ff39c ("afs: Locally edit directory data for mkdir/create/unlink/...")
cc: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
afs_wake_up_async_call() can incur lock recursion. The problem is that it
is called from AF_RXRPC whilst holding the ->notify_lock, but it tries to
take a ref on the afs_call struct in order to pass it to a work queue - but
if the afs_call is already queued, we then have an extraneous ref that must
be put... calling afs_put_call() may call back down into AF_RXRPC through
rxrpc_kernel_shutdown_call(), however, which might try taking the
->notify_lock again.
This case isn't very common, however, so defer it to a workqueue. The oops
looks something like:
BUG: spinlock recursion on CPU#0, krxrpcio/7001/1646
lock: 0xffff888141399b30, .magic: dead4ead, .owner: krxrpcio/7001/1646, .owner_cpu: 0
CPU: 0 UID: 0 PID: 1646 Comm: krxrpcio/7001 Not tainted 6.12.0-rc2-build3+ #4351
Hardware name: ASUS All Series/H97-PLUS, BIOS 2306 10/09/2014
Call Trace:
<TASK>
dump_stack_lvl+0x47/0x70
do_raw_spin_lock+0x3c/0x90
rxrpc_kernel_shutdown_call+0x83/0xb0
afs_put_call+0xd7/0x180
rxrpc_notify_socket+0xa0/0x190
rxrpc_input_split_jumbo+0x198/0x1d0
rxrpc_input_data+0x14b/0x1e0
? rxrpc_input_call_packet+0xc2/0x1f0
rxrpc_input_call_event+0xad/0x6b0
rxrpc_input_packet_on_conn+0x1e1/0x210
rxrpc_input_packet+0x3f2/0x4d0
rxrpc_io_thread+0x243/0x410
? __pfx_rxrpc_io_thread+0x10/0x10
kthread+0xcf/0xe0
? __pfx_kthread+0x10/0x10
ret_from_fork+0x24/0x40
? __pfx_kthread+0x10/0x10
ret_from_fork_asm+0x1a/0x30
</TASK>
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/1394602.1729162732@warthog.procyon.org.uk
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
In afs_wait_for_operation(), we set transcribe the call responded flag to
the server record that we used after doing the fileserver iteration loop -
but it's possible to exit the loop having had a response from the server
that we've discarded (e.g. it returned an abort or we started receiving
data, but the call didn't complete).
This means that op->server might be NULL, but we don't check that before
attempting to set the server flag.
Fixes: 98f9fda2057b ("afs: Fold the afs_addr_cursor struct in")
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/r/20240923150756.902363-7-dhowells@redhat.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
