Pull dcache busy loop updates from Al Viro:
"Fix livelocks in shrink_dcache_tree()
If shrink_dcache_tree() finds a dentry in the middle of being killed
by another thread, it has to wait until the victim finishes dying,
gets detached from the tree and ceases to pin its parent.
The way we used to deal with that amounted to busy-wait;
unfortunately, it's not just inefficient but can lead to reliably
reproducible hard livelocks.
Solved by having shrink_dentry_tree() attach a completion to such
dentry, with dentry_unlist() calling complete() on all objects
attached to it. With a bit of care it can be done without growing
struct dentry or adding overhead in normal case"
* tag 'pull-dcache-busy-wait' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
get rid of busy-waiting in shrink_dcache_tree()
dcache.c: more idiomatic "positives are not allowed" sanity checks
struct dentry: make ->d_u anonymous
for_each_alias(): helper macro for iterating through dentries of given inode
Most of the places using d_alias are loops iterating through all aliases for
given inode; introduce a helper macro (for_each_alias(dentry, inode))
and convert open-coded instances of such loop to it.
They are easier to read that way and it reduces the noise on the next steps.
You _must_ hold inode->i_lock over that thing.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
On 32-bit architectures, unsigned long is only 32 bits wide, which
causes 64-bit inode numbers to be silently truncated. Several
filesystems (NFS, XFS, BTRFS, etc.) can generate inode numbers that
exceed 32 bits, and this truncation can lead to inode number collisions
and other subtle bugs on 32-bit systems.
Change the type of inode->i_ino from unsigned long to u64 to ensure that
inode numbers are always represented as 64-bit values regardless of
architecture. Update all format specifiers treewide from %lu/%lx to
%llu/%llx to match the new type, along with corresponding local variable
types.
This is the bulk treewide conversion. Earlier patches in this series
handled trace events separately to allow trace field reordering for
better struct packing on 32-bit.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Link: https://patch.msgid.link/20260304-iino-u64-v3-12-2257ad83d372@kernel.org
Acked-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Pull misc vfs updates from Christian Brauner:
"This contains the usual selections of misc updates for this cycle.
Features:
- Use folios for symlinks in the page cache
FUSE already uses folios for its symlinks. Mirror that conversion
in the generic code and the NFS code. That lets us get rid of a few
folio->page->folio conversions in this path, and some of the few
remaining users of read_cache_page() / read_mapping_page()
- Try and make a few filesystem operations killable on the VFS
inode->i_mutex level
- Add sysctl vfs_cache_pressure_denom for bulk file operations
Some workloads need to preserve more dentries than we currently
allow through out sysctl interface
A HDFS servers with 12 HDDs per server, on a HDFS datanode startup
involves scanning all files and caching their metadata (including
dentries and inodes) in memory. Each HDD contains approximately 2
million files, resulting in a total of ~20 million cached dentries
after initialization
To minimize dentry reclamation, they set vfs_cache_pressure to 1.
Despite this configuration, memory pressure conditions can still
trigger reclamation of up to 50% of cached dentries, reducing the
cache from 20 million to approximately 10 million entries. During
the subsequent cache rebuild period, any HDFS datanode restart
operation incurs substantial latency penalties until full cache
recovery completes
To maintain service stability, more dentries need to be preserved
during memory reclamation. The current minimum reclaim ratio (1/100
of total dentries) remains too aggressive for such workload. This
patch introduces vfs_cache_pressure_denom for more granular cache
pressure control
The configuration [vfs_cache_pressure=1,
vfs_cache_pressure_denom=10000] effectively maintains the full 20
million dentry cache under memory pressure, preventing datanode
restart performance degradation
- Avoid some jumps in inode_permission() using likely()/unlikely()
- Avid a memory access which is most likely a cache miss when
descending into devcgroup_inode_permission()
- Add fastpath predicts for stat() and fdput()
- Anonymous inodes currently don't come with a proper mode causing
issues in the kernel when we want to add useful VFS debug assert.
Fix that by giving them a proper mode and masking it off when we
report it to userspace which relies on them not having any mode
- Anonymous inodes currently allow to change inode attributes because
the VFS falls back to simple_setattr() if i_op->setattr isn't
implemented. This means the ownership and mode for every single
user of anon_inode_inode can be changed. Block that as it's either
useless or actively harmful. If specific ownership is needed the
respective subsystem should allocate anonymous inodes from their
own private superblock
- Raise SB_I_NODEV and SB_I_NOEXEC on the anonymous inode superblock
- Add proper tests for anonymous inode behavior
- Make it easy to detect proper anonymous inodes and to ensure that
we can detect them in codepaths such as readahead()
Cleanups:
- Port pidfs to the new anon_inode_{g,s}etattr() helpers
- Try to remove the uselib() system call
- Add unlikely branch hint return path for poll
- Add unlikely branch hint on return path for core_sys_select
- Don't allow signals to interrupt getdents copying for fuse
- Provide a size hint to dir_context for during readdir()
- Use writeback_iter directly in mpage_writepages
- Update compression and mtime descriptions in initramfs
documentation
- Update main netfs API document
- Remove useless plus one in super_cache_scan()
- Remove unnecessary NULL-check guards during setns()
- Add separate separate {get,put}_cgroup_ns no-op cases
Fixes:
- Fix typo in root= kernel parameter description
- Use KERN_INFO for infof()|info_plog()|infofc()
- Correct comments of fs_validate_description()
- Mark an unlikely if condition with unlikely() in
vfs_parse_monolithic_sep()
- Delete macro fsparam_u32hex()
- Remove unused and problematic validate_constant_table()
- Fix potential unsigned integer underflow in fs_name()
- Make file-nr output the total allocated file handles"
* tag 'vfs-6.16-rc1.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (43 commits)
fs: Pass a folio to page_put_link()
nfs: Use a folio in nfs_get_link()
fs: Convert __page_get_link() to use a folio
fs/read_write: make default_llseek() killable
fs/open: make do_truncate() killable
fs/open: make chmod_common() and chown_common() killable
include/linux/fs.h: add inode_lock_killable()
readdir: supply dir_context.count as readdir buffer size hint
vfs: Add sysctl vfs_cache_pressure_denom for bulk file operations
fuse: don't allow signals to interrupt getdents copying
Documentation: fix typo in root= kernel parameter description
include/cgroup: separate {get,put}_cgroup_ns no-op case
kernel/nsproxy: remove unnecessary guards
fs: use writeback_iter directly in mpage_writepages
fs: remove useless plus one in super_cache_scan()
fs: add S_ANON_INODE
fs: remove uselib() system call
device_cgroup: avoid access to ->i_rdev in the common case in devcgroup_inode_permission()
fs/fs_parse: Remove unused and problematic validate_constant_table()
fs: touch up predicts in inode_permission()
...
The family of functions:
lookup_one()
lookup_one_unlocked()
lookup_one_positive_unlocked()
appear designed to be used by external clients of the filesystem rather
than by filesystems acting on themselves as the lookup_one_len family
are used.
They are used by:
btrfs/ioctl - which is a user-space interface rather than an internal
activity
exportfs - i.e. from nfsd or the open_by_handle_at interface
overlayfs - at access the underlying filesystems
smb/server - for file service
They should be used by nfsd (more than just the exportfs path) and
cachefs but aren't.
It would help if the documentation didn't claim they should "not be
called by generic code".
Also the path component name is passed as "name" and "len" which are
(confusingly?) separate by the "base". In some cases the len in simply
"strlen" and so passing a qstr using QSTR() would make the calling
clearer.
Other callers do pass separate name and len which are stored in a
struct. Sometimes these are already stored in a qstr, other times it
easily could be.
So this patch changes these three functions to receive a 'struct qstr *',
and improves the documentation.
QSTR_LEN() is added to make it easy to pass a QSTR containing a known
len.
[brauner@kernel.org: take a struct qstr pointer]
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://lore.kernel.org/r/20250319031545.2999807-2-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
The locking around the ->get_parent() call brings no value.
We are locking a child which is only used to find an inode and thence the
parent inode number. All further activity involves the parent inode
which may have several children so locking one child cannot protect the
parent in any useful way.
The filesystem must already ensure that only one 'struct inode' exists
for a given inode, and will call d_obtain_alias() which contains the
required locking to ensure only one dentry will be attached to that
inode.
So remove the unnecessary locking.
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://lore.kernel.org/r/174190497326.9342.9313518146512158587@noble.neil.brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
We would like to use the high 16bit of the handle_type field to encode
file handle traits, such as "connectable".
In preparation for this change, make sure that filesystems do not return
a handle_type value with upper bits set and that the open_by_handle_at(2)
syscall rejects these handle types.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/r/20241011090023.655623-2-amir73il@gmail.com
Fixes: 570df4e9c2 ("ceph: snapshot nfs re-export")
Acked-by:
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
A current limitation of open_by_handle_at() is that it's currently not possible
to use it from within containers at all because we require CAP_DAC_READ_SEARCH
in the initial namespace. That's unfortunate because there are scenarios where
using open_by_handle_at() from within containers.
Two examples:
(1) cgroupfs allows to encode cgroups to file handles and reopen them with
open_by_handle_at().
(2) Fanotify allows placing filesystem watches they currently aren't usable in
containers because the returned file handles cannot be used.
Here's a proposal for relaxing the permission check for open_by_handle_at().
(1) Opening file handles when the caller has privileges over the filesystem
(1.1) The caller has an unobstructed view of the filesystem.
(1.2) The caller has permissions to follow a path to the file handle.
This doesn't address the problem of opening a file handle when only a portion
of a filesystem is exposed as is common in containers by e.g., bind-mounting a
subtree. The proposal to solve this use-case is:
(2) Opening file handles when the caller has privileges over a subtree
(2.1) The caller is able to reach the file from the provided mount fd.
(2.2) The caller has permissions to construct an unobstructed path to the
file handle.
(2.3) The caller has permissions to follow a path to the file handle.
The relaxed permission checks are currently restricted to directory file
handles which are what both cgroupfs and fanotify need. Handling disconnected
non-directory file handles would lead to a potentially non-deterministic api.
If a disconnected non-directory file handle is provided we may fail to decode
a valid path that we could use for permission checking. That in itself isn't a
problem as we would just return EACCES in that case. However, confusion may
arise if a non-disconnected dentry ends up in the cache later and those opening
the file handle would suddenly succeed.
* It's potentially possible to use timing information (side-channel) to infer
whether a given inode exists. I don't think that's particularly
problematic. Thanks to Jann for bringing this to my attention.
* An unrelated note (IOW, these are thoughts that apply to
open_by_handle_at() generically and are unrelated to the changes here):
Jann pointed out that we should verify whether deleted files could
potentially be reopened through open_by_handle_at(). I don't think that's
possible though.
Another potential thing to check is whether open_by_handle_at() could be
abused to open internal stuff like memfds or gpu stuff. I don't think so
but I haven't had the time to completely verify this.
This dates back to discussions Amir and I had quite some time ago and thanks to
him for providing a lot of details around the export code and related patches!
Link: https://lore.kernel.org/r/20240524-vfs-open_by_handle_at-v1-1-3d4b7d22736b@kernel.org
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
The fallback implementation for the get_name export operation uses
readdir() to try to match the inode number to a filename. That filename
is then used together with lookup_one() to produce a dentry.
A problem arises when we match the '.' or '..' entries, since that
causes lookup_one() to fail. This has sometimes been seen to occur for
filesystems that violate POSIX requirements around uniqueness of inode
numbers, something that is common for snapshot directories.
This patch just ensures that we skip '.' and '..' rather than allowing a
match.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/linux-nfs/CAOQ4uxiOZobN76OKB-VBNXWeFKVwLW_eK5QtthGyYzWU9mjb7Q@mail.gmail.com/
Acked-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
AT_HANDLE_FID was added as an API for name_to_handle_at() that request
the encoding of a file id, which is not intended to be decoded.
This file id is used by fanotify to describe objects in events.
So far, overlayfs is the only filesystem that supports encoding
non-decodeable file ids, by providing export_operations with an
->encode_fh() method and without a ->decode_fh() method.
Add support for encoding non-decodeable file ids to all the filesystems
that do not provide export_operations, by encoding a file id of type
FILEID_INO64_GEN from { i_ino, i_generation }.
A filesystem may that does not support NFS export, can opt-out of
encoding non-decodeable file ids for fanotify by defining an empty
export_operations struct (i.e. with a NULL ->encode_fh() method).
This allows the use of fanotify events with file ids on filesystems
like 9p which do not support NFS export to bring fanotify in feature
parity with inotify on those filesystems.
Note that fanotify also requires that the filesystems report a non-null
fsid. Currently, many simple filesystems that have support for inotify
(e.g. debugfs, tracefs, sysfs) report a null fsid, so can still not be
used with fanotify in file id reporting mode.
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/r/20231023180801.2953446-5-amir73il@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
Rename the default helper for encoding FILEID_INO32_GEN* file handles to
generic_encode_ino32_fh() and convert the filesystems that used the
default implementation to use the generic helper explicitly.
After this change, exportfs_encode_inode_fh() no longer has a default
implementation to encode FILEID_INO32_GEN* file handles.
This is a step towards allowing filesystems to encode non-decodeable
file handles for fanotify without having to implement any
export_operations.
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/r/20231023180801.2953446-3-amir73il@gmail.com
Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Remove kernel-doc warning in exportfs:
fs/exportfs/expfs.c:395: warning: Function parameter or member 'parent'
not described in 'exportfs_encode_inode_fh'
Signed-off-by: Zhu Wang <wangzhu9@huawei.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
All users now just use '->iterate_shared()', which only takes the
directory inode lock for reading.
Filesystems that never got convered to shared mode now instead use a
wrapper that drops the lock, re-takes it in write mode, calls the old
function, and then downgrades the lock back to read mode.
This way the VFS layer and other callers no longer need to care about
filesystems that never got converted to the modern era.
The filesystems that use the new wrapper are ceph, coda, exfat, jfs,
ntfs, ocfs2, overlayfs, and vboxsf.
Honestly, several of them look like they really could just iterate their
directories in shared mode and skip the wrapper entirely, but the point
of this change is to not change semantics or fix filesystems that
haven't been fixed in the last 7+ years, but to finally get rid of the
dual iterators.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
So far, all callers of exportfs_encode_inode_fh(), except for fsnotify's
show_mark_fhandle(), check that filesystem can decode file handles, but
we would like to add more callers that do not require a file handle that
can be decoded.
Introduce a flag to explicitly request a file handle that may not to be
decoded later and a wrapper exportfs_encode_fid() that sets this flag
and convert show_mark_fhandle() to use the new wrapper.
This will be used to allow adding fanotify support to filesystems that
do not support NFS export.
Acked-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Message-Id: <20230502124817.3070545-3-amir73il@gmail.com>
Convert the bool connectable arguemnt into a bit flags argument and
define the EXPORT_FS_CONNECTABLE flag as a requested property of the
file handle.
We are going to add a flag for requesting non-decodeable file handles.
Acked-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Message-Id: <20230502124817.3070545-2-amir73il@gmail.com>
Convert to struct mnt_idmap.
Last cycle we merged the necessary infrastructure in
256c8aed2b ("fs: introduce dedicated idmap type for mounts").
This is just the conversion to struct mnt_idmap.
Currently we still pass around the plain namespace that was attached to a
mount. This is in general pretty convenient but it makes it easy to
conflate namespaces that are relevant on the filesystem with namespaces
that are relevent on the mount level. Especially for non-vfs developers
without detailed knowledge in this area this can be a potential source for
bugs.
Once the conversion to struct mnt_idmap is done all helpers down to the
really low-level helpers will take a struct mnt_idmap argument instead of
two namespace arguments. This way it becomes impossible to conflate the two
eliminating the possibility of any bugs. All of the vfs and all filesystems
only operate on struct mnt_idmap.
Acked-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
expfs.c has a bunch of dprintk statements which are unusable due to:
#define dprintk(fmt, args...) do{}while(0)
Use pr_debug so that they can be enabled dynamically.
Also make some minor changes to the debug statements to fix some
incorrect types, and remove __func__ which can be handled by dynamic
debug separately.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Reviewed-by: NeilBrown <neilb@suse.de>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
filldir_t instances (directory iterators callbacks) used to return 0 for
"OK, keep going" or -E... for "stop". Note that it's *NOT* how the
error values are reported - the rules for those are callback-dependent
and ->iterate{,_shared}() instances only care about zero vs. non-zero
(look at emit_dir() and friends).
So let's just return bool ("should we keep going?") - it's less confusing
that way. The choice between "true means keep going" and "true means
stop" is bikesheddable; we have two groups of callbacks -
do something for everything in directory, until we run into problem
and
find an entry in directory and do something to it.
The former tended to use 0/-E... conventions - -E<something> on failure.
The latter tended to use 0/1, 1 being "stop, we are done".
The callers treated anything non-zero as "stop", ignoring which
non-zero value did they get.
"true means stop" would be more natural for the second group; "true
means keep going" - for the first one. I tried both variants and
the things like
if allocation failed
something = -ENOMEM;
return true;
just looked unnatural and asking for trouble.
[folded suggestion from Matthew Wilcox <willy@infradead.org>]
Acked-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
