| Age | Commit message (Collapse) | Author |
|---|
|
Clang warns (or errors with CONFIG_WERROR=y):
fs/ntfs3/inode.c:1021:6: error: variable 'folio' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
1021 | if (is_resident(ni)) {
| ^~~~~~~~~~~~~~~
fs/ntfs3/inode.c:1024:48: note: uninitialized use occurs here
1024 | while ((folio = writeback_iter(mapping, wbc, folio, &err)))
| ^~~~~
folio should be initialized to NULL for the first iteration of
writeback_iter() to start the loop properly. Restore the NULL
initialization of folio that was lost in the recent iomap conversion to
clear up the warning.
Fixes: 099ef9a ("fs/ntfs3: implement iomap-based file operations")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/oe-kbuild-all/202601010644.FIhOXy6Y-lkp@intel.com/
Closes: https://lore.kernel.org/r/202601010513.axd56bks-lkp@intel.com/
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
[almaz.alexandrovich@paragon-software.com: added a few more tags]
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
|
|
When a filesystem is frozen, quotactl_block() enters a retry loop
waiting for the filesystem to thaw. It acquires s_umount, checks the
freeze state, drops s_umount and uses sb_start_write() - sb_end_write()
pair to wait for the unfreeze.
However, this retry loop can trigger a livelock issue, specifically on
kernels with preemption disabled.
The mechanism is as follows:
1. freeze_super() sets SB_FREEZE_WRITE and calls sb_wait_write().
2. sb_wait_write() calls percpu_down_write(), which initiates
synchronize_rcu().
3. Simultaneously, quotactl_block() spins in its retry loop, immediately
executing the sb_start_write() - sb_end_write() pair.
4. Because the kernel is non-preemptible and the loop contains no
scheduling points, quotactl_block() never yields the CPU. This
prevents that CPU from reaching an RCU quiescent state.
5. synchronize_rcu() in the freezer thread waits indefinitely for the
quotactl_block() CPU to report a quiescent state.
6. quotactl_block() spins indefinitely waiting for the freezer to
advance, which it cannot do as it is blocked on the RCU sync.
This results in a hang of the freezer process and 100% CPU usage by the
quota process.
While this can occur intermittently on multi-core systems, it is
reliably reproducing on a node with the following script, running both
the freezer and the quota toggle on the same CPU:
# mkfs.ext4 -O quota /dev/sda 2g && mkdir a_mount
# mount /dev/sda -o quota,usrquota,grpquota a_mount
# taskset -c 3 bash -c "while true; do xfs_freeze -f a_mount; \
xfs_freeze -u a_mount; done" &
# taskset -c 3 bash -c "while true; do quotaon a_mount; \
quotaoff a_mount; done" &
Adding cond_resched() to the retry loop fixes the issue. It acts as an
RCU quiescent state, allowing synchronize_rcu() in percpu_down_write()
to complete.
Fixes: 576215cffdef ("fs: Drop wait_unfrozen wait queue")
Signed-off-by: Abhishek Bapat <abhishekbapat@google.com>
Link: https://patch.msgid.link/20260115213103.1089129-1-abhishekbapat@google.com
Signed-off-by: Jan Kara <jack@suse.cz>
|
|
With a mix of include dependencies, the compiler warns that:
fs/fuse/dir.c:35:9: warning: ?HASH_BITS? redefined
35 | #define HASH_BITS 5
| ^~~~~~~~~
In file included from ./include/linux/io_uring_types.h:5,
from ./include/linux/bpf.h:34,
from ./include/linux/security.h:35,
from ./include/linux/fs_context.h:14,
from fs/fuse/dir.c:13:
./include/linux/hashtable.h:28:9: note: this is the location of the previous definition
28 | #define HASH_BITS(name) ilog2(HASH_SIZE(name))
| ^~~~~~~~~
fs/fuse/dir.c:36:9: warning: ?HASH_SIZE? redefined
36 | #define HASH_SIZE (1 << HASH_BITS)
| ^~~~~~~~~
./include/linux/hashtable.h:27:9: note: this is the location of the previous definition
27 | #define HASH_SIZE(name) (ARRAY_SIZE(name))
| ^~~~~~~~~
Hence rename the HASH_SIZE/HASH_BITS in fuse, by prefixing them with
FUSE_ instead.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Link: https://patch.msgid.link/195c9525-281c-4302-9549-f3d9259416c6@kernel.dk
Acked-by: Miklos Szeredi <mszeredi@redhat.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Without exception all caller do that. So move the allocation into the
helper.
This reduces boilerplate and removes unnecessary error checking.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Link: https://patch.msgid.link/20260115122341.556026-1-mszeredi@redhat.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
This reverts commit 196c81fdd438f7ac429d5639090a9816abb9760a.
Original patch may cause below deadlock, revert it.
write remount
- write_begin
- lock_page --- lock A
- prepare_write_begin
- f2fs_map_lock
- f2fs_enable_checkpoint
- down_write(cp_enable_rwsem) --- lock B
- sync_inode_sb
- writepages
- lock_page --- lock A
- down_read(cp_enable_rwsem) --- lock A
Cc: stable@kernel.org
Fixes: 196c81fdd438 ("f2fs: block cache/dio write during f2fs_enable_checkpoint()")
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
Pull NFS client fixes from Trond Myklebust:
- Fix another deadlock involving nfs_release_folio()
- localio:
- Stop I/O upon hitting a fatal error
- Deal with page offsets that are > PAGE_SIZE
- Fix size read races in truncate, fallocate and copy offload
- Several bugfixes for the NFSv4.x directory delegation client code
- pNFS:
- Fix a deadlock when returning delegations during open
- Fix memory leaks in various error paths
* tag 'nfs-for-6.19-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
NFS: Fix size read races in truncate, fallocate and copy offload
NFS: Don't immediately return directory delegations when disabled
NFS/localio: Deal with page bases that are > PAGE_SIZE
NFS/localio: Stop further I/O upon hitting an error
NFSv4.x: Directory delegations don't require any state recovery
NFSv4: Don't free slots prematurely if requesting a directory delegation
NFSv4: Fix nfs_clear_verifier_delegated() for delegated directories
NFS: Fix directory delegation verifier checks
pnfs/blocklayout: Fix memory leak in bl_parse_scsi()
pnfs/flexfiles: Fix memory leak in nfs4_ff_alloc_deviceid_node()
NFS: Fix a deadlock involving nfs_release_folio()
pNFS: Fix a deadlock when returning a delegation during open()
|
|
If the pre-operation file size is read before locking the inode and
quiescing O_DIRECT writes, then nfs_truncate_last_folio() might end up
overwriting valid file data.
Fixes: b1817b18ff20 ("NFS: Protect against 'eof page pollution'")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
|
|
Extract the repeated pattern of duplicating attribute and updating
OpenAttr into a helper function to reduce code duplication and improve
maintainability.
Signed-off-by: Baolin Liu <liubaolin@kylinos.cn>
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
|
|
When processing valid within the range [valid : pos), if valid cannot
be retrieved correctly, for example, if the retrieved valid value is
always the same, this can trigger a potential infinite loop, similar
to the hung problem reported by syzbot [1].
Adding a check for the valid value within the loop body, and terminating
the loop and returning -EINVAL if the value is the same as the current
value, can prevent this.
[1]
INFO: task syz.4.21:6056 blocked for more than 143 seconds.
Call Trace:
rwbase_write_lock+0x14f/0x750 kernel/locking/rwbase_rt.c:244
inode_lock include/linux/fs.h:1027 [inline]
ntfs_file_write_iter+0xe6/0x870 fs/ntfs3/file.c:1284
Fixes: 4342306f0f0d ("fs/ntfs3: Add file operations and implementation")
Reported-by: syzbot+bcf9e1868c1a0c7e04f1@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=bcf9e1868c1a0c7e04f1
Signed-off-by: Edward Adam Davis <eadavis@qq.com>
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
|
|
Fix below sparse warnings:
fs/ntfs3/inode.c:972:34: sparse: sparse: symbol 'ntfs_writeback_ops' was not declared.
Should it be static?
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202601061424.nbKLNwC5-lkp@intel.com/
Signed-off-by: sunliming <sunliming@kylinos.cn>
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
|
|
Syzbot reported a circular locking dependency between wnd->rw_lock
(sbi->used.bitmap) and ni->file.run_lock.
The deadlock scenario:
1. ntfs_extend_mft() takes ni->file.run_lock then wnd->rw_lock.
2. run_unpack_ex() takes wnd->rw_lock then tries to acquire
ni->file.run_lock inside ntfs_refresh_zone().
This creates an AB-BA deadlock.
Fix this by using down_read_trylock() instead of down_read() when
acquiring run_lock in run_unpack_ex(). If the lock is contended,
skip ntfs_refresh_zone() - the MFT zone will be refreshed on the
next MFT operation. This breaks the circular dependency since we
never block waiting for run_lock while holding wnd->rw_lock.
Reported-by: syzbot+d27edf9f96ae85939222@syzkaller.appspotmail.com
Tested-by: syzbot+d27edf9f96ae85939222@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=d27edf9f96ae85939222
Signed-off-by: Szymon Wilczek <swilczek.lx@gmail.com>
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
|
|
Cross-merge BPF and other fixes after downstream PR.
No conflicts.
Adjacent:
Auto-merging MAINTAINERS
Auto-merging Makefile
Auto-merging kernel/bpf/verifier.c
Auto-merging kernel/sched/ext.c
Auto-merging mm/memcontrol.c
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Add the __counted_by() compiler attribute to the flexible array member
'data' to improve access bounds-checking via CONFIG_UBSAN_BOUNDS and
CONFIG_FORTIFY_SOURCE.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Acked-by: Tyler Hicks <code@tyhicks.com>
Link: https://patch.msgid.link/20260112115314.739612-2-thorsten.blum@linux.dev
Signed-off-by: Kees Cook <kees@kernel.org>
|
|
Remove the newly introduced zoned statistics from sysfs, as sysfs can
only show a single page this will truncate the output on a busy
filesystem.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
There is no longer the need to use btrfs_may_create(), which was a copy
of the VFS private function may_create(), since now that functionality
is exported by the VFS as a function named may_create_dentry(). So change
btrfs_mksubvol() to use the VFS function and remove btrfs_may_create().
Note that the btrfs copy was missing an audit_inode_child() call that we
have in the VFS function. This only reinforces the need to use a common
function, as it's very easy for the btrfs copy to get out of sync and
therefore a maintenance burden.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Link: https://patch.msgid.link/adf8c802c77ec1c855ea9fe12491120ccc29a294.1768307858.git.fdmanana@suse.com
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
There is no longer the need to use btrfs_may_delete(), which was a copy
of the VFS private function may_delete(), since now that functionality
is exported by the VFS as a function named may_delete_dentry(). In fact
our local copy of may_delete() lacks an update that happened to that
function which is point number 7 in that function's comment:
"7. If the victim has an unknown uid or gid we can't change the inode."
which corresponds to this code:
/* Inode writeback is not safe when the uid or gid are invalid. */
if (!vfsuid_valid(i_uid_into_vfsuid(idmap, inode)) ||
!vfsgid_valid(i_gid_into_vfsgid(idmap, inode)))
return -EOVERFLOW;
As long as we keep a separate copy, duplicating code, we are also prone
to updates to the VFS being missed in our local copy.
So change btrfs_ioctl_snap_destroy() to use the VFS function and remove
btrfs_may_delete().
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Link: https://patch.msgid.link/46b13dc5c957deb72a7f085916757a20878a8e73.1768307858.git.fdmanana@suse.com
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
For many years btrfs as been using a copy of may_create() in
fs/btrfs/ioctl.c:btrfs_may_create(). Everytime may_create() is updated we
need to update the btrfs copy, and this is a maintenance burden. Currently
there are minor differences between both because the btrfs side lacks
updates done in may_create().
Export may_create() so that btrfs can use it and with the less generic
name may_create_dentry().
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Link: https://patch.msgid.link/ce5174bca079f4cdcbb8dd145f0924feb1f227cd.1768307858.git.fdmanana@suse.com
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
For many years btrfs as been using a copy of may_delete() in
fs/btrfs/ioctl.c:btrfs_may_delete(). Everytime may_delete() is updated we
need to update the btrfs copy, and this is a maintenance burden. Currently
there are minor differences between both because the btrfs side lacks
updates done in may_delete().
Export may_delete() so that btrfs can use it and with the less generic
name may_delete_dentry(). While at it change the calls in vfs_rmdir() to
pass a boolean literal instead of 1 and 0 as the last argument since the
argument has a bool type.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Link: https://patch.msgid.link/e09128fd53f01b19d0a58f0e7d24739f79f47f6d.1768307858.git.fdmanana@suse.com
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The dirtytime_work is a background housekeeping task that flushes dirty
inodes, using round_jiffies_relative() will allow kernel to batch this
work with other aligned system tasks, reducing power consumption.
Signed-off-by: Zhao Mengmeng <zhaomengmeng@kylinos.cn>
Link: https://patch.msgid.link/20260113082614.231580-1-zhaomzhao@126.com
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
__iomap_get_folio needs to wait for writeback to finish if the file
requires folios to be stable for writes. For the regular path this is
taken care of by __filemap_get_folio, but for the newly added batch
lookup it has to be done manually.
This fixes xfs/131 failures when running on PI-capable hardware.
Fixes: 395ed1ef0012 ("iomap: optional zero range dirty folio processing")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://patch.msgid.link/20260113153943.3323869-1-hch@lst.de
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
This is the only routine which instead skipped instead of waiting.
The current behavior is arguably a bug as it results in a corner case
where the inode hash can have *two* matching inodes, one of which is on
its way out.
Ironing out this difference is an incremental step towards sanitizing
the API.
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Link: https://patch.msgid.link/20260114094717.236202-1-mjguzik@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Use get_kernel_nofault() to safely access inode and related structures
(superblock, file_system_type) to avoid crashing when the inode pointer
is invalid. This allows the same pattern as dump_mapping().
Note: The original access method for i_state and i_count is preserved,
as get_kernel_nofault() is unnecessary once the inode structure is
verified accessible.
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Yuto Ohnuki <ytohnuki@amazon.com>
Link: https://patch.msgid.link/20260112181443.81286-1-ytohnuki@amazon.com
Reviewed-by: Mateusz Guzik <mjguzik@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The init_fs symbol is defined in <linux/init_task.h> but was
not included in fs/fs_struct.c so fix by adding the include.
Fixes the following sparse warning:
fs/fs_struct.c:150:18: warning: symbol 'init_fs' was not declared. Should it be static?
Fixes: 3e93cd671813e ("Take fs_struct handling to new file")
Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Link: https://patch.msgid.link/20260108115856.238027-1-ben.dooks@codethink.co.uk
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The comment above file_ref_init() says:
"We're SLAB_TYPESAFE_BY_RCU so initialize f_ref last."
but file_set_fsnotify_mode() was added after file_ref_init().
Move it right after setting f_mode, where it makes more sense.
Fixes: 711f9b8fbe4f4 ("fsnotify: disable pre-content and permission events by default")
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://patch.msgid.link/20260109211536.3565697-1-amir73il@gmail.com
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Introduce `struct erofs_iomap_iter_ctx` to hold both `struct page *`
and `void *base`, avoiding bogus use of `kmap_to_page()` in
`erofs_iomap_end()`.
With this change, fiemap and bmap no longer need to read inline data.
Additionally, the upcoming page cache sharing mechanism requires
passing the backing inode pointer to `erofs_iomap_{begin,end}()`, as
I/O accesses must apply to backing inodes rather than anon inodes.
Signed-off-by: Hongbo Li <lihongbo22@huawei.com>
Link: https://patch.msgid.link/20260109102856.598531-3-lihongbo22@huawei.com
Reviewed-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
It's useful to get filesystem-specific information using the
existing private field in the @iomap_iter passed to iomap_{begin,end}
for advanced usage for iomap buffered reads, which is much like the
current iomap DIO.
For example, EROFS needs it to:
- implement an efficient page cache sharing feature, since iomap
needs to apply to anon inode page cache but we'd like to get the
backing inode/fs instead, so filesystem-specific private data is
needed to keep such information;
- pass in both struct page * and void * for inline data to avoid
kmap_to_page() usage (which is bogus).
Signed-off-by: Hongbo Li <lihongbo22@huawei.com>
Link: https://patch.msgid.link/20260109102856.598531-2-lihongbo22@huawei.com
Reviewed-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Remove the "nullfs_rootfs" boot parameter and simply always use nullfs.
The mutable rootfs will be mounted on top of it. Systems that don't use
pivot_root() to pivot away from the real rootfs will have an additional
mount stick around but that shouldn't be a problem at all. If it is
we'll rever this commit.
This also simplifies the boot process and removes the need for the
traditional switch_root workarounds.
Suggested-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
All linux_binprm instances come from alloc_bprm() and are unconditionally
destroyed by free_bprm() in the end of the same scope. IOW, CLASS()
machinery is a decent fit for those.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
... and convert its callers to CLASS(filename...)
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
All of them are wrappers for do_execveat_common() and each has
exactly one caller. The only difference is in the way they are
constructing argv/envp arguments for do_execveat_common() and
that's easy to do with less boilerplate.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
alloc_bprm() starts with do_open_execat() and it will do the right
thing if given ERR_PTR() for name. Allows to drop such checks in
its callers...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
It accepts ERR_PTR() for name and does the right thing in that case.
That allows to simplify the logics in callers, making them trivial
to switch to CLASS(filename).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
do_file_open() will do the right thing when given ERR_PTR() as name...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
"filp" thing never made sense; seeing that there are exactly 4 callers
in the entire tree (and it's neither exported nor even declared in
linux/*/*.h), there's no point keeping that ugliness.
FWIW, the 'filp' thing did originate in OSD&I; for some reason Tanenbaum
decided to call the object representing an opened file 'struct filp',
the last letter standing for 'position'. In all Unices, Linux included,
the corresponding object had always been 'struct file'...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
The rest of the set_nameidata() callers treat IS_ERR(pathname) as
"bail out immediately with PTR_ERR(pathname) as error". Makes
life simpler for callers; do_filp_open() is the only exception
and its callers would also benefit from such calling conventions
change.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
no need to check in the caller
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
no need to check in the caller
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
no need to check it in the caller
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
no need to check it in the caller
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
no need to check it in the caller
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
... or visible outside of audit, really. Note that references
held in delayed_filename always have refcount 1, and from the
moment of complete_getname() or equivalent point in getname...()
there won't be any references to struct filename instance left
in places visible to other threads.
Acked-by: Paul Moore <paul@paul-moore.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
There are two filename-related problems in io_uring and its
interplay with audit.
Filenames are imported when request is submitted and used when
it is processed. Unfortunately, the latter may very well
happen in a different thread. In that case the reference to
filename is put into the wrong audit_context - that of submitting
thread, not the processing one. Audit logics is called by
the latter, and it really wants to be able to find the names
in audit_context current (== processing) thread.
Another related problem is the headache with refcounts -
normally all references to given struct filename are visible
only to one thread (the one that uses that struct filename).
io_uring violates that - an extra reference is stashed in
audit_context of submitter. It gets dropped when submitter
returns to userland, which can happen simultaneously with
processing thread deciding to drop the reference it got.
We paper over that by making refcount atomic, but that means
pointless headache for everyone.
Solution: the notion of partially imported filenames. Namely,
already copied from userland, but *not* exposed to audit yet.
io_uring can create that in submitter thread, and complete the
import (obtaining the usual reference to struct filename) in
processing thread.
Object: struct delayed_filename.
Primitives for working with it:
delayed_getname(&delayed_filename, user_string) - copies the name from
userland, returning 0 and stashing the address of (still incomplete)
struct filename in delayed_filename on success and returning -E... on
error.
delayed_getname_uflags(&delayed_filename, user_string, atflags) -
similar, in the same relation to delayed_getname() as getname_uflags()
is to getname()
complete_getname(&delayed_filename) - completes the import of filename
stashed in delayed_filename and returns struct filename to caller,
emptying delayed_filename.
CLASS(filename_complete_delayed, name)(&delayed_filename) - variant of
CLASS(filename) with complete_getname() for constructor.
dismiss_delayed_filename(&delayed_filename) - destructor; drops whatever
might be stashed in delayed_filename, emptying it.
putname_to_delayed(&delayed_filename, name) - if name is shared, stashes
its copy into delayed_filename and drops the reference to name, otherwise
stashes the name itself in there.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
s/names_cachep/names_cache/ for consistency with dentry cache.
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Always allocate struct filename from names_cachep, long name or short;
short names would be embedded into struct filename. Longer ones do
not cannibalize the original struct filename - put them into kmalloc'ed
buffers (PATH_MAX-sized for import from userland, strlen() + 1 - for
ones originating kernel-side, where we know the length beforehand).
Cutoff length for short names is chosen so that struct filename would be
192 bytes long - that's both a multiple of 64 and large enough to cover
the majority of real-world uses.
Simplifies logics in getname()/putname() and friends.
[fixed an embarrassing braino in EMBEDDED_NAME_MAX, first reported by
Dan Carpenter]
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Instances of struct filename come from names_cachep (via
__getname()). That is done by getname_flags() and getname_kernel()
and these two are the main callers of __getname(). However, there are
other callers that simply want to allocate PATH_MAX bytes for uses that
have nothing to do with struct filename.
We want saner allocation rules for long pathnames, so that struct
filename would *always* come from names_cachep, with the out-of-line
pathname getting kmalloc'ed. For that we need to be able to change the
size of objects allocated by getname_flags()/getname_kernel().
That requires the rest of __getname() users to stop using
names_cachep; we could explicitly switch all of those to kmalloc(),
but that would cause quite a bit of noise. So the plan is to switch
getname_...() to new helpers and turn __getname() into a wrapper for
kmalloc(). Remaining __getname() users could be converted to explicit
kmalloc() at leisure, hopefully along with figuring out what size do
they really want - PATH_MAX is an overkill for some of them, used out
of laziness ("we have a convenient helper that does 4K allocations and
that's large enough, let's use it").
As a side benefit, names_cachep is no longer used outside
of fs/namei.c, so we can move it there and be done with that.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Take the "long name" case into a helper (getname_long()). In
case of failure have the caller deal with freeing the original
struct filename.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
In case of long name don't reread what we'd already copied.
memmove() it instead. That avoids the possibility of ending
up with empty name there and the need to look at the flags
on the slow path.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
... so don't use __getname() there. Switch it (and ntfs_d_hash(), while
we are at it) to kmalloc(PATH_MAX, GFP_NOWAIT). Yes, ntfs_d_hash()
almost certainly can do with smaller allocations, but let ntfs folks
deal with that - keep the allocation size as-is for now.
Stop abusing names_cachep in ntfs, period - various uses of that thing
in there have nothing to do with pathnames; just use k[mz]alloc() and
be done with that. For now let's keep sizes as-in, but AFAICS none of
the users actually want PATH_MAX.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
