The inode reclaiming process(See function prune_icache_sb) collects all
reclaimable inodes and mark them with I_FREEING flag at first, at that
time, other processes will be stuck if they try getting these inodes
(See function find_inode_fast), then the reclaiming process destroy the
inodes by function dispose_list(). Some filesystems(eg. ext4 with
ea_inode feature, ubifs with xattr) may do inode lookup in the inode
evicting callback function, if the inode lookup is operated under the
inode lru traversing context, deadlock problems may happen.
Case 1: In function ext4_evict_inode(), the ea inode lookup could happen
if ea_inode feature is enabled, the lookup process will be stuck
under the evicting context like this:
1. File A has inode i_reg and an ea inode i_ea
2. getfattr(A, xattr_buf) // i_ea is added into lru // lru->i_ea
3. Then, following three processes running like this:
PA PB
echo 2 > /proc/sys/vm/drop_caches
shrink_slab
prune_dcache_sb
// i_reg is added into lru, lru->i_ea->i_reg
prune_icache_sb
list_lru_walk_one
inode_lru_isolate
i_ea->i_state |= I_FREEING // set inode state
inode_lru_isolate
__iget(i_reg)
spin_unlock(&i_reg->i_lock)
spin_unlock(lru_lock)
rm file A
i_reg->nlink = 0
iput(i_reg) // i_reg->nlink is 0, do evict
ext4_evict_inode
ext4_xattr_delete_inode
ext4_xattr_inode_dec_ref_all
ext4_xattr_inode_iget
ext4_iget(i_ea->i_ino)
iget_locked
find_inode_fast
__wait_on_freeing_inode(i_ea) ----→ AA deadlock
dispose_list // cannot be executed by prune_icache_sb
wake_up_bit(&i_ea->i_state)
Case 2: In deleted inode writing function ubifs_jnl_write_inode(), file
deleting process holds BASEHD's wbuf->io_mutex while getting the
xattr inode, which could race with inode reclaiming process(The
reclaiming process could try locking BASEHD's wbuf->io_mutex in
inode evicting function), then an ABBA deadlock problem would
happen as following:
1. File A has inode ia and a xattr(with inode ixa), regular file B has
inode ib and a xattr.
2. getfattr(A, xattr_buf) // ixa is added into lru // lru->ixa
3. Then, following three processes running like this:
PA PB PC
echo 2 > /proc/sys/vm/drop_caches
shrink_slab
prune_dcache_sb
// ib and ia are added into lru, lru->ixa->ib->ia
prune_icache_sb
list_lru_walk_one
inode_lru_isolate
ixa->i_state |= I_FREEING // set inode state
inode_lru_isolate
__iget(ib)
spin_unlock(&ib->i_lock)
spin_unlock(lru_lock)
rm file B
ib->nlink = 0
rm file A
iput(ia)
ubifs_evict_inode(ia)
ubifs_jnl_delete_inode(ia)
ubifs_jnl_write_inode(ia)
make_reservation(BASEHD) // Lock wbuf->io_mutex
ubifs_iget(ixa->i_ino)
iget_locked
find_inode_fast
__wait_on_freeing_inode(ixa)
| iput(ib) // ib->nlink is 0, do evict
| ubifs_evict_inode
| ubifs_jnl_delete_inode(ib)
↓ ubifs_jnl_write_inode
ABBA deadlock ←-----make_reservation(BASEHD)
dispose_list // cannot be executed by prune_icache_sb
wake_up_bit(&ixa->i_state)
Fix the possible deadlock by using new inode state flag I_LRU_ISOLATING
to pin the inode in memory while inode_lru_isolate() reclaims its pages
instead of using ordinary inode reference. This way inode deletion
cannot be triggered from inode_lru_isolate() thus avoiding the deadlock.
evict() is made to wait for I_LRU_ISOLATING to be cleared before
proceeding with inode cleanup.
Link: https://lore.kernel.org/all/37c29c42-7685-d1f0-067d-63582ffac405@huaweicloud.com/
Link: https://bugzilla.kernel.org/show_bug.cgi?id=219022
Fixes: e50e5129f384 ("ext4: xattr-in-inode support")
Fixes: 7959cf3a7506 ("ubifs: journal: Handle xattrs like files")
Cc: stable@vger.kernel.org
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Link: https://lore.kernel.org/r/20240809031628.1069873-1-chengzhihao@huaweicloud.com
Reviewed-by: Jan Kara <jack@suse.cz>
Suggested-by: Jan Kara <jack@suse.cz>
Suggested-by: Mateusz Guzik <mjguzik@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
const qualify the struct ctl_table argument in the proc_handler function
signatures. This is a prerequisite to moving the static ctl_table
structs into .rodata data which will ensure that proc_handler function
pointers cannot be modified.
This patch has been generated by the following coccinelle script:
```
virtual patch
@r1@
identifier ctl, write, buffer, lenp, ppos;
identifier func !~ "appldata_(timer|interval)_handler|sched_(rt|rr)_handler|rds_tcp_skbuf_handler|proc_sctp_do_(hmac_alg|rto_min|rto_max|udp_port|alpha_beta|auth|probe_interval)";
@@
int func(
- struct ctl_table *ctl
+ const struct ctl_table *ctl
,int write, void *buffer, size_t *lenp, loff_t *ppos);
@r2@
identifier func, ctl, write, buffer, lenp, ppos;
@@
int func(
- struct ctl_table *ctl
+ const struct ctl_table *ctl
,int write, void *buffer, size_t *lenp, loff_t *ppos)
{ ... }
@r3@
identifier func;
@@
int func(
- struct ctl_table *
+ const struct ctl_table *
,int , void *, size_t *, loff_t *);
@r4@
identifier func, ctl;
@@
int func(
- struct ctl_table *ctl
+ const struct ctl_table *ctl
,int , void *, size_t *, loff_t *);
@r5@
identifier func, write, buffer, lenp, ppos;
@@
int func(
- struct ctl_table *
+ const struct ctl_table *
,int write, void *buffer, size_t *lenp, loff_t *ppos);
```
* Code formatting was adjusted in xfs_sysctl.c to comply with code
conventions. The xfs_stats_clear_proc_handler,
xfs_panic_mask_proc_handler and xfs_deprecated_dointvec_minmax where
adjusted.
* The ctl_table argument in proc_watchdog_common was const qualified.
This is called from a proc_handler itself and is calling back into
another proc_handler, making it necessary to change it as part of the
proc_handler migration.
Co-developed-by: Thomas Weißschuh <linux@weissschuh.net>
Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Co-developed-by: Joel Granados <j.granados@samsung.com>
Signed-off-by: Joel Granados <j.granados@samsung.com>
In __wait_on_freeing_inode() we warn in case the inode_hash_lock is held but the inode is unhashed. We then release the inode_lock. So using "locked" as parameter name is confusing. Use is_inode_hash_locked as parameter name instead. Signed-off-by: Christian Brauner <brauner@kernel.org>
Lockless hash lookup can find and lock the inode after it gets the
I_FREEING flag set, at which point it blocks waiting for teardown in
evict() to finish.
However, the flag is still set even after evict() wakes up all waiters.
This results in a race where if the inode lock is taken late enough, it
can happen after both hash removal and wakeups, meaning there is nobody
to wake the racing thread up.
This worked prior to RCU-based lookup because the entire ordeal was
synchronized with the inode hash lock.
Since unhashing requires the inode lock, we can safely check whether it
happened after acquiring it.
Link: https://lore.kernel.org/v9fs/20240717102458.649b60be@kernel.org/
Reported-by: Dominique Martinet <asmadeus@codewreck.org>
Fixes: 7180f8d91fcb ("vfs: add rcu-based find_inode variants for iget ops")
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Link: https://lore.kernel.org/r/20240718151838.611807-1-mjguzik@gmail.com
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Pull vfs inode / dentry updates from Christian Brauner:
"This contains smaller performance improvements to inodes and dentries:
inode:
- Add rcu based inode lookup variants.
They avoid one inode hash lock acquire in the common case thereby
significantly reducing contention. We already support RCU-based
operations but didn't take advantage of them during inode
insertion.
Callers of iget_locked() get the improvement without any code
changes. Callers that need a custom callback can switch to
iget5_locked_rcu() as e.g., did btrfs.
With 20 threads each walking a dedicated 1000 dirs * 1000 files
directory tree to stat(2) on a 32 core + 24GB ram vm:
before: 3.54s user 892.30s system 1966% cpu 45.549 total
after: 3.28s user 738.66s system 1955% cpu 37.932 total (-16.7%)
Long-term we should pick up the effort to introduce more
fine-grained locking and possibly improve on the currently used
hash implementation.
- Start zeroing i_state in inode_init_always() instead of doing it in
individual filesystems.
This allows us to remove an unneeded lock acquire in new_inode()
and not burden individual filesystems with this.
dcache:
- Move d_lockref out of the area used by RCU lookup to avoid
cacheline ping poing because the embedded name is sharing a
cacheline with d_lockref.
- Fix dentry size on 32bit with CONFIG_SMP=y so it does actually end
up with 128 bytes in total"
* tag 'vfs-6.11.inode' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
fs: fix dentry size
vfs: move d_lockref out of the area used by RCU lookup
bcachefs: remove now spurious i_state initialization
xfs: remove now spurious i_state initialization in xfs_inode_alloc
vfs: partially sanitize i_state zeroing on inode creation
xfs: preserve i_state around inode_init_always in xfs_reinit_inode
btrfs: use iget5_locked_rcu
vfs: add rcu-based find_inode variants for iget ops
Export in_group_or_capable() as a VFS helper function. Signed-off-by: Youling Tang <tangyouling@kylinos.cn> Link: https://lore.kernel.org/r/20240620032335.147136-1-youling.tang@linux.dev Signed-off-by: Christian Brauner <brauner@kernel.org>
new_inode used to have the following: spin_lock(&inode_lock); inodes_stat.nr_inodes++; list_add(&inode->i_list, &inode_in_use); list_add(&inode->i_sb_list, &sb->s_inodes); inode->i_ino = ++last_ino; inode->i_state = 0; spin_unlock(&inode_lock); over time things disappeared, got moved around or got replaced (global inode lock with a per-inode lock), eventually this got reduced to: spin_lock(&inode->i_lock); inode->i_state = 0; spin_unlock(&inode->i_lock); But the lock acquire here does not synchronize against anyone. Additionally iget5_locked performs i_state = 0 assignment without any locks to begin with, the two combined look confusing at best. It looks like the current state is a leftover which was not cleaned up. Ideally it would be an invariant that i_state == 0 to begin with, but achieving that would require dealing with all filesystem alloc handlers one by one. In the meantime drop the misleading locking and move i_state zeroing to inode_init_always so that others don't need to deal with it by hand. Signed-off-by: Mateusz Guzik <mjguzik@gmail.com> Link: https://lore.kernel.org/r/20240611120626.513952-3-mjguzik@gmail.com Reviewed-by: Jan Kara <jack@suse.cz> Signed-off-by: Christian Brauner <brauner@kernel.org>
This avoids one inode hash lock acquire in the common case on inode
creation, in effect significantly reducing contention.
On the stock kernel said lock is typically taken twice:
1. once to check if the inode happens to already be present
2. once to add it to the hash
The back-to-back lock/unlock pattern is known to degrade performance
significantly, which is further exacerbated if the hash is heavily
populated (long chains to walk, extending hold time). Arguably hash
sizing and hashing algo need to be revisited, but that's beyond the
scope of this patch.
With the acquire from step 1 eliminated with RCU lookup throughput
increases significantly at the scale of 20 cores (benchmark results at
the bottom).
So happens the hash already supports RCU-based operation, but lookups on
inode insertions didn't take advantage of it.
This of course has its limits as the global lock is still a bottleneck.
There was a patchset posted which introduced fine-grained locking[1] but
it appears staled. Apart from that doubt was expressed whether a
handrolled hash implementation is appropriate to begin with, suggesting
replacement with rhashtables. Nobody committed to carrying [1] across
the finish line or implementing anything better, thus the bandaid below.
iget_locked consumers (notably ext4) get away without any changes
because inode comparison method is built-in.
iget5_locked consumers pass a custom callback. Since removal of locking
adds more problems (inode can be changing) it's not safe to assume all
filesystems happen to cope. Thus iget5_locked_rcu gets added, requiring
manual conversion of interested filesystems.
In order to reduce code duplication find_inode and find_inode_fast grow
an argument indicating whether inode hash lock is held, which is passed
down in case sleeping is necessary. They always rcu_read_lock, which is
redundant but harmless. Doing it conditionally reduces readability for
no real gain that I can see. RCU-alike restrictions were already put on
callbacks due to the hash spinlock being held.
Benchmarking:
There is a real cache-busting workload scanning millions of files in
parallel (it's a backup appliance), where the initial lookup is
guaranteed to fail resulting in the two lock acquires on stock kernel
(and one with the patch at hand).
Implemented below is a synthetic benchmark providing the same behavior.
[I shall note the workload is not running on Linux, instead it was
causing trouble elsewhere. Benchmark below was used while addressing
said problems and was found to adequately represent the real workload.]
Total real time fluctuates by 1-2s.
With 20 threads each walking a dedicated 1000 dirs * 1000 files
directory tree to stat(2) on a 32 core + 24GB RAM vm:
ext4 (needed mkfs.ext4 -N 24000000):
before: 3.77s user 890.90s system 1939% cpu 46.118 total
after: 3.24s user 397.73s system 1858% cpu 21.581 total (-53%)
That's 20 million files to visit, while the machine can only cache about
15 million at a time (obtained from ext4_inode_cache object count in
/proc/slabinfo). Since each terminal inode is only visited once per run
this amounts to 0% hit ratio for the dentry cache and the hash table
(there are however hits for the intermediate directories).
On repeated runs the kernel caches the last ~15 mln, meaning there is ~5
mln of uncached inodes which are going to be visited first, evicting the
previously cached state as it happens.
Lack of hits can be trivially verified with bpftrace, like so:
bpftrace -e 'kretprobe:find_inode_fast { @[kstack(), retval != 0] = count(); }'\
-c "/bin/sh walktrees /testfs 20"
Best ran more than once.
Expected results after "warmup":
[snip]
@[
__ext4_iget+275
ext4_lookup+224
__lookup_slow+130
walk_component+219
link_path_walk.part.0.constprop.0+614
path_lookupat+62
filename_lookup+204
vfs_statx+128
vfs_fstatat+131
__do_sys_newfstatat+38
do_syscall_64+87
entry_SYSCALL_64_after_hwframe+118
, 1]: 20000
@[
__ext4_iget+275
ext4_lookup+224
__lookup_slow+130
walk_component+219
path_lookupat+106
filename_lookup+204
vfs_statx+128
vfs_fstatat+131
__do_sys_newfstatat+38
do_syscall_64+87
entry_SYSCALL_64_after_hwframe+118
, 1]: 20000000
That is 20 million calls for the initial lookup and 20 million after
allocating a new inode, all of them failing to return a value != 0
(i.e., they are returning NULL -- no match found).
Of course aborting the benchmark in the middle and starting it again (or
messing with the state in other ways) is going to alter these results.
Benchmark can be found here: https://people.freebsd.org/~mjg/fstree.tgz
[1] https://lore.kernel.org/all/20231206060629.2827226-1-david@fromorbit.com/
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Link: https://lore.kernel.org/r/20240611173824.535995-2-mjguzik@gmail.com
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Pull bcachefs updates from Kent Overstreet:
- Subvolume children btree; this is needed for providing a userspace
interface for walking subvolumes, which will come later
- Lots of improvements to directory structure checking
- Improved journal pipelining, significantly improving performance on
high iodepth write workloads
- Discard path improvements: the discard path is more efficient, and no
longer flushes the journal unnecessarily
- Buffered write path can now avoid taking the inode lock
- new mm helper: memalloc_flags_{save|restore}
- mempool now does kvmalloc mempools
* tag 'bcachefs-2024-03-13' of https://evilpiepirate.org/git/bcachefs: (128 commits)
bcachefs: time_stats: shrink time_stat_buffer for better alignment
bcachefs: time_stats: split stats-with-quantiles into a separate structure
bcachefs: mean_and_variance: put struct mean_and_variance_weighted on a diet
bcachefs: time_stats: add larger units
bcachefs: pull out time_stats.[ch]
bcachefs: reconstruct_alloc cleanup
bcachefs: fix bch_folio_sector padding
bcachefs: Fix btree key cache coherency during replay
bcachefs: Always flush write buffer in delete_dead_inodes()
bcachefs: Fix order of gc_done passes
bcachefs: fix deletion of indirect extents in btree_gc
bcachefs: Prefer struct_size over open coded arithmetic
bcachefs: Kill unused flags argument to btree_split()
bcachefs: Check for writing superblocks with nonsense member seq fields
bcachefs: fix bch2_journal_buf_to_text()
lib/generic-radix-tree.c: Make nodes more reasonably sized
bcachefs: copy_(to|from)_user_errcode()
bcachefs: Split out bkey_types.h
bcachefs: fix lost journal buf wakeup due to improved pipelining
bcachefs: intercept mountoption value for bool type
...
Rename and export __file_remove_privs(); for a buffered write path that doesn't take the inode lock we need to be able to check if the operation needs to do work first. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Christian Brauner <brauner@kernel.org>