On Mon, Aug 5, 2024 at 3:24 AM Zhihao Cheng <chengzhihao@xxxxxxxxxxxxxxx> wrote: > > From: Zhihao Cheng <chengzhihao1@xxxxxxxxxx> > > 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 it by forbidding inode evicting under the inode lru traversing > context. In details, we import a new inode state flag 'I_LRU_ISOLATING' > to pin inode without holding i_count under the inode lru traversing > context, the inode evicting process will wait until this flag is > cleared from i_state. > > Link: https://lore.kernel.org/all/37c29c42-7685-d1f0-067d-63582ffac405@xxxxxxxxxxxxxxx/ > 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") I only have some tidy-ups with stuff I neglected to mention when typing up my proposal. > --- > fs/inode.c | 37 +++++++++++++++++++++++++++++++++++-- > include/linux/fs.h | 5 +++++ > 2 files changed, 40 insertions(+), 2 deletions(-) > > diff --git a/fs/inode.c b/fs/inode.c > index 86670941884b..f1c6e8072f39 100644 > --- a/fs/inode.c > +++ b/fs/inode.c > @@ -488,6 +488,36 @@ static void inode_lru_list_del(struct inode *inode) > this_cpu_dec(nr_unused); > } > > +static void inode_lru_isolating(struct inode *inode) > +{ lockdep_assert_held(&inode->i_lock); > + BUG_ON(inode->i_state & (I_LRU_ISOLATING | I_FREEING | I_WILL_FREE)); > + inode->i_state |= I_LRU_ISOLATING; > +} > + > +static void inode_lru_finish_isolating(struct inode *inode) > +{ > + spin_lock(&inode->i_lock); > + BUG_ON(!(inode->i_state & I_LRU_ISOLATING)); > + inode->i_state &= ~I_LRU_ISOLATING; > + wake_up_bit(&inode->i_state, __I_LRU_ISOLATING); > + spin_unlock(&inode->i_lock); > +} > + > +static void inode_wait_for_lru_isolating(struct inode *inode) > +{ > + DEFINE_WAIT_BIT(wq, &inode->i_state, __I_LRU_ISOLATING); > + wait_queue_head_t *wqh; > + Top of evict() asserts on I_FREEING being set, used to decide it's not legit to pin above. This dependency can be documented it in the routine as well: BUG_ON(!(inode->i_state & I_FREEING)); > + spin_lock(&inode->i_lock); This lock acquire is avoidable, which is always nice to do for single-threaded perf. Probably can be also done for the writeback code below. Maybe I'll massage it myself after the patch lands. > + wqh = bit_waitqueue(&inode->i_state, __I_LRU_ISOLATING); > + while (inode->i_state & I_LRU_ISOLATING) { > + spin_unlock(&inode->i_lock); > + __wait_on_bit(wqh, &wq, bit_wait, TASK_UNINTERRUPTIBLE); > + spin_lock(&inode->i_lock); > + } > + spin_unlock(&inode->i_lock); > +} So new arrivals *are* blocked by this point thanks to I_FREEING being set on entry to evict(). This also means the flag can show up at most once. Thus instead of looping this should merely go to sleep once and assert the I_LRU_ISOLATING flag is no longer set after waking up. > + > /** > * inode_sb_list_add - add inode to the superblock list of inodes > * @inode: inode to add > @@ -657,6 +687,9 @@ static void evict(struct inode *inode) > > inode_sb_list_del(inode); > > + /* Wait for LRU isolating to finish. */ I don't think this comment adds anything given the name of the func. > + inode_wait_for_lru_isolating(inode); > + > /* > * Wait for flusher thread to be done with the inode so that filesystem > * does not start destroying it while writeback is still running. Since > @@ -855,7 +888,7 @@ static enum lru_status inode_lru_isolate(struct list_head *item, > * be under pressure before the cache inside the highmem zone. > */ > if (inode_has_buffers(inode) || !mapping_empty(&inode->i_data)) { > - __iget(inode); > + inode_lru_isolating(inode); > spin_unlock(&inode->i_lock); > spin_unlock(lru_lock); > if (remove_inode_buffers(inode)) { > @@ -867,7 +900,7 @@ static enum lru_status inode_lru_isolate(struct list_head *item, > __count_vm_events(PGINODESTEAL, reap); > mm_account_reclaimed_pages(reap); > } > - iput(inode); > + inode_lru_finish_isolating(inode); > spin_lock(lru_lock); > return LRU_RETRY; > } > diff --git a/include/linux/fs.h b/include/linux/fs.h > index fd34b5755c0b..fb0426f349fc 100644 > --- a/include/linux/fs.h > +++ b/include/linux/fs.h > @@ -2392,6 +2392,9 @@ static inline void kiocb_clone(struct kiocb *kiocb, struct kiocb *kiocb_src, > * > * I_PINNING_FSCACHE_WB Inode is pinning an fscache object for writeback. > * > + * I_LRU_ISOLATING Inode is pinned being isolated from LRU without holding > + * i_count. > + * > * Q: What is the difference between I_WILL_FREE and I_FREEING? > */ > #define I_DIRTY_SYNC (1 << 0) > @@ -2415,6 +2418,8 @@ static inline void kiocb_clone(struct kiocb *kiocb, struct kiocb *kiocb_src, > #define I_DONTCACHE (1 << 16) > #define I_SYNC_QUEUED (1 << 17) > #define I_PINNING_NETFS_WB (1 << 18) > +#define __I_LRU_ISOLATING 19 > +#define I_LRU_ISOLATING (1 << __I_LRU_ISOLATING) > > #define I_DIRTY_INODE (I_DIRTY_SYNC | I_DIRTY_DATASYNC) > #define I_DIRTY (I_DIRTY_INODE | I_DIRTY_PAGES) > -- > 2.39.2 > -- Mateusz Guzik <mjguzik gmail.com>
