diff options
author | John Ogness <john.ogness@linutronix.de> | 2018-02-23 00:50:24 +0100 |
---|---|---|
committer | Al Viro <viro@zeniv.linux.org.uk> | 2018-03-29 15:07:44 -0400 |
commit | 8f04da2adbdffed8dc4b2feb00ec3b3d84683885 (patch) | |
tree | 856f7627b33aea9f930931ff143da1e150f3b629 | |
parent | f657a666fd1b1b9fe59963943c74c245ae66f4cc (diff) | |
download | linux-8f04da2adbdffed8dc4b2feb00ec3b3d84683885.tar.gz linux-8f04da2adbdffed8dc4b2feb00ec3b3d84683885.tar.bz2 linux-8f04da2adbdffed8dc4b2feb00ec3b3d84683885.zip |
fs/dcache: Avoid a try_lock loop in shrink_dentry_list()
shrink_dentry_list() holds dentry->d_lock and needs to acquire
dentry->d_inode->i_lock. This cannot be done with a spin_lock()
operation because it's the reverse of the regular lock order.
To avoid ABBA deadlocks it is done with a trylock loop.
Trylock loops are problematic in two scenarios:
1) PREEMPT_RT converts spinlocks to 'sleeping' spinlocks, which are
preemptible. As a consequence the i_lock holder can be preempted
by a higher priority task. If that task executes the trylock loop
it will do so forever and live lock.
2) In virtual machines trylock loops are problematic as well. The
VCPU on which the i_lock holder runs can be scheduled out and a
task on a different VCPU can loop for a whole time slice. In the
worst case this can lead to starvation. Commits 47be61845c77
("fs/dcache.c: avoid soft-lockup in dput()") and 046b961b45f9
("shrink_dentry_list(): take parent's d_lock earlier") are
addressing exactly those symptoms.
Avoid the trylock loop by using dentry_kill(). When pruning ancestors,
the same code applies that is used to kill a dentry in dput(). This
also has the benefit that the locking order is now the same. First
the inode is locked, then the parent.
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-rw-r--r-- | fs/dcache.c | 41 |
1 files changed, 10 insertions, 31 deletions
diff --git a/fs/dcache.c b/fs/dcache.c index f2d945688025..9b4e7fd0ee66 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -992,9 +992,11 @@ EXPORT_SYMBOL(d_prune_aliases); /* * Lock a dentry from shrink list. + * Called under rcu_read_lock() and dentry->d_lock; the former + * guarantees that nothing we access will be freed under us. * Note that dentry is *not* protected from concurrent dentry_kill(), - * d_delete(), etc. It is protected from freeing (by the fact of - * being on a shrink list), but everything else is fair game. + * d_delete(), etc. + * * Return false if dentry has been disrupted or grabbed, leaving * the caller to kick it off-list. Otherwise, return true and have * that dentry's inode and parent both locked. @@ -1009,7 +1011,6 @@ static bool shrink_lock_dentry(struct dentry *dentry) inode = dentry->d_inode; if (inode && unlikely(!spin_trylock(&inode->i_lock))) { - rcu_read_lock(); /* to protect inode */ spin_unlock(&dentry->d_lock); spin_lock(&inode->i_lock); spin_lock(&dentry->d_lock); @@ -1018,16 +1019,13 @@ static bool shrink_lock_dentry(struct dentry *dentry) /* changed inode means that somebody had grabbed it */ if (unlikely(inode != dentry->d_inode)) goto out; - rcu_read_unlock(); } parent = dentry->d_parent; if (IS_ROOT(dentry) || likely(spin_trylock(&parent->d_lock))) return true; - rcu_read_lock(); /* to protect parent */ spin_unlock(&dentry->d_lock); - parent = READ_ONCE(dentry->d_parent); spin_lock(&parent->d_lock); if (unlikely(parent != dentry->d_parent)) { spin_unlock(&parent->d_lock); @@ -1035,15 +1033,12 @@ static bool shrink_lock_dentry(struct dentry *dentry) goto out; } spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); - if (likely(!dentry->d_lockref.count)) { - rcu_read_unlock(); + if (likely(!dentry->d_lockref.count)) return true; - } spin_unlock(&parent->d_lock); out: if (inode) spin_unlock(&inode->i_lock); - rcu_read_unlock(); return false; } @@ -1051,12 +1046,13 @@ static void shrink_dentry_list(struct list_head *list) { while (!list_empty(list)) { struct dentry *dentry, *parent; - struct inode *inode; dentry = list_entry(list->prev, struct dentry, d_lru); spin_lock(&dentry->d_lock); + rcu_read_lock(); if (!shrink_lock_dentry(dentry)) { bool can_free = false; + rcu_read_unlock(); d_shrink_del(dentry); if (dentry->d_lockref.count < 0) can_free = dentry->d_flags & DCACHE_MAY_FREE; @@ -1065,6 +1061,7 @@ static void shrink_dentry_list(struct list_head *list) dentry_free(dentry); continue; } + rcu_read_unlock(); d_shrink_del(dentry); parent = dentry->d_parent; __dentry_kill(dentry); @@ -1077,26 +1074,8 @@ static void shrink_dentry_list(struct list_head *list) * fragmentation. */ dentry = parent; - while (dentry && !lockref_put_or_lock(&dentry->d_lockref)) { - parent = lock_parent(dentry); - if (dentry->d_lockref.count != 1) { - dentry->d_lockref.count--; - spin_unlock(&dentry->d_lock); - if (parent) - spin_unlock(&parent->d_lock); - break; - } - inode = dentry->d_inode; /* can't be NULL */ - if (unlikely(!spin_trylock(&inode->i_lock))) { - spin_unlock(&dentry->d_lock); - if (parent) - spin_unlock(&parent->d_lock); - cpu_relax(); - continue; - } - __dentry_kill(dentry); - dentry = parent; - } + while (dentry && !lockref_put_or_lock(&dentry->d_lockref)) + dentry = dentry_kill(dentry); } } |