Btrfs: fix race when updating existing ref head

While we update an existing ref head's extent_op, we're not holding
its spinlock, so while we're updating its extent_op contents (key,
flags) we can have a task running __btrfs_run_delayed_refs() that
holds the ref head's lock and sets its extent_op to NULL right after
the task updating the ref head just checked its extent_op was not NULL.

Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Chris Mason <clm@fb.com>
This commit is contained in:
Filipe Manana 2014-03-14 20:55:01 +00:00 committed by Chris Mason
parent c3a468915a
commit 21543baddc

View file

@ -495,6 +495,7 @@ update_existing_head_ref(struct btrfs_delayed_ref_node *existing,
ref = btrfs_delayed_node_to_head(update);
BUG_ON(existing_ref->is_data != ref->is_data);
spin_lock(&existing_ref->lock);
if (ref->must_insert_reserved) {
/* if the extent was freed and then
* reallocated before the delayed ref
@ -536,7 +537,6 @@ update_existing_head_ref(struct btrfs_delayed_ref_node *existing,
* only need the lock for this case cause we could be processing it
* currently, for refs we just added we know we're a-ok.
*/
spin_lock(&existing_ref->lock);
existing->ref_mod += update->ref_mod;
spin_unlock(&existing_ref->lock);
}