btrfs: destroy qgroup extent records on transaction abort
authorJeff Mahoney <jeffm@suse.com>
Tue, 11 Feb 2020 07:25:37 +0000 (15:25 +0800)
committerDavid Sterba <dsterba@suse.com>
Tue, 18 Feb 2020 23:35:54 +0000 (00:35 +0100)
commit81f7eb00ff5bb8326e82503a32809421d14abb8a
tree6c2ddb770d4bcdbd85db0881d040db3bbd21eb7b
parent52e29e331070cd7d52a64cbf1b0958212a340e28
btrfs: destroy qgroup extent records on transaction abort

We clean up the delayed references when we abort a transaction but we
leave the pending qgroup extent records behind, leaking memory.

This patch destroys the extent records when we destroy the delayed refs
and makes sure ensure they're gone before releasing the transaction.

Fixes: 3368d001ba5d ("btrfs: qgroup: Record possible quota-related extent for qgroup.")
CC: stable@vger.kernel.org # 4.4+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
[ Rebased to latest upstream, remove to_qgroup() helper, use
  rbtree_postorder_for_each_entry_safe() wrapper ]
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
fs/btrfs/disk-io.c
fs/btrfs/qgroup.c
fs/btrfs/qgroup.h
fs/btrfs/transaction.c