Commit d01730d7 authored by Jiaying Zhang's avatar Jiaying Zhang Committed by Jan Kara

quota: Fix possible deadlock during parallel quotaon and quotaoff

The following test script triggers a deadlock on ext2 filesystem:
while true; do quotaon /dev/hda >&/dev/null; usleep $RANDOM; done &
while true; do quotaoff /dev/hda >&/dev/null; usleep $RANDOM; done &

I found there is a potential deadlock between quotaon and quotaoff (or
quotasync). Basically, all of quotactl operations need to be protected by
dqonoff_mutex. vfs_quota_off and vfs_quota_sync also call sb->s_op->quota_write
that needs to grab the i_mutex of the quota file.  But in vfs_quota_on_inode
(called from quotaon operation), the current code tries to grab  the i_mutex of
the quota file first before getting quonoff_mutex.

Reverse the order in which we take locks in vfs_quota_on_inode().

Jan Kara: Changed changelog to be more readable, made lockdep happy with
  I_MUTEX_QUOTA.
Signed-off-by: default avatarJiaying Zhang <jiayingz@google.com>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
parent faf80d62
......@@ -2042,8 +2042,8 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
* changes */
invalidate_bdev(sb->s_bdev);
}
mutex_lock(&inode->i_mutex);
mutex_lock(&dqopt->dqonoff_mutex);
mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA);
if (sb_has_quota_loaded(sb, type)) {
error = -EBUSY;
goto out_lock;
......@@ -2094,7 +2094,6 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
dqopt->files[type] = NULL;
iput(inode);
out_lock:
mutex_unlock(&dqopt->dqonoff_mutex);
if (oldflags != -1) {
down_write(&dqopt->dqptr_sem);
/* Set the flags back (in the case of accidental quotaon()
......@@ -2104,6 +2103,7 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
up_write(&dqopt->dqptr_sem);
}
mutex_unlock(&inode->i_mutex);
mutex_unlock(&dqopt->dqonoff_mutex);
out_fmt:
put_quota_format(fmt);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment