提交 7bfac9ec 编写于 作者: M Miklos Szeredi 提交者: Linus Torvalds

splice: fix deadlock in splicing to file

There's a possible deadlock in generic_file_splice_write(),
splice_from_pipe() and ocfs2_file_splice_write():

 - task A calls generic_file_splice_write()
 - this calls inode_double_lock(), which locks i_mutex on both
   pipe->inode and target inode
 - ordering depends on inode pointers, can happen that pipe->inode is
   locked first
 - __splice_from_pipe() needs more data, calls pipe_wait()
 - this releases lock on pipe->inode, goes to interruptible sleep
 - task B calls generic_file_splice_write(), similarly to the first
 - this locks pipe->inode, then tries to lock inode, but that is
   already held by task A
 - task A is interrupted, it tries to lock pipe->inode, but fails, as
   it is already held by task B
 - ABBA deadlock

Fix this by explicitly ordering locks: the outer lock must be on
target inode and the inner lock (which is later unlocked and relocked)
must be on pipe->inode.  This is OK, pipe inodes and target inodes
form two nonoverlapping sets, generic_file_splice_write() and friends
are not called with a target which is a pipe.
Signed-off-by: NMiklos Szeredi <mszeredi@suse.cz>
Acked-by: NMark Fasheh <mfasheh@suse.com>
Acked-by: NJens Axboe <jens.axboe@oracle.com>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
上级 61239230
......@@ -1926,7 +1926,7 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe,
out->f_path.dentry->d_name.len,
out->f_path.dentry->d_name.name);
inode_double_lock(inode, pipe->inode);
mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT);
ret = ocfs2_rw_lock(inode, 1);
if (ret < 0) {
......@@ -1941,12 +1941,16 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe,
goto out_unlock;
}
if (pipe->inode)
mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_CHILD);
ret = generic_file_splice_write_nolock(pipe, out, ppos, len, flags);
if (pipe->inode)
mutex_unlock(&pipe->inode->i_mutex);
out_unlock:
ocfs2_rw_unlock(inode, 1);
out:
inode_double_unlock(inode, pipe->inode);
mutex_unlock(&inode->i_mutex);
mlog_exit(ret);
return ret;
......
......@@ -737,10 +737,19 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out,
* ->write_end. Most of the time, these expect i_mutex to
* be held. Since this may result in an ABBA deadlock with
* pipe->inode, we have to order lock acquiry here.
*
* Outer lock must be inode->i_mutex, as pipe_wait() will
* release and reacquire pipe->inode->i_mutex, AND inode must
* never be a pipe.
*/
inode_double_lock(inode, pipe->inode);
WARN_ON(S_ISFIFO(inode->i_mode));
mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT);
if (pipe->inode)
mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_CHILD);
ret = __splice_from_pipe(pipe, &sd, actor);
inode_double_unlock(inode, pipe->inode);
if (pipe->inode)
mutex_unlock(&pipe->inode->i_mutex);
mutex_unlock(&inode->i_mutex);
return ret;
}
......@@ -831,11 +840,17 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
};
ssize_t ret;
inode_double_lock(inode, pipe->inode);
WARN_ON(S_ISFIFO(inode->i_mode));
mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT);
ret = file_remove_suid(out);
if (likely(!ret))
if (likely(!ret)) {
if (pipe->inode)
mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_CHILD);
ret = __splice_from_pipe(pipe, &sd, pipe_to_file);
inode_double_unlock(inode, pipe->inode);
if (pipe->inode)
mutex_unlock(&pipe->inode->i_mutex);
}
mutex_unlock(&inode->i_mutex);
if (ret > 0) {
unsigned long nr_pages;
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册