lockd: fix races in client GRANTED_MSG wait logic

After the wait for a grant is done (for whatever reason), nlmclnt_block
updates the status of the nlm_rqst with the status of the block. At the
point it does this, however, the block is still queued its status could
change at any time.

This is particularly a problem when the waiting task is signaled during
the wait. We can end up giving up on the lock just before the GRANTED_MSG
callback comes in, and accept it even though the lock request gets back
an error, leaving a dangling lock on the server.

Since the nlm_wait never lives beyond the end of nlmclnt_lock, put it on
the stack and add functions to allow us to enqueue and dequeue the
block. Enqueue it just before the lock/wait loop, and dequeue it
just after we exit the loop instead of waiting until the end of
the function. Also, scrape the status at the time that we dequeue it to
ensure that it's final.

Reported-by: Yongcheng Yang <yoyang@redhat.com>
Link: https://bugzilla.redhat.com/show_bug.cgi?id=2063818
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

1 files changed, 21 insertions, 21 deletions

diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c
index 464cb15c1a06..c374ee072db3 100644
--- a/fs/lockd/clntlock.c
+++ b/fs/lockd/clntlock.c
@@ -82,41 +82,42 @@ void nlmclnt_done(struct nlm_host *host)
 }
 EXPORT_SYMBOL_GPL(nlmclnt_done);
 
+void nlmclnt_prepare_block(struct nlm_wait *block, struct nlm_host *host, struct file_lock *fl)
+{
+	block->b_host = host;
+	block->b_lock = fl;
+	init_waitqueue_head(&block->b_wait);
+	block->b_status = nlm_lck_blocked;
+}
+
 /*
  * Queue up a lock for blocking so that the GRANTED request can see it
  */
-struct nlm_wait *nlmclnt_prepare_block(struct nlm_host *host, struct file_lock *fl)
+void nlmclnt_queue_block(struct nlm_wait *block)
 {
-	struct nlm_wait *block;
-
-	block = kmalloc(sizeof(*block), GFP_KERNEL);
-	if (block != NULL) {
-		block->b_host = host;
-		block->b_lock = fl;
-		init_waitqueue_head(&block->b_wait);
-		block->b_status = nlm_lck_blocked;
-
-		spin_lock(&nlm_blocked_lock);
-		list_add(&block->b_list, &nlm_blocked);
-		spin_unlock(&nlm_blocked_lock);
-	}
-	return block;
+	spin_lock(&nlm_blocked_lock);
+	list_add(&block->b_list, &nlm_blocked);
+	spin_unlock(&nlm_blocked_lock);
 }
 
-void nlmclnt_finish_block(struct nlm_wait *block)
+/*
+ * Dequeue the block and return its final status
+ */
+__be32 nlmclnt_dequeue_block(struct nlm_wait *block)
 {
-	if (block == NULL)
-		return;
+	__be32 status;
+
 	spin_lock(&nlm_blocked_lock);
 	list_del(&block->b_list);
+	status = block->b_status;
 	spin_unlock(&nlm_blocked_lock);
-	kfree(block);
+	return status;
 }
 
 /*
  * Block on a lock
  */
-int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout)
+int nlmclnt_wait(struct nlm_wait *block, struct nlm_rqst *req, long timeout)
 {
 	long ret;
 
@@ -142,7 +143,6 @@ int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout)
 	/* Reset the lock status after a server reboot so we resend */
 	if (block->b_status == nlm_lck_denied_grace_period)
 		block->b_status = nlm_lck_blocked;
-	req->a_res.status = block->b_status;
 	return 0;
 }