From: Jeff Layton <jlayton@redhat.com>
Date: Wed, 28 May 2008 11:07:09 -0400
Subject: [fs] nlm: canceled inflight GRANT_MSG shouldn't requeue
Message-id: 1211987231-9749-2-git-send-email-jlayton@redhat.com
O-Subject: [RHEL5.3 PATCH 1/3] BZ#254195: NLM: don't requeue block if it was invalidated while GRANT_MSG was in flight
Bugzilla: 254195
RH-Acked-by: Anton Arapov <aarapov@redhat.com>
RH-Acked-by: Peter Staubach <staubach@redhat.com>

It's possible for lockd to catch a SIGKILL while a GRANT_MSG callback
is in flight. If this happens we don't want the callback to insert the
block back into the nlm_blocked list.

diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c
index 48acce0..52a84da 100644
--- a/fs/lockd/svclock.c
+++ b/fs/lockd/svclock.c
@@ -786,6 +786,15 @@ static void nlmsvc_grant_callback(struct rpc_task *task, void *data)
 
 	dprintk("lockd: GRANT_MSG RPC callback\n");
 
+	/* if the block is not queued at this point then it has
+	 * been invalidated. Don't try to requeue it.
+	 *
+	 * The BKL makes sure that lockd doesn't dequeue the block
+	 * after we check this.
+	 */
+	if (!block->b_queued)
+		return;
+
 	/* Technically, we should down the file semaphore here. Since we
 	 * move the block towards the head of the queue only, no harm
 	 * can be done, though. */