commit e1253be0ece1a95a02c7f5843194877471af8179
author Trond Myklebust <trond.myklebust@primarydata.com> Wed Mar 05 08:44:23 2014 -0500
committer Trond Myklebust <trond.myklebust@primarydata.com> Wed Mar 05 11:55:06 2014 -0500
tree d5f22de8fec53058a624f308df1375973c264c79
parent 755a48a7a4eb05b9c8424e3017d947b2961a60e0

NFSv4: nfs4_stateid_is_current should return 'true' for an invalid stateid

When nfs4_set_rw_stateid() can fails by returning EIO to indicate that
the stateid is completely invalid, then it makes no sense to have it
trigger a retry of the READ or WRITE operation. Instead, we should just
have it fall through and attempt a recovery.

This fixes an infinite loop in which the client keeps replaying the same
bad stateid back to the server.

Reported-by: Andy Adamson <andros@netapp.com>
Link: http://lkml.kernel.org/r/1393954269-3974-1-git-send-email-andros@netapp.com
Cc: stable@vger.kernel.org # 3.10+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

fs/nfs/nfs4proc.c

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 44e088d..4ae8141 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4011,8 +4011,9 @@
 {
 	nfs4_stateid current_stateid;
 
-	if (nfs4_set_rw_stateid(&current_stateid, ctx, l_ctx, fmode))
-		return false;
+	/* If the current stateid represents a lost lock, then exit */
+	if (nfs4_set_rw_stateid(&current_stateid, ctx, l_ctx, fmode) == -EIO)
+		return true;
 	return nfs4_stateid_match(stateid, &current_stateid);
 }