[DLM] fix requestqueue race

Red Hat BZ 211914

There's a race between dlm_recoverd (1) enabling locking and (2) clearing
out the requestqueue, and dlm_recvd (1) checking if locking is enabled and
(2) adding a message to the requestqueue.  An order of recoverd(1),
recvd(1), recvd(2), recoverd(2) will result in a message being left on the
requestqueue.  The fix is to have dlm_recvd check if dlm_recoverd has
enabled locking after taking the mutex for the requestqueue and if it has
processing the message instead of queueing it.

Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

1 files changed, 11 insertions, 4 deletions

diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c
index 3f2befa4797b..6088a16926bf 100644
--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -3028,10 +3028,17 @@ int dlm_receive_message(struct dlm_header *hd, int nodeid, int recovery)
 
         while (1) {
                 if (dlm_locking_stopped(ls)) {
-                        if (!recovery)
-                                dlm_add_requestqueue(ls, nodeid, hd);
-                        error = -EINTR;
-                        goto out;
+                        if (recovery) {
+                                error = -EINTR;
+                                goto out;
+                        }
+                        error = dlm_add_requestqueue(ls, nodeid, hd);
+                        if (error == -EAGAIN)
+                                continue;
+                        else {
+                                error = -EINTR;
+                                goto out;
+                        }
                 }
 
                 if (lock_recovery_try(ls))