Merge branch 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6

* 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6:
  NFSv4: Fix a regression in the NFSv4 state manager
  NFSv4: Release the sequence id before restarting a CLOSE rpc call
  nfs41: fix session fore channel negotiation
  nfs41: do not zero seqid portion of stateid on close
  nfs: run state manager in privileged mode
  nfs: make recovery state manager operations privileged
  nfs: enforce FIFO ordering of operations trying to acquire slot
  rpc: add a new priority in RPC task
  nfs: remove rpc_task argument from nfs4_find_slot
  rpc: add rpc_queue_empty function
  nfs: change nfs4_do_setlk params to identify recovery type
  nfs: do not do a LOOKUP after open
  nfs: minor cleanup of session draining

fs/nfs/nfs4_fs.h

@@ -108,6 +108,10 @@ enum {
 	NFS_OWNER_RECLAIM_NOGRACE
 };
 
+#define NFS_LOCK_NEW		0
+#define NFS_LOCK_RECLAIM	1
+#define NFS_LOCK_EXPIRED	2
+
 /*
  * struct nfs4_state maintains the client-side state for a given
  * (state_owner,inode) tuple (OPEN) or state_owner (LOCK).
@@ -282,6 +286,7 @@ extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter);
 extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task);
 extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid);
 extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid);
+extern void nfs_release_seqid(struct nfs_seqid *seqid);
 extern void nfs_free_seqid(struct nfs_seqid *seqid);
 
 extern const nfs4_stateid zero_stateid;

fs/nfs/nfs4state.c

@@ -135,16 +135,30 @@ static int nfs41_setup_state_renewal(struct nfs_client *clp)
 	return status;
 }
 
-static void nfs41_end_drain_session(struct nfs_client *clp,
-		struct nfs4_session *ses)
+static void nfs4_end_drain_session(struct nfs_client *clp)
 {
-	if (test_and_clear_bit(NFS4CLNT_SESSION_DRAINING, &clp->cl_state))
-		rpc_wake_up(&ses->fc_slot_table.slot_tbl_waitq);
+	struct nfs4_session *ses = clp->cl_session;
+	int max_slots;
+
+	if (test_and_clear_bit(NFS4CLNT_SESSION_DRAINING, &clp->cl_state)) {
+		spin_lock(&ses->fc_slot_table.slot_tbl_lock);
+		max_slots = ses->fc_slot_table.max_slots;
+		while (max_slots--) {
+			struct rpc_task *task;
+
+			task = rpc_wake_up_next(&ses->fc_slot_table.
+						slot_tbl_waitq);
+			if (!task)
+				break;
+			rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
+		}
+		spin_unlock(&ses->fc_slot_table.slot_tbl_lock);
+	}
 }
 
-static int nfs41_begin_drain_session(struct nfs_client *clp,
-		struct nfs4_session *ses)
+static int nfs4_begin_drain_session(struct nfs_client *clp)
 {
+	struct nfs4_session *ses = clp->cl_session;
 	struct nfs4_slot_table *tbl = &ses->fc_slot_table;
 
 	spin_lock(&tbl->slot_tbl_lock);
@@ -162,16 +176,13 @@ int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	int status;
 
-	status = nfs41_begin_drain_session(clp, clp->cl_session);
-	if (status != 0)
-		goto out;
+	nfs4_begin_drain_session(clp);
 	status = nfs4_proc_exchange_id(clp, cred);
 	if (status != 0)
 		goto out;
 	status = nfs4_proc_create_session(clp);
 	if (status != 0)
 		goto out;
-	nfs41_end_drain_session(clp, clp->cl_session);
 	nfs41_setup_state_renewal(clp);
 	nfs_mark_client_ready(clp, NFS_CS_READY);
 out:
@@ -755,16 +766,21 @@ struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter)
 	return new;
 }
 
-void nfs_free_seqid(struct nfs_seqid *seqid)
+void nfs_release_seqid(struct nfs_seqid *seqid)
 {
 	if (!list_empty(&seqid->list)) {
 		struct rpc_sequence *sequence = seqid->sequence->sequence;
 
 		spin_lock(&sequence->lock);
-		list_del(&seqid->list);
+		list_del_init(&seqid->list);
 		spin_unlock(&sequence->lock);
 		rpc_wake_up(&sequence->wait);
 	}
+}
+
+void nfs_free_seqid(struct nfs_seqid *seqid)
+{
+	nfs_release_seqid(seqid);
 	kfree(seqid);
 }
 
@@ -1257,13 +1273,9 @@ void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags)
 
 static int nfs4_reset_session(struct nfs_client *clp)
 {
-	struct nfs4_session *ses = clp->cl_session;
 	int status;
 
-	status = nfs41_begin_drain_session(clp, ses);
-	if (status != 0)
-		return status;
-
+	nfs4_begin_drain_session(clp);
 	status = nfs4_proc_destroy_session(clp->cl_session);
 	if (status && status != -NFS4ERR_BADSESSION &&
 	    status != -NFS4ERR_DEADSESSION) {
@@ -1279,19 +1291,17 @@ static int nfs4_reset_session(struct nfs_client *clp)
 out:
 	/*
 	 * Let the state manager reestablish state
-	 * without waking other tasks yet.
 	 */
-	if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) {
-		/* Wake up the next rpc task */
-		nfs41_end_drain_session(clp, ses);
-		if (status == 0)
-			nfs41_setup_state_renewal(clp);
-	}
+	if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) &&
+	    status == 0)
+		nfs41_setup_state_renewal(clp);
+
 	return status;
 }
 
 #else /* CONFIG_NFS_V4_1 */
 static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
+static int nfs4_end_drain_session(struct nfs_client *clp) { return 0; }
 #endif /* CONFIG_NFS_V4_1 */
 
 /* Set NFS4CLNT_LEASE_EXPIRED for all v4.0 errors and for recoverable errors
@@ -1382,6 +1392,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 				goto out_error;
 		}
 
+		nfs4_end_drain_session(clp);
 		if (test_and_clear_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) {
 			nfs_client_return_marked_delegations(clp);
 			continue;
@@ -1398,6 +1409,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 out_error:
 	printk(KERN_WARNING "Error: state manager failed on NFSv4 server %s"
 			" with error %d\n", clp->cl_hostname, -status);
+	nfs4_end_drain_session(clp);
 	nfs4_clear_state_manager_bit(clp);
 }
 

include/linux/sunrpc/sched.h

@@ -173,7 +173,8 @@ struct rpc_task_setup {
 #define RPC_PRIORITY_LOW	(-1)
 #define RPC_PRIORITY_NORMAL	(0)
 #define RPC_PRIORITY_HIGH	(1)
-#define RPC_NR_PRIORITY		(1 + RPC_PRIORITY_HIGH - RPC_PRIORITY_LOW)
+#define RPC_PRIORITY_PRIVILEGED	(2)
+#define RPC_NR_PRIORITY		(1 + RPC_PRIORITY_PRIVILEGED - RPC_PRIORITY_LOW)
 
 struct rpc_timer {
 	struct timer_list timer;
@@ -229,6 +230,7 @@ void		rpc_wake_up_queued_task(struct rpc_wait_queue *,
 void		rpc_wake_up(struct rpc_wait_queue *);
 struct rpc_task *rpc_wake_up_next(struct rpc_wait_queue *);
 void		rpc_wake_up_status(struct rpc_wait_queue *, int);
+int		rpc_queue_empty(struct rpc_wait_queue *);
 void		rpc_delay(struct rpc_task *, unsigned long);
 void *		rpc_malloc(struct rpc_task *, size_t);
 void		rpc_free(void *);
@@ -254,6 +256,16 @@ static inline int rpc_wait_for_completion_task(struct rpc_task *task)
 	return __rpc_wait_for_completion_task(task, NULL);
 }
 
+static inline void rpc_task_set_priority(struct rpc_task *task, unsigned char prio)
+{
+	task->tk_priority = prio - RPC_PRIORITY_LOW;
+}
+
+static inline int rpc_task_has_priority(struct rpc_task *task, unsigned char prio)
+{
+	return (task->tk_priority + RPC_PRIORITY_LOW == prio);
+}
+
 #ifdef RPC_DEBUG
 static inline const char * rpc_qname(struct rpc_wait_queue *q)
 {

net/sunrpc/sched.c

@@ -210,6 +210,7 @@ void rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qnam
 {
 	__rpc_init_priority_wait_queue(queue, qname, RPC_NR_PRIORITY);
 }
+EXPORT_SYMBOL_GPL(rpc_init_priority_wait_queue);
 
 void rpc_init_wait_queue(struct rpc_wait_queue *queue, const char *qname)
 {
@@ -384,6 +385,20 @@ static void rpc_wake_up_task_queue_locked(struct rpc_wait_queue *queue, struct r
 		__rpc_do_wake_up_task(queue, task);
 }
 
+/*
+ * Tests whether rpc queue is empty
+ */
+int rpc_queue_empty(struct rpc_wait_queue *queue)
+{
+	int res;
+
+	spin_lock_bh(&queue->lock);
+	res = queue->qlen;
+	spin_unlock_bh(&queue->lock);
+	return (res == 0);
+}
+EXPORT_SYMBOL_GPL(rpc_queue_empty);
+
 /*
  * Wake up a task on a specific queue
  */