Merge branch 'nfs-for-2.6.34' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6

* 'nfs-for-2.6.34' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6: (44 commits)
  NFS: Remove requirement for inode->i_mutex from nfs_invalidate_mapping
  NFS: Clean up nfs_sync_mapping
  NFS: Simplify nfs_wb_page()
  NFS: Replace __nfs_write_mapping with sync_inode()
  NFS: Simplify nfs_wb_page_cancel()
  NFS: Ensure inode is always marked I_DIRTY_DATASYNC, if it has unstable pages
  NFS: Run COMMIT as an asynchronous RPC call when wbc->for_background is set
  NFS: Reduce the number of unnecessary COMMIT calls
  NFS: Add a count of the number of unstable writes carried by an inode
  NFS: Cleanup - move nfs_write_inode() into fs/nfs/write.c
  nfs41 fix NFS4ERR_CLID_INUSE for exchange id
  NFS: Fix an allocation-under-spinlock bug
  SUNRPC: Handle EINVAL error returns from the TCP connect operation
  NFSv4.1: Various fixes to the sequence flag error handling
  nfs4: renewd renew operations should take/put a client reference
  nfs41: renewd sequence operations should take/put client reference
  nfs: prevent backlogging of renewd requests
  nfs: kill renewd before clearing client minor version
  NFS: Make close(2) asynchronous when closing NFS O_DIRECT files
  NFS: Improve NFS iostat byte count accuracy for writes
  ...

17 files changed, 600 insertions, 422 deletions

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index d4036be0b589..85a7cfd1b8dd 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -119,6 +119,14 @@ struct cb_recallanyargs {
 };
 
 extern unsigned nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy);
+
+struct cb_recallslotargs {
+	struct sockaddr	*crsa_addr;
+	uint32_t	crsa_target_max_slots;
+};
+extern unsigned nfs4_callback_recallslot(struct cb_recallslotargs *args,
+					  void *dummy);
+
 #endif /* CONFIG_NFS_V4_1 */
 
 extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res);
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index defa9b4c470e..84761b5bb8e2 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -143,44 +143,49 @@ int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation, const n
  * Return success if the sequenceID is one more than what we last saw on
  * this slot, accounting for wraparound.  Increments the slot's sequence.
  *
- * We don't yet implement a duplicate request cache, so at this time
- * we will log replays, and process them as if we had not seen them before,
- * but we don't bump the sequence in the slot.  Not too worried about it,
+ * We don't yet implement a duplicate request cache, instead we set the
+ * back channel ca_maxresponsesize_cached to zero. This is OK for now
  * since we only currently implement idempotent callbacks anyway.
  *
  * We have a single slot backchannel at this time, so we don't bother
  * checking the used_slots bit array on the table.  The lower layer guarantees
  * a single outstanding callback request at a time.
  */
-static int
-validate_seqid(struct nfs4_slot_table *tbl, u32 slotid, u32 seqid)
+static __be32
+validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args)
 {
 	struct nfs4_slot *slot;
 
 	dprintk("%s enter. slotid %d seqid %d\n",
-		__func__, slotid, seqid);
+		__func__, args->csa_slotid, args->csa_sequenceid);
 
-	if (slotid > NFS41_BC_MAX_CALLBACKS)
+	if (args->csa_slotid > NFS41_BC_MAX_CALLBACKS)
 		return htonl(NFS4ERR_BADSLOT);
 
-	slot = tbl->slots + slotid;
+	slot = tbl->slots + args->csa_slotid;
 	dprintk("%s slot table seqid: %d\n", __func__, slot->seq_nr);
 
 	/* Normal */
-	if (likely(seqid == slot->seq_nr + 1)) {
+	if (likely(args->csa_sequenceid == slot->seq_nr + 1)) {
 		slot->seq_nr++;
 		return htonl(NFS4_OK);
 	}
 
 	/* Replay */
-	if (seqid == slot->seq_nr) {
-		dprintk("%s seqid %d is a replay - no DRC available\n",
-			__func__, seqid);
-		return htonl(NFS4_OK);
+	if (args->csa_sequenceid == slot->seq_nr) {
+		dprintk("%s seqid %d is a replay\n",
+			__func__, args->csa_sequenceid);
+		/* Signal process_op to set this error on next op */
+		if (args->csa_cachethis == 0)
+			return htonl(NFS4ERR_RETRY_UNCACHED_REP);
+
+		/* The ca_maxresponsesize_cached is 0 with no DRC */
+		else if (args->csa_cachethis == 1)
+			return htonl(NFS4ERR_REP_TOO_BIG_TO_CACHE);
 	}
 
 	/* Wraparound */
-	if (seqid == 1 && (slot->seq_nr + 1) == 0) {
+	if (args->csa_sequenceid == 1 && (slot->seq_nr + 1) == 0) {
 		slot->seq_nr = 1;
 		return htonl(NFS4_OK);
 	}
@@ -225,27 +230,87 @@ validate_seqid(struct nfs4_slot_table *tbl, u32 slotid, u32 seqid)
 	return NULL;
 }
 
-/* FIXME: referring calls should be processed */
-unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
+/*
+ * For each referring call triple, check the session's slot table for
+ * a match.  If the slot is in use and the sequence numbers match, the
+ * client is still waiting for a response to the original request.
+ */
+static bool referring_call_exists(struct nfs_client *clp,
+				  uint32_t nrclists,
+				  struct referring_call_list *rclists)
+{
+	bool status = 0;
+	int i, j;
+	struct nfs4_session *session;
+	struct nfs4_slot_table *tbl;
+	struct referring_call_list *rclist;
+	struct referring_call *ref;
+
+	/*
+	 * XXX When client trunking is implemented, this becomes
+	 * a session lookup from within the loop
+	 */
+	session = clp->cl_session;
+	tbl = &session->fc_slot_table;
+
+	for (i = 0; i < nrclists; i++) {
+		rclist = &rclists[i];
+		if (memcmp(session->sess_id.data,
+			   rclist->rcl_sessionid.data,
+			   NFS4_MAX_SESSIONID_LEN) != 0)
+			continue;
+
+		for (j = 0; j < rclist->rcl_nrefcalls; j++) {
+			ref = &rclist->rcl_refcalls[j];
+
+			dprintk("%s: sessionid %x:%x:%x:%x sequenceid %u "
+				"slotid %u\n", __func__,
+				((u32 *)&rclist->rcl_sessionid.data)[0],
+				((u32 *)&rclist->rcl_sessionid.data)[1],
+				((u32 *)&rclist->rcl_sessionid.data)[2],
+				((u32 *)&rclist->rcl_sessionid.data)[3],
+				ref->rc_sequenceid, ref->rc_slotid);
+
+			spin_lock(&tbl->slot_tbl_lock);
+			status = (test_bit(ref->rc_slotid, tbl->used_slots) &&
+				  tbl->slots[ref->rc_slotid].seq_nr ==
+					ref->rc_sequenceid);
+			spin_unlock(&tbl->slot_tbl_lock);
+			if (status)
+				goto out;
+		}
+	}
+
+out:
+	return status;
+}
+
+__be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 				struct cb_sequenceres *res)
 {
 	struct nfs_client *clp;
-	int i, status;
-
-	for (i = 0; i < args->csa_nrclists; i++)
-		kfree(args->csa_rclists[i].rcl_refcalls);
-	kfree(args->csa_rclists);
+	int i;
+	__be32 status;
 
 	status = htonl(NFS4ERR_BADSESSION);
 	clp = find_client_with_session(args->csa_addr, 4, &args->csa_sessionid);
 	if (clp == NULL)
 		goto out;
 
-	status = validate_seqid(&clp->cl_session->bc_slot_table,
-				args->csa_slotid, args->csa_sequenceid);
+	status = validate_seqid(&clp->cl_session->bc_slot_table, args);
 	if (status)
 		goto out_putclient;
 
+	/*
+	 * Check for pending referring calls.  If a match is found, a
+	 * related callback was received before the response to the original
+	 * call.
+	 */
+	if (referring_call_exists(clp, args->csa_nrclists, args->csa_rclists)) {
+		status = htonl(NFS4ERR_DELAY);
+		goto out_putclient;
+	}
+
 	memcpy(&res->csr_sessionid, &args->csa_sessionid,
 	       sizeof(res->csr_sessionid));
 	res->csr_sequenceid = args->csa_sequenceid;
@@ -256,15 +321,23 @@ unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
 out_putclient:
 	nfs_put_client(clp);
 out:
-	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
-	res->csr_status = status;
-	return res->csr_status;
+	for (i = 0; i < args->csa_nrclists; i++)
+		kfree(args->csa_rclists[i].rcl_refcalls);
+	kfree(args->csa_rclists);
+
+	if (status == htonl(NFS4ERR_RETRY_UNCACHED_REP))
+		res->csr_status = 0;
+	else
+		res->csr_status = status;
+	dprintk("%s: exit with status = %d res->csr_status %d\n", __func__,
+		ntohl(status), ntohl(res->csr_status));
+	return status;
 }
 
-unsigned nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy)
+__be32 nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy)
 {
 	struct nfs_client *clp;
-	int status;
+	__be32 status;
 	fmode_t flags = 0;
 
 	status = htonl(NFS4ERR_OP_NOT_IN_SESSION);
@@ -289,4 +362,40 @@ out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
 	return status;
 }
+
+/* Reduce the fore channel's max_slots to the target value */
+__be32 nfs4_callback_recallslot(struct cb_recallslotargs *args, void *dummy)
+{
+	struct nfs_client *clp;
+	struct nfs4_slot_table *fc_tbl;
+	__be32 status;
+
+	status = htonl(NFS4ERR_OP_NOT_IN_SESSION);
+	clp = nfs_find_client(args->crsa_addr, 4);
+	if (clp == NULL)
+		goto out;
+
+	dprintk("NFS: CB_RECALL_SLOT request from %s target max slots %d\n",
+		rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR),
+		args->crsa_target_max_slots);
+
+	fc_tbl = &clp->cl_session->fc_slot_table;
+
+	status = htonl(NFS4ERR_BAD_HIGH_SLOT);
+	if (args->crsa_target_max_slots > fc_tbl->max_slots ||
+	    args->crsa_target_max_slots < 1)
+		goto out_putclient;
+
+	status = htonl(NFS4_OK);
+	if (args->crsa_target_max_slots == fc_tbl->max_slots)
+		goto out_putclient;
+
+	fc_tbl->target_max_slots = args->crsa_target_max_slots;
+	nfs41_handle_recall_slot(clp);
+out_putclient:
+	nfs_put_client(clp);	/* balance nfs_find_client */
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 8e1a2511c8be..db30c0b398b5 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -24,10 +24,14 @@
 #define CB_OP_SEQUENCE_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ + \
 					4 + 1 + 3)
 #define CB_OP_RECALLANY_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
+#define CB_OP_RECALLSLOT_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 #endif /* CONFIG_NFS_V4_1 */
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
 
+/* Internal error code */
+#define NFS4ERR_RESOURCE_HDR	11050
+
 typedef __be32 (*callback_process_op_t)(void *, void *);
 typedef __be32 (*callback_decode_arg_t)(struct svc_rqst *, struct xdr_stream *, void *);
 typedef __be32 (*callback_encode_res_t)(struct svc_rqst *, struct xdr_stream *, void *);
@@ -173,7 +177,7 @@ static __be32 decode_op_hdr(struct xdr_stream *xdr, unsigned int *op)
 	__be32 *p;
 	p = read_buf(xdr, 4);
 	if (unlikely(p == NULL))
-		return htonl(NFS4ERR_RESOURCE);
+		return htonl(NFS4ERR_RESOURCE_HDR);
 	*op = ntohl(*p);
 	return 0;
 }
@@ -215,10 +219,10 @@ out:
 
 #if defined(CONFIG_NFS_V4_1)
 
-static unsigned decode_sessionid(struct xdr_stream *xdr,
+static __be32 decode_sessionid(struct xdr_stream *xdr,
 				 struct nfs4_sessionid *sid)
 {
-	uint32_t *p;
+	__be32 *p;
 	int len = NFS4_MAX_SESSIONID_LEN;
 
 	p = read_buf(xdr, len);
@@ -229,12 +233,12 @@ static unsigned decode_sessionid(struct xdr_stream *xdr,
 	return 0;
 }
 
-static unsigned decode_rc_list(struct xdr_stream *xdr,
+static __be32 decode_rc_list(struct xdr_stream *xdr,
 			       struct referring_call_list *rc_list)
 {
-	uint32_t *p;
+	__be32 *p;
 	int i;
-	unsigned status;
+	__be32 status;
 
 	status = decode_sessionid(xdr, &rc_list->rcl_sessionid);
 	if (status)
@@ -267,13 +271,13 @@ out:
 	return status;
 }
 
-static unsigned decode_cb_sequence_args(struct svc_rqst *rqstp,
+static __be32 decode_cb_sequence_args(struct svc_rqst *rqstp,
 					struct xdr_stream *xdr,
 					struct cb_sequenceargs *args)
 {
-	uint32_t *p;
+	__be32 *p;
 	int i;
-	unsigned status;
+	__be32 status;
 
 	status = decode_sessionid(xdr, &args->csa_sessionid);
 	if (status)
@@ -327,11 +331,11 @@ out_free:
 	goto out;
 }
 
-static unsigned decode_recallany_args(struct svc_rqst *rqstp,
+static __be32 decode_recallany_args(struct svc_rqst *rqstp,
 				      struct xdr_stream *xdr,
 				      struct cb_recallanyargs *args)
 {
-	uint32_t *p;
+	__be32 *p;
 
 	args->craa_addr = svc_addr(rqstp);
 	p = read_buf(xdr, 4);
@@ -346,6 +350,20 @@ static unsigned decode_recallany_args(struct svc_rqst *rqstp,
 	return 0;
 }
 
+static __be32 decode_recallslot_args(struct svc_rqst *rqstp,
+					struct xdr_stream *xdr,
+					struct cb_recallslotargs *args)
+{
+	__be32 *p;
+
+	args->crsa_addr = svc_addr(rqstp);
+	p = read_buf(xdr, 4);
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_BADXDR);
+	args->crsa_target_max_slots = ntohl(*p++);
+	return 0;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str)
@@ -465,7 +483,7 @@ static __be32 encode_op_hdr(struct xdr_stream *xdr, uint32_t op, __be32 res)
 	
 	p = xdr_reserve_space(xdr, 8);
 	if (unlikely(p == NULL))
-		return htonl(NFS4ERR_RESOURCE);
+		return htonl(NFS4ERR_RESOURCE_HDR);
 	*p++ = htonl(op);
 	*p = res;
 	return 0;
@@ -499,10 +517,10 @@ out:
 
 #if defined(CONFIG_NFS_V4_1)
 
-static unsigned encode_sessionid(struct xdr_stream *xdr,
+static __be32 encode_sessionid(struct xdr_stream *xdr,
 				 const struct nfs4_sessionid *sid)
 {
-	uint32_t *p;
+	__be32 *p;
 	int len = NFS4_MAX_SESSIONID_LEN;
 
 	p = xdr_reserve_space(xdr, len);
@@ -513,11 +531,11 @@ static unsigned encode_sessionid(struct xdr_stream *xdr,
 	return 0;
 }
 
-static unsigned encode_cb_sequence_res(struct svc_rqst *rqstp,
+static __be32 encode_cb_sequence_res(struct svc_rqst *rqstp,
 				       struct xdr_stream *xdr,
 				       const struct cb_sequenceres *res)
 {
-	uint32_t *p;
+	__be32 *p;
 	unsigned status = res->csr_status;
 
 	if (unlikely(status != 0))
@@ -554,6 +572,7 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	case OP_CB_RECALL:
 	case OP_CB_SEQUENCE:
 	case OP_CB_RECALL_ANY:
+	case OP_CB_RECALL_SLOT:
 		*op = &callback_ops[op_nr];
 		break;
 
@@ -562,7 +581,6 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	case OP_CB_NOTIFY:
 	case OP_CB_PUSH_DELEG:
 	case OP_CB_RECALLABLE_OBJ_AVAIL:
-	case OP_CB_RECALL_SLOT:
 	case OP_CB_WANTS_CANCELLED:
 	case OP_CB_NOTIFY_LOCK:
 		return htonl(NFS4ERR_NOTSUPP);
@@ -602,20 +620,18 @@ preprocess_nfs4_op(unsigned int op_nr, struct callback_op **op)
 static __be32 process_op(uint32_t minorversion, int nop,
 		struct svc_rqst *rqstp,
 		struct xdr_stream *xdr_in, void *argp,
-		struct xdr_stream *xdr_out, void *resp)
+		struct xdr_stream *xdr_out, void *resp, int* drc_status)
 {
 	struct callback_op *op = &callback_ops[0];
-	unsigned int op_nr = OP_CB_ILLEGAL;
+	unsigned int op_nr;
 	__be32 status;
 	long maxlen;
 	__be32 res;
 
 	dprintk("%s: start\n", __func__);
 	status = decode_op_hdr(xdr_in, &op_nr);
-	if (unlikely(status)) {
-		status = htonl(NFS4ERR_OP_ILLEGAL);
-		goto out;
-	}
+	if (unlikely(status))
+		return status;
 
 	dprintk("%s: minorversion=%d nop=%d op_nr=%u\n",
 		__func__, minorversion, nop, op_nr);
@@ -624,19 +640,32 @@ static __be32 process_op(uint32_t minorversion, int nop,
 				preprocess_nfs4_op(op_nr, &op);
 	if (status == htonl(NFS4ERR_OP_ILLEGAL))
 		op_nr = OP_CB_ILLEGAL;
-out:
+	if (status)
+		goto encode_hdr;
+
+	if (*drc_status) {
+		status = *drc_status;
+		goto encode_hdr;
+	}
+
 	maxlen = xdr_out->end - xdr_out->p;
 	if (maxlen > 0 && maxlen < PAGE_SIZE) {
-		if (likely(status == 0 && op->decode_args != NULL))
-			status = op->decode_args(rqstp, xdr_in, argp);
-		if (likely(status == 0 && op->process_op != NULL))
+		status = op->decode_args(rqstp, xdr_in, argp);
+		if (likely(status == 0))
 			status = op->process_op(argp, resp);
 	} else
 		status = htonl(NFS4ERR_RESOURCE);
 
+	/* Only set by OP_CB_SEQUENCE processing */
+	if (status == htonl(NFS4ERR_RETRY_UNCACHED_REP)) {
+		*drc_status = status;
+		status = 0;
+	}
+
+encode_hdr:
 	res = encode_op_hdr(xdr_out, op_nr, status);
-	if (status == 0)
-		status = res;
+	if (unlikely(res))
+		return res;
 	if (op->encode_res != NULL && status == 0)
 		status = op->encode_res(rqstp, xdr_out, resp);
 	dprintk("%s: done, status = %d\n", __func__, ntohl(status));
@@ -652,7 +681,7 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r
 	struct cb_compound_hdr_res hdr_res = { NULL };
 	struct xdr_stream xdr_in, xdr_out;
 	__be32 *p;
-	__be32 status;
+	__be32 status, drc_status = 0;
 	unsigned int nops = 0;
 
 	dprintk("%s: start\n", __func__);
@@ -672,11 +701,18 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r
 		return rpc_system_err;
 
 	while (status == 0 && nops != hdr_arg.nops) {
-		status = process_op(hdr_arg.minorversion, nops,
-				    rqstp, &xdr_in, argp, &xdr_out, resp);
+		status = process_op(hdr_arg.minorversion, nops, rqstp,
+				    &xdr_in, argp, &xdr_out, resp, &drc_status);
 		nops++;
 	}
 
+	/* Buffer overflow in decode_ops_hdr or encode_ops_hdr. Return
+	* resource error in cb_compound status without returning op */
+	if (unlikely(status == htonl(NFS4ERR_RESOURCE_HDR))) {
+		status = htonl(NFS4ERR_RESOURCE);
+		nops--;
+	}
+
 	*hdr_res.status = status;
 	*hdr_res.nops = htonl(nops);
 	dprintk("%s: done, status = %u\n", __func__, ntohl(status));
@@ -713,6 +749,11 @@ static struct callback_op callback_ops[] = {
 		.decode_args = (callback_decode_arg_t)decode_recallany_args,
 		.res_maxsize = CB_OP_RECALLANY_RES_MAXSZ,
 	},
+	[OP_CB_RECALL_SLOT] = {
+		.process_op = (callback_process_op_t)nfs4_callback_recallslot,
+		.decode_args = (callback_decode_arg_t)decode_recallslot_args,
+		.res_maxsize = CB_OP_RECALLSLOT_RES_MAXSZ,
+	},
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index ee77713ce68b..2274f1737336 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -164,30 +164,7 @@ error_0:
 	return ERR_PTR(err);
 }
 
-static void nfs4_shutdown_client(struct nfs_client *clp)
-{
-#ifdef CONFIG_NFS_V4
-	if (__test_and_clear_bit(NFS_CS_RENEWD, &clp->cl_res_state))
-		nfs4_kill_renewd(clp);
-	BUG_ON(!RB_EMPTY_ROOT(&clp->cl_state_owners));
-	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state))
-		nfs_idmap_delete(clp);
-
-	rpc_destroy_wait_queue(&clp->cl_rpcwaitq);
-#endif
-}
-
-/*
- * Destroy the NFS4 callback service
- */
-static void nfs4_destroy_callback(struct nfs_client *clp)
-{
 #ifdef CONFIG_NFS_V4
-	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state))
-		nfs_callback_down(clp->cl_minorversion);
-#endif /* CONFIG_NFS_V4 */
-}
-
 /*
  * Clears/puts all minor version specific parts from an nfs_client struct
  * reverting it to minorversion 0.
@@ -202,9 +179,33 @@ static void nfs4_clear_client_minor_version(struct nfs_client *clp)
 
 	clp->cl_call_sync = _nfs4_call_sync;
 #endif /* CONFIG_NFS_V4_1 */
+}
 
+/*
+ * Destroy the NFS4 callback service
+ */
+static void nfs4_destroy_callback(struct nfs_client *clp)
+{
+	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state))
+		nfs_callback_down(clp->cl_minorversion);
+}
+
+static void nfs4_shutdown_client(struct nfs_client *clp)
+{
+	if (__test_and_clear_bit(NFS_CS_RENEWD, &clp->cl_res_state))
+		nfs4_kill_renewd(clp);
+	nfs4_clear_client_minor_version(clp);
 	nfs4_destroy_callback(clp);
+	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state))
+		nfs_idmap_delete(clp);
+
+	rpc_destroy_wait_queue(&clp->cl_rpcwaitq);
 }
+#else
+static void nfs4_shutdown_client(struct nfs_client *clp)
+{
+}
+#endif /* CONFIG_NFS_V4 */
 
 /*
  * Destroy a shared client record
@@ -213,7 +214,6 @@ static void nfs_free_client(struct nfs_client *clp)
 {
 	dprintk("--> nfs_free_client(%u)\n", clp->rpc_ops->version);
 
-	nfs4_clear_client_minor_version(clp);
 	nfs4_shutdown_client(clp);
 
 	nfs_fscache_release_client_cookie(clp);
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 3c7f03b669fb..a1f6b4438fb1 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -560,7 +560,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	desc->entry = &my_entry;
 
 	nfs_block_sillyrename(dentry);
-	res = nfs_revalidate_mapping_nolock(inode, filp->f_mapping);
+	res = nfs_revalidate_mapping(inode, filp->f_mapping);
 	if (res < 0)
 		goto out;
 
diff --git a/fs/nfs/dns_resolve.c b/fs/nfs/dns_resolve.c
index 95e1ca765d47..3f0cd4dfddaf 100644
--- a/fs/nfs/dns_resolve.c
+++ b/fs/nfs/dns_resolve.c
@@ -36,6 +36,19 @@ struct nfs_dns_ent {
 };
 
 
+static void nfs_dns_ent_update(struct cache_head *cnew,
+		struct cache_head *ckey)
+{
+	struct nfs_dns_ent *new;
+	struct nfs_dns_ent *key;
+
+	new = container_of(cnew, struct nfs_dns_ent, h);
+	key = container_of(ckey, struct nfs_dns_ent, h);
+
+	memcpy(&new->addr, &key->addr, key->addrlen);
+	new->addrlen = key->addrlen;
+}
+
 static void nfs_dns_ent_init(struct cache_head *cnew,
 		struct cache_head *ckey)
 {
@@ -49,8 +62,7 @@ static void nfs_dns_ent_init(struct cache_head *cnew,
 	new->hostname = kstrndup(key->hostname, key->namelen, GFP_KERNEL);
 	if (new->hostname) {
 		new->namelen = key->namelen;
-		memcpy(&new->addr, &key->addr, key->addrlen);
-		new->addrlen = key->addrlen;
+		nfs_dns_ent_update(cnew, ckey);
 	} else {
 		new->namelen = 0;
 		new->addrlen = 0;
@@ -234,7 +246,7 @@ static struct cache_detail nfs_dns_resolve = {
 	.cache_show = nfs_dns_show,
 	.match = nfs_dns_match,
 	.init = nfs_dns_ent_init,
-	.update = nfs_dns_ent_init,
+	.update = nfs_dns_ent_update,
 	.alloc = nfs_dns_ent_alloc,
 };
 
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 63f2071d6445..ae8d02294e46 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -123,11 +123,11 @@ nfs_file_open(struct inode *inode, struct file *filp)
 			filp->f_path.dentry->d_parent->d_name.name,
 			filp->f_path.dentry->d_name.name);
 
+	nfs_inc_stats(inode, NFSIOS_VFSOPEN);
 	res = nfs_check_flags(filp->f_flags);
 	if (res)
 		return res;
 
-	nfs_inc_stats(inode, NFSIOS_VFSOPEN);
 	res = nfs_open(inode, filp);
 	return res;
 }
@@ -237,9 +237,9 @@ nfs_file_flush(struct file *file, fl_owner_t id)
 			dentry->d_parent->d_name.name,
 			dentry->d_name.name);
 
+	nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
 	if ((file->f_mode & FMODE_WRITE) == 0)
 		return 0;
-	nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
 
 	/* Flush writes to the server and return any errors */
 	return nfs_do_fsync(ctx, inode);
@@ -262,9 +262,11 @@ nfs_file_read(struct kiocb *iocb, const struct iovec *iov,
 		(unsigned long) count, (unsigned long) pos);
 
 	result = nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping);
-	nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, count);
-	if (!result)
+	if (!result) {
 		result = generic_file_aio_read(iocb, iov, nr_segs, pos);
+		if (result > 0)
+			nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, result);
+	}
 	return result;
 }
 
@@ -282,8 +284,11 @@ nfs_file_splice_read(struct file *filp, loff_t *ppos,
 		(unsigned long) count, (unsigned long long) *ppos);
 
 	res = nfs_revalidate_mapping(inode, filp->f_mapping);
-	if (!res)
+	if (!res) {
 		res = generic_file_splice_read(filp, ppos, pipe, count, flags);
+		if (res > 0)
+			nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, res);
+	}
 	return res;
 }
 
@@ -596,6 +601,7 @@ static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov,
 {
 	struct dentry * dentry = iocb->ki_filp->f_path.dentry;
 	struct inode * inode = dentry->d_inode;
+	unsigned long written = 0;
 	ssize_t result;
 	size_t count = iov_length(iov, nr_segs);
 
@@ -622,14 +628,18 @@ static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov,
 	if (!count)
 		goto out;
 
-	nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count);
 	result = generic_file_aio_write(iocb, iov, nr_segs, pos);
+	if (result > 0)
+		written = result;
+
 	/* Return error values for O_DSYNC and IS_SYNC() */
 	if (result >= 0 && nfs_need_sync_write(iocb->ki_filp, inode)) {
 		int err = nfs_do_fsync(nfs_file_open_context(iocb->ki_filp), inode);
 		if (err < 0)
 			result = err;
 	}
+	if (result > 0)
+		nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, written);
 out:
 	return result;
 
@@ -644,6 +654,7 @@ static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe,
 {
 	struct dentry *dentry = filp->f_path.dentry;
 	struct inode *inode = dentry->d_inode;
+	unsigned long written = 0;
 	ssize_t ret;
 
 	dprintk("NFS splice_write(%s/%s, %lu@%llu)\n",
@@ -654,14 +665,17 @@ static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe,
 	 * The combination of splice and an O_APPEND destination is disallowed.
 	 */
 
-	nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count);
-
 	ret = generic_file_splice_write(pipe, filp, ppos, count, flags);
+	if (ret > 0)
+		written = ret;
+
 	if (ret >= 0 && nfs_need_sync_write(filp, inode)) {
 		int err = nfs_do_fsync(nfs_file_open_context(filp), inode);
 		if (err < 0)
 			ret = err;
 	}
+	if (ret > 0)
+		nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, written);
 	return ret;
 }
 
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 7f9ecc46f3fb..657201acda84 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -97,18 +97,6 @@ u64 nfs_compat_user_ino64(u64 fileid)
 	return ino;
 }
 
-int nfs_write_inode(struct inode *inode, struct writeback_control *wbc)
-{
-	int ret;
-
-	ret = nfs_commit_inode(inode,
-			wbc->sync_mode == WB_SYNC_ALL ? FLUSH_SYNC : 0);
-	if (ret >= 0)
-		return 0;
-	__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
-	return ret;
-}
-
 void nfs_clear_inode(struct inode *inode)
 {
 	/*
@@ -126,16 +114,12 @@ void nfs_clear_inode(struct inode *inode)
  */
 int nfs_sync_mapping(struct address_space *mapping)
 {
-	int ret;
+	int ret = 0;
 
-	if (mapping->nrpages == 0)
-		return 0;
-	unmap_mapping_range(mapping, 0, 0, 0);
-	ret = filemap_write_and_wait(mapping);
-	if (ret != 0)
-		goto out;
-	ret = nfs_wb_all(mapping->host);
-out:
+	if (mapping->nrpages != 0) {
+		unmap_mapping_range(mapping, 0, 0, 0);
+		ret = nfs_wb_all(mapping->host);
+	}
 	return ret;
 }
 
@@ -507,17 +491,11 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
 	int need_atime = NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATIME;
 	int err;
 
-	/*
-	 * Flush out writes to the server in order to update c/mtime.
-	 *
-	 * Hold the i_mutex to suspend application writes temporarily;
-	 * this prevents long-running writing applications from blocking
-	 * nfs_wb_nocommit.
-	 */
+	/* Flush out writes to the server in order to update c/mtime.  */
 	if (S_ISREG(inode->i_mode)) {
-		mutex_lock(&inode->i_mutex);
-		nfs_wb_nocommit(inode);
-		mutex_unlock(&inode->i_mutex);
+		err = filemap_write_and_wait(inode->i_mapping);
+		if (err)
+			goto out;
 	}
 
 	/*
@@ -541,6 +519,7 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
 		generic_fillattr(inode, stat);
 		stat->ino = nfs_compat_user_ino64(NFS_FILEID(inode));
 	}
+out:
 	return err;
 }
 
@@ -616,11 +595,6 @@ void put_nfs_open_context(struct nfs_open_context *ctx)
 	__put_nfs_open_context(ctx, 0);
 }
 
-static void put_nfs_open_context_sync(struct nfs_open_context *ctx)
-{
-	__put_nfs_open_context(ctx, 1);
-}
-
 /*
  * Ensure that mmap has a recent RPC credential for use when writing out
  * shared pages
@@ -667,7 +641,7 @@ static void nfs_file_clear_open_context(struct file *filp)
 		spin_lock(&inode->i_lock);
 		list_move_tail(&ctx->list, &NFS_I(inode)->open_files);
 		spin_unlock(&inode->i_lock);
-		put_nfs_open_context_sync(ctx);
+		__put_nfs_open_context(ctx, filp->f_flags & O_DIRECT ? 0 : 1);
 	}
 }
 
@@ -775,7 +749,7 @@ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 	return __nfs_revalidate_inode(server, inode);
 }
 
-static int nfs_invalidate_mapping_nolock(struct inode *inode, struct address_space *mapping)
+static int nfs_invalidate_mapping(struct inode *inode, struct address_space *mapping)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	
@@ -796,49 +770,10 @@ static int nfs_invalidate_mapping_nolock(struct inode *inode, struct address_spa
 	return 0;
 }
 
-static int nfs_invalidate_mapping(struct inode *inode, struct address_space *mapping)
-{
-	int ret = 0;
-
-	mutex_lock(&inode->i_mutex);
-	if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_DATA) {
-		ret = nfs_sync_mapping(mapping);
-		if (ret == 0)
-			ret = nfs_invalidate_mapping_nolock(inode, mapping);
-	}
-	mutex_unlock(&inode->i_mutex);
-	return ret;
-}
-
-/**
- * nfs_revalidate_mapping_nolock - Revalidate the pagecache
- * @inode - pointer to host inode
- * @mapping - pointer to mapping
- */
-int nfs_revalidate_mapping_nolock(struct inode *inode, struct address_space *mapping)
-{
-	struct nfs_inode *nfsi = NFS_I(inode);
-	int ret = 0;
-
-	if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE)
-			|| nfs_attribute_timeout(inode) || NFS_STALE(inode)) {
-		ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode);
-		if (ret < 0)
-			goto out;
-	}
-	if (nfsi->cache_validity & NFS_INO_INVALID_DATA)
-		ret = nfs_invalidate_mapping_nolock(inode, mapping);
-out:
-	return ret;
-}
-
 /**
  * nfs_revalidate_mapping - Revalidate the pagecache
  * @inode - pointer to host inode
  * @mapping - pointer to mapping
- *
- * This version of the function will take the inode->i_mutex and attempt to
- * flush out all dirty data if it needs to invalidate the page cache.
  */
 int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
 {
@@ -1416,6 +1351,7 @@ static void init_once(void *foo)
 	INIT_LIST_HEAD(&nfsi->access_cache_inode_lru);
 	INIT_RADIX_TREE(&nfsi->nfs_page_tree, GFP_ATOMIC);
 	nfsi->npages = 0;
+	nfsi->ncommit = 0;
 	atomic_set(&nfsi->silly_count, 1);
 	INIT_HLIST_HEAD(&nfsi->silly_list);
 	init_waitqueue_head(&nfsi->waitqueue);
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 3f8881d1a050..24992f0a29f2 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -22,14 +22,14 @@
 
 #define NFSDBG_FACILITY		NFSDBG_PROC
 
-/* A wrapper to handle the EJUKEBOX error message */
+/* A wrapper to handle the EJUKEBOX and EKEYEXPIRED error messages */
 static int
 nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags)
 {
 	int res;
 	do {
 		res = rpc_call_sync(clnt, msg, flags);
-		if (res != -EJUKEBOX)
+		if (res != -EJUKEBOX && res != -EKEYEXPIRED)
 			break;
 		schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME);
 		res = -ERESTARTSYS;
@@ -42,9 +42,10 @@ nfs3_rpc_wrapper(stru