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(struct rpc_clnt *clnt, struct rpc_message *msg, int flags)
 static int
 nfs3_async_handle_jukebox(struct rpc_task *task, struct inode *inode)
 {
-	if (task->tk_status != -EJUKEBOX)
+	if (task->tk_status != -EJUKEBOX && task->tk_status != -EKEYEXPIRED)
 		return 0;
-	nfs_inc_stats(inode, NFSIOS_DELAY);
+	if (task->tk_status == -EJUKEBOX)
+		nfs_inc_stats(inode, NFSIOS_DELAY);
 	task->tk_status = 0;
 	rpc_restart_call(task);
 	rpc_delay(task, NFS_JUKEBOX_RETRY_TIME);
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 0c6fda33d66e..a187200a7aac 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -46,6 +46,7 @@ enum nfs4_client_state {
 	NFS4CLNT_DELEGRETURN,
 	NFS4CLNT_SESSION_RESET,
 	NFS4CLNT_SESSION_DRAINING,
+	NFS4CLNT_RECALL_SLOT,
 };
 
 /*
@@ -280,6 +281,7 @@ extern void nfs4_schedule_state_manager(struct nfs_client *);
 extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state);
 extern int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state);
 extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
+extern void nfs41_handle_recall_slot(struct nfs_client *clp);
 extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 84d83be25a98..eda74c42d552 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -281,6 +281,7 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 			}
 		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
+		case -EKEYEXPIRED:
 			ret = nfs4_delay(server->client, &exception->timeout);
 			if (ret != 0)
 				break;
@@ -418,7 +419,8 @@ static void nfs41_sequence_done(struct nfs_client *clp,
 			clp->cl_last_renewal = timestamp;
 		spin_unlock(&clp->cl_lock);
 		/* Check sequence flags */
-		nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
+		if (atomic_read(&clp->cl_count) > 1)
+			nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
 	}
 out:
 	/* The session may be reset by one of the error handlers. */
@@ -1163,7 +1165,7 @@ static int nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state
 	int err;
 	do {
 		err = _nfs4_do_open_reclaim(ctx, state);
-		if (err != -NFS4ERR_DELAY)
+		if (err != -NFS4ERR_DELAY && err != -EKEYEXPIRED)
 			break;
 		nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
@@ -1582,6 +1584,7 @@ static int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state
 			goto out;
 		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
+		case -EKEYEXPIRED:
 			nfs4_handle_exception(server, err, &exception);
 			err = 0;
 		}
@@ -3145,10 +3148,19 @@ static void nfs4_proc_commit_setup(struct nfs_write_data *data, struct rpc_messa
  * nfs4_proc_async_renew(): This is not one of the nfs_rpc_ops; it is a special
  * standalone procedure for queueing an asynchronous RENEW.
  */
+static void nfs4_renew_release(void *data)
+{
+	struct nfs_client *clp = data;
+
+	if (atomic_read(&clp->cl_count) > 1)
+		nfs4_schedule_state_renewal(clp);
+	nfs_put_client(clp);
+}
+
 static void nfs4_renew_done(struct rpc_task *task, void *data)
 {
-	struct nfs_client *clp = (struct nfs_client *)task->tk_msg.rpc_argp;
-	unsigned long timestamp = (unsigned long)data;
+	struct nfs_client *clp = data;
+	unsigned long timestamp = task->tk_start;
 
 	if (task->tk_status < 0) {
 		/* Unless we're shutting down, schedule state recovery! */
@@ -3164,6 +3176,7 @@ static void nfs4_renew_done(struct rpc_task *task, void *data)
 
 static const struct rpc_call_ops nfs4_renew_ops = {
 	.rpc_call_done = nfs4_renew_done,
+	.rpc_release = nfs4_renew_release,
 };
 
 int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred)
@@ -3174,8 +3187,10 @@ int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred)
 		.rpc_cred	= cred,
 	};
 
+	if (!atomic_inc_not_zero(&clp->cl_count))
+		return -EIO;
 	return rpc_call_async(clp->cl_rpcclient, &msg, RPC_TASK_SOFT,
-			&nfs4_renew_ops, (void *)jiffies);
+			&nfs4_renew_ops, clp);
 }
 
 int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred)
@@ -3452,6 +3467,7 @@ _nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 			if (server)
 				nfs_inc_server_stats(server, NFSIOS_DELAY);
 		case -NFS4ERR_GRACE:
+		case -EKEYEXPIRED:
 			rpc_delay(task, NFS4_POLL_RETRY_MAX);
 			task->tk_status = 0;
 			return -EAGAIN;
@@ -3564,6 +3580,7 @@ int nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred)
 			case -NFS4ERR_RESOURCE:
 				/* The IBM lawyers misread another document! */
 			case -NFS4ERR_DELAY:
+			case -EKEYEXPIRED:
 				err = nfs4_delay(clp->cl_rpcclient, &timeout);
 		}
 	} while (err == 0);
@@ -4179,7 +4196,7 @@ static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request
 		if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0)
 			return 0;
 		err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_RECLAIM);
-		if (err != -NFS4ERR_DELAY)
+		if (err != -NFS4ERR_DELAY && err != -EKEYEXPIRED)
 			break;
 		nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
@@ -4204,6 +4221,7 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request
 			goto out;
 		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
+		case -EKEYEXPIRED:
 			nfs4_handle_exception(server, err, &exception);
 			err = 0;
 		}
@@ -4355,6 +4373,7 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 				err = 0;
 				goto out;
 			case -NFS4ERR_DELAY:
+			case -EKEYEXPIRED:
 				break;
 		}
 		err = nfs4_handle_exception(server, err, &exception);
@@ -4500,7 +4519,7 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 
 		status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
 
-		if (status != NFS4ERR_CLID_INUSE)
+		if (status != -NFS4ERR_CLID_INUSE)
 			break;
 
 		if (signalled())
@@ -4554,6 +4573,7 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
 	switch (task->tk_status) {
 	case -NFS4ERR_DELAY:
 	case -NFS4ERR_GRACE:
+	case -EKEYEXPIRED:
 		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
 		rpc_delay(task, NFS4_POLL_RETRY_MIN);
 		task->tk_status = 0;
@@ -4611,26 +4631,32 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 /*
  * Reset a slot table
  */
-static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, int max_slots,
-		int old_max_slots, int ivalue)
+static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, u32 max_reqs,
+				 int ivalue)
 {
+	struct nfs4_slot *new = NULL;
 	int i;
 	int ret = 0;
 
-	dprintk("--> %s: max_reqs=%u, tbl %p\n", __func__, max_slots, tbl);
+	dprintk("--> %s: max_reqs=%u, tbl->max_slots %d\n", __func__,
+		max_reqs, tbl->max_slots);
 
-	/*
-	 * Until we have dynamic slot table adjustment, insist
-	 * upon the same slot table size
-	 */
-	if (max_slots != old_max_slots) {
-		dprintk("%s reset slot table does't match old\n",
-			__func__);
-		ret = -EINVAL; /*XXX NFS4ERR_REQ_TOO_BIG ? */
-		goto out;
+	/* Does the newly negotiated max_reqs match the existing slot table? */
+	if (max_reqs != tbl->max_slots) {
+		ret = -ENOMEM;
+		new = kmalloc(max_reqs * sizeof(struct nfs4_slot),
+			      GFP_KERNEL);
+		if (!new)
+			goto out;
+		ret = 0;
+		kfree(tbl->slots);
 	}
 	spin_lock(&tbl->slot_tbl_lock);
-	for (i = 0; i < max_slots; ++i)
+	if (new) {
+		tbl->slots = new;
+		tbl->max_slots = max_reqs;
+	}
+	for (i = 0; i < tbl->max_slots; ++i)
 		tbl->slots[i].seq_nr = ivalue;
 	spin_unlock(&tbl->slot_tbl_lock);
 	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
@@ -4648,16 +4674,12 @@ static int nfs4_reset_slot_tables(struct nfs4_session *session)
 	int status;
 
 	status = nfs4_reset_slot_table(&session->fc_slot_table,
-			session->fc_attrs.max_reqs,
-			session->fc_slot_table.max_slots,
-			1);
+			session->fc_attrs.max_reqs, 1);
 	if (status)
 		return status;
 
 	status = nfs4_reset_slot_table(&session->bc_slot_table,
-			session->bc_attrs.max_reqs,
-			session->bc_slot_table.max_slots,
-			0);
+			session->bc_attrs.max_reqs, 0);
 	return status;
 }
 
@@ -4798,16 +4820,14 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args)
 	args->fc_attrs.headerpadsz = 0;
 	args->fc_attrs.max_rqst_sz = mxrqst_sz;
 	args->fc_attrs.max_resp_sz = mxresp_sz;
-	args->fc_attrs.max_resp_sz_cached = mxresp_sz;
 	args->fc_attrs.max_ops = NFS4_MAX_OPS;
 	args->fc_attrs.max_reqs = session->clp->cl_rpcclient->cl_xprt->max_reqs;
 
 	dprintk("%s: Fore Channel : max_rqst_sz=%u max_resp_sz=%u "
-		"max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n",
+		"max_ops=%u max_reqs=%u\n",
 		__func__,
 		args->fc_attrs.max_rqst_sz, args->fc_attrs.max_resp_sz,
-		args->fc_attrs.max_resp_sz_cached, args->fc_attrs.max_ops,
-		args->fc_attrs.max_reqs);
+		args->fc_attrs.max_ops, args->fc_attrs.max_reqs);
 
 	/* Back channel attributes */
 	args->bc_attrs.headerpadsz = 0;
@@ -5016,7 +5036,16 @@ static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred)
 				       &res, args.sa_cache_this, 1);
 }
 
-void nfs41_sequence_call_done(struct rpc_task *task, void *data)
+static void nfs41_sequence_release(void *data)
+{
+	struct nfs_client *clp = (struct nfs_client *)data;
+
+	if (atomic_read(&clp->cl_count) > 1)
+		nfs4_schedule_state_renewal(clp);
+	nfs_put_client(clp);
+}
+
+static void nfs41_sequence_call_done(struct rpc_task *task, void *data)
 {
 	struct nfs_client *clp = (struct nfs_client *)data;
 
@@ -5024,6 +5053,8 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data)
 
 	if (task->tk_status < 0) {
 		dprintk("%s ERROR %d\n", __func__, task->tk_status);
+		if (atomic_read(&clp->cl_count) == 1)
+			goto out;
 
 		if (_nfs4_async_handle_error(task, NULL, clp, NULL)
 								== -EAGAIN) {
@@ -5032,7 +5063,7 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data)
 		}
 	}
 	dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred);
-
+out:
 	kfree(task->tk_msg.rpc_argp);
 	kfree(task->tk_msg.rpc_resp);
 
@@ -5057,6 +5088,7 @@ static void nfs41_sequence_prepare(struct rpc_task *task, void *data)
 static const struct rpc_call_ops nfs41_sequence_ops = {
 	.rpc_call_done = nfs41_sequence_call_done,
 	.rpc_call_prepare = nfs41_sequence_prepare,
+	.rpc_release = nfs41_sequence_release,
 };
 
 static int nfs41_proc_async_sequence(struct nfs_client *clp,
@@ -5069,12 +5101,13 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp,
 		.rpc_cred = cred,
 	};
 
+	if (!atomic_inc_not_zero(&clp->cl_count))
+		return -EIO;
 	args = kzalloc(sizeof(*args), GFP_KERNEL);
-	if (!args)
-		return -ENOMEM;
 	res = kzalloc(sizeof(*res), GFP_KERNEL);
-	if (!res) {
+	if (!args || !res) {
 		kfree(args);
+		nfs_put_client(clp);
 		return -ENOMEM;
 	}
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 0156c01c212c..d87f10327b72 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -36,11 +36,6 @@
  * as an rpc_task, not a real kernel thread, so it always runs in rpciod's
  * context.  There is one renewd per nfs_server.
  *
- * TODO: If the send queue gets backlogged (e.g., if the server goes down),
- * we will keep filling the queue with periodic RENEW requests.  We need a
- * mechanism for ensuring that if renewd successfully sends off a request,
- * then it only wakes up when the request is finished.  Maybe use the
- * child task framework of the RPC layer?
  */
 
 #include <linux/mm.h>
@@ -63,7 +58,7 @@ nfs4_renew_state(struct work_struct *work)
 	struct nfs_client *clp =
 		container_of(work, struct nfs_client, cl_renewd.work);
 	struct rpc_cred *cred;
-	long lease, timeout;
+	long lease;
 	unsigned long last, now;
 
 	ops = nfs4_state_renewal_ops[clp->cl_minorversion];
@@ -75,7 +70,6 @@ nfs4_renew_state(struct work_struct *work)
 	lease = clp->cl_lease_time;
 	last = clp->cl_last_renewal;
 	now = jiffies;
-	timeout = (2 * lease) / 3 + (long)last - (long)now;
 	/* Are we close to a lease timeout? */
 	if (time_after(now, last + lease/3)) {
 		cred = ops->get_state_renewal_cred_locked(clp);
@@ -90,19 +84,15 @@ nfs4_renew_state(struct work_struct *work)
 			/* Queue an asynchronous RENEW. */
 			ops->sched_state_renewal(clp, cred);
 			put_rpccred(cred);
+			goto out_exp;
 		}
-		timeout = (2 * lease) / 3;
-		spin_lock(&clp->cl_lock);
-	} else
+	} else {
 		dprintk("%s: failed to call renewd. Reason: lease not expired \n",
 				__func__);
-	if (timeout < 5 * HZ)    /* safeguard */
-		timeout = 5 * HZ;
-	dprintk("%s: requeueing work. Lease period = %ld\n",
-			__func__, (timeout + HZ - 1) / HZ);
-	cancel_delayed_work(&clp->cl_renewd);
-	schedule_delayed_work(&clp->cl_renewd, timeout);
-	spin_unlock(&clp->cl_lock);
+		spin_unlock(&clp->cl_lock);
+	}
+	nfs4_schedule_state_renewal(clp);
+out_exp:
 	nfs_expire_unreferenced_delegations(clp);
 out:
 	dprintk("%s: done\n", __func__);
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index c1e2733f4fa4..6c5ed51f105e 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1249,26 +1249,65 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 }
 
 #ifdef CONFIG_NFS_V4_1
+void nfs41_handle_recall_slot(struct nfs_client *clp)
+{
+	set_bit(NFS4CLNT_RECALL_SLOT, &clp->cl_state);
+	nfs4_schedule_state_recovery(clp);
+}
+
+static void nfs4_reset_all_state(struct nfs_client *clp)
+{
+	if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) {
+		clp->cl_boot_time = CURRENT_TIME;
+		nfs4_state_start_reclaim_nograce(clp);
+		nfs4_schedule_state_recovery(clp);
+	}
+}
+
+static void nfs41_handle_server_reboot(struct nfs_client *clp)
+{
+	if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) {
+		nfs4_state_start_reclaim_reboot(clp);
+		nfs4_schedule_state_recovery(clp);
+	}
+}
+
+static void nfs41_handle_state_revoked(struct nfs_client *clp)
+{
+	/* Temporary */
+	nfs4_reset_all_state(clp);
+}
+
+static void nfs41_handle_recallable_state_revoked(struct nfs_client *clp)
+{
+	/* This will need to handle layouts too */
+	nfs_expire_all_delegations(clp);
+}
+
+static void nfs41_handle_cb_path_down(struct nfs_client *clp)
+{
+	nfs_expire_all_delegations(clp);
+	if (test_and_set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state) == 0)
+		nfs4_schedule_state_recovery(clp);
+}
+
 void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags)
 {
 	if (!flags)
 		return;
-	else if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED) {
-		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		nfs4_state_start_reclaim_reboot(clp);
-		nfs4_schedule_state_recovery(clp);
-	} else if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED |
+	else if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED)
+		nfs41_handle_server_reboot(clp);
+	else if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED |
 			    SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED |
 			    SEQ4_STATUS_ADMIN_STATE_REVOKED |
-			    SEQ4_STATUS_RECALLABLE_STATE_REVOKED |
-			    SEQ4_STATUS_LEASE_MOVED)) {
-		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		nfs4_state_start_reclaim_nograce(clp);
-		nfs4_schedule_state_recovery(clp);
-	} else if (flags & (SEQ4_STATUS_CB_PATH_DOWN |
+			    SEQ4_STATUS_LEASE_MOVED))
+		nfs41_handle_state_revoked(clp);
+	else if (flags & SEQ4_STATUS_RECALLABLE_STATE_REVOKED)
+		nfs41_handle_recallable_state_revoked(clp);
+	else if (flags & (SEQ4_STATUS_CB_PATH_DOWN |
 			    SEQ4_STATUS_BACKCHANNEL_FAULT |
 			    SEQ4_STATUS_CB_PATH_DOWN_SESSION))
-		nfs_expire_all_delegations(clp);
+		nfs41_handle_cb_path_down(clp);
 }
 
 static int nfs4_reset_session(struct nfs_client *clp)
@@ -1285,23 +1324,52 @@ static int nfs4_reset_session(struct nfs_client *clp)
 
 	memset(clp->cl_session->sess_id.data, 0, NFS4_MAX_SESSIONID_LEN);
 	status = nfs4_proc_create_session(clp);
-	if (status)
+	if (status) {
 		status = nfs4_recovery_handle_error(clp, status);
+		goto out;
+	}
+	/* create_session negotiated new slot table */
+	clear_bit(NFS4CLNT_RECALL_SLOT, &clp->cl_state);
 
-out:
-	/*
-	 * Let the state manager reestablish state
-	 */
-	if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) &&
-	    status == 0)
+	 /* Let the state manager reestablish state */
+	if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 		nfs41_setup_state_renewal(clp);
-
+out:
 	return status;
 }
 
+static int nfs4_recall_slot(struct nfs_client *clp)
+{
+	struct nfs4_slot_table *fc_tbl = &clp->cl_session->fc_slot_table;
+	struct nfs4_channel_attrs *fc_attrs = &clp->cl_session->fc_attrs;
+	struct nfs4_slot *new, *old;
+	int i;
+
+	nfs4_begin_drain_session(clp);
+	new = kmalloc(fc_tbl->target_max_slots * sizeof(struct nfs4_slot),
+		      GFP_KERNEL);
+        if (!new)
+		return -ENOMEM;
+
+	spin_lock(&fc_tbl->slot_tbl_lock);
+	for (i = 0; i < fc_tbl->target_max_slots; i++)
+		new[i].seq_nr = fc_tbl->slots[i].seq_nr;
+	old = fc_tbl->slots;
+	fc_tbl->slots = new;
+	fc_tbl->max_slots = fc_tbl->target_max_slots;
+	fc_tbl->target_max_slots = 0;
+	fc_attrs->max_reqs = fc_tbl->max_slots;
+	spin_unlock(&fc_tbl->slot_tbl_lock);
+
+	kfree(old);
+	nfs4_end_drain_session(clp);
+	return 0;
+}
+
 #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; }
+static int nfs4_recall_slot(struct nfs_client *clp) { return 0; }
 #endif /* CONFIG_NFS_V4_1 */
 
 /* Set NFS4CLNT_LEASE_EXPIRED for all v4.0 errors and for recoverable errors
@@ -1314,6 +1382,7 @@ static void nfs4_set_lease_expired(struct nfs_client *clp, int status)
 		case -NFS4ERR_DELAY:
 		case -NFS4ERR_CLID_INUSE:
 		case -EAGAIN:
+		case -EKEYEXPIRED:
 			break;
 
 		case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery
@@ -1397,6 +1466,15 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			nfs_client_return_marked_delegations(clp);
 			continue;
 		}
+		/* Recall session slots */
+		if (test_and_clear_bit(NFS4CLNT_RECALL_SLOT, &clp->cl_state)
+		   && nfs4_has_session(clp)) {
+			status = nfs4_recall_slot(clp);
+			if (status < 0)
+				goto out_error;
+			continue;
+		}
+
 
 		nfs4_clear_state_manager_bit(clp);
 		/* Did we race with an attempt to give us more work? */
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 5cd5184b56db..4d338be492cb 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1578,6 +1578,14 @@ static void encode_create_session(struct xdr_stream *xdr,
 	char machine_name[NFS4_MAX_MACHINE_NAME_LEN];
 	uint32_t len;
 	struct nfs_client *clp = args->client;
+	u32 max_resp_sz_cached;
+
+	/*
+	 * Assumes OPEN is the biggest non-idempotent compound.
+	 * 2 is the verifier.
+	 */
+	max_resp_sz_cached = (NFS4_dec_open_sz + RPC_REPHDRSIZE +
+			      RPC_MAX_AUTH_SIZE + 2) * XDR_UNIT;
 
 	len = scnprintf(machine_name, sizeof(machine_name), "%s",
 			clp->cl_ipaddr);
@@ -1592,7 +1600,7 @@ static void encode_create_session(struct xdr_stream *xdr,
 	*p++ = cpu_to_be32(args->fc_attrs.headerpadsz);	/* header padding size */
 	*p++ = cpu_to_be32(args->fc_attrs.max_rqst_sz);	/* max req size */
 	*p++ = cpu_to_be32(args->fc_attrs.max_resp_sz);	/* max resp size */
-	*p++ = cpu_to_be32(args->fc_attrs.max_resp_sz_cached);	/* Max resp sz cached */
+	*p++ = cpu_to_be32(max_resp_sz_cached);		/* Max resp sz cached */
 	*p++ = cpu_to_be32(args->fc_attrs.max_ops);	/* max operations */
 	*p++ = cpu_to_be32(args->fc_attrs.max_reqs);	/* max requests */
 	*p++ = cpu_to_be32(0);				/* rdmachannel_attrs */
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index ef583854d8d0..c752d944fe9e 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -47,6 +47,39 @@
 #define NFSDBG_FACILITY		NFSDBG_PROC
 
 /*
+ * wrapper to handle the -EKEYEXPIRED error message. This should generally
+ * only happen if using krb5 auth and a user's TGT expires. NFSv2 doesn't
+ * support the NFSERR_JUKEBOX error code, but we handle this situation in the
+ * same way that we handle that error with NFSv3.
+ */
+static int
+nfs_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags)
+{
+	int res;
+	do {
+		res = rpc_call_sync(clnt, msg, flags);
+		if (res != -EKEYEXPIRED)
+			break;
+		schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME);
+		res = -ERESTARTSYS;
+	} while (!fatal_signal_pending(current));
+	return res;
+}
+
+#define rpc_call_sync(clnt, msg, flags)	nfs_rpc_wrapper(clnt, msg, flags)
+
+static int
+nfs_async_handle_expired_key(struct rpc_task *task)
+{
+	if (task->tk_status != -EKEYEXPIRED)
+		return 0;
+	task->tk_status = 0;
+	rpc_restart_call(task);
+	rpc_delay(task, NFS_JUKEBOX_RETRY_TIME);
+	return 1;
+}
+
+/*
  * Bare-bones access to getattr: this is for nfs_read_super.
  */
 static int
@@ -307,6 +340,8 @@ nfs_proc_unlink_setup(struct rpc_message *msg, struct inode *dir)
 
 static int nfs_proc_unlink_done(struct rpc_task *task, struct inode *dir)
 {
+	if (nfs_async_handle_expired_key(task))
+		return 0;
 	nfs_mark_for_revalidate(dir);
 	return 1;
 }
@@ -560,6 +595,9 @@ nfs_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
 
 static int nfs_read_done(struct rpc_task *task, struct nfs_read_data *data)
 {
+	if (nfs_async_handle_expired_key(task))
+		return -EAGAIN;
+
 	nfs_invalidate_atime(data->inode);
 	if (task->tk_status >= 0) {
 		nfs_refresh_inode(data->inode, data->res.fattr);
@@ -579,6 +617,9 @@ static void nfs_proc_read_setup(struct nfs_read_data *data, struct rpc_message *
 
 static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data)
 {
+	if (nfs_async_handle_expired_key(task))
+		return -EAGAIN;
+
 	if (task->tk_status >= 0)
 		nfs_post_op_update_inode_force_wcc(data->inode, data->res.fattr);
 	return 0;
diff --git a/fs/nfs/symlink.c b/fs/nfs/symlink.c
index 412738dbfbc7..2ea9e5c27e55 100644
--- a/fs/nfs/symlink.c
+++ b/fs/nfs/symlink.c
@@ -50,7 +50,7 @@ static void *nfs_follow_link(struct dentry *dentry, struct nameidata *nd)
 	struct page *page;
 	void *err;
 
-	err = ERR_PTR(nfs_revalidate_mapping_nolock(inode, inode->i_mapping));
+	err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping));
 	if (err)
 		goto read_failed;
 	page = read_cache_page(&inode->i_data, 0,
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index d63d964a0392..53ff70e23993 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -438,6 +438,7 @@ nfs_mark_request_commit(struct nfs_page *req)
 	radix_tree_tag_set(&nfsi->nfs_page_tree,
 			req->wb_index,
 			NFS_PAGE_TAG_COMMIT);
+	nfsi->ncommit++;
 	spin_unlock(&inode->i_lock);
 	inc_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
 	inc_bdi_stat(req->wb_page->mapping->backing_dev_info, BDI_RECLAIMABLE);
@@ -501,57 +502,6 @@ int nfs_reschedule_unstable_write(struct nfs_page *req)
 }
 #endif
 
-/*
- * Wait for a request to complete.
- *
- * Interruptible by fatal signals only.
- */
-static int nfs_wait_on_requests_locked(struct inode *inode, pgoff_t idx_start, unsigned int npages)
-{
-	struct nfs_inode *nfsi = NFS_I(inode);
-	struct nfs_page *req;
-	pgoff_t idx_end, next;
-	unsigned int		res = 0;
-	int			error;
-
-	if (npages == 0)
-		idx_end = ~0;
-	else
-		idx_end = idx_start + npages - 1;
-
-	next = idx_start;
-	while (radix_tree_gang_lookup_tag(&nfsi->nfs_page_tree, (void **)&req, next, 1, NFS_PAGE_TAG_LOCKED)) {
-		if (req->wb_index > idx_end)
-			break;
-
-		next = req->wb_index + 1;
-		BUG_ON(!NFS_WBACK_BUSY(req));
-
-		kref_get(&req->wb_kref);
-		spin_unlock(&inode->i_lock);
-		error = nfs_wait_on_request(req);
-		nfs_release_request(req);
-		spin_lock(&inode->i_lock);
-		if (error < 0)
-			return error;
-		res++;
-	}
-	return res;
-}
-
-static void nfs_cancel_commit_list(struct list_head *head)
-{
-	struct nfs_page *req;
-
-	while(!list_empty(head)) {
-		req = nfs_list_entry(head->next);
-		nfs_list_remove_request(req);
-		nfs_clear_request_commit(req);
-		nfs_inode_remove_request(req);
-		nfs_unlock_request(req);
-	}
-}
-
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 static int
 nfs_need_commit(struct nfs_inode *nfsi)
@@ -573,11 +523,17 @@ static int
 nfs_scan_commit(struct inode *inode, struct list_head *dst, pgoff_t idx_start, unsigned int npages)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
+	int ret;
 
 	if (!nfs_need_commit(nfsi))
 		return 0;
 
-	return nfs_scan_list(nfsi, dst, idx_start, npages, NFS_PAGE_TAG_COMMIT);
+	ret = nfs_scan_list(nfsi, dst, idx_start, npages, NFS_PAGE_TAG_COMMIT);
+	if (ret > 0)
+		nfsi->ncommit -= ret;
+	if (nfs_need_commit(NFS_I(inode)))
+		__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
+	return ret;
 }
 #else
 static inline int nfs_need_commit(struct nfs_inode *nfsi)
@@ -642,9 +598,10 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode,
 		spin_lock(&inode->i_lock);
 	}
 
-	if (nfs_clear_request_commit(req))
-		radix_tree_tag_clear(&NFS_I(inode)->nfs_page_tree,
-				req->wb_index, NFS_PAGE_TAG_COMMIT);
+	if (nfs_clear_request_commit(req) &&
+			radix_tree_tag_clear(&NFS_I(inode)->nfs_page_tree,
+				req->wb_index, NFS_PAGE_TAG_COMMIT) != NULL)
+		NFS_I(inode)->ncommit--;
 
 	/* Okay, the request matches. Update the region */
 	if (offset < req->wb_offset) {
@@ -1391,7 +1348,7 @@ static const struct rpc_call_ops nfs_commit_ops = {
 	.rpc_release = nfs_commit_release,
 };
 
-int nfs_commit_inode(struct inode *inode, int how)
+static int nfs_commit_inode(struct inode *inode, int how)
 {
 	LIST_HEAD(head);
 	int res;
@@ -1406,92 +1363,51 @@ int nfs_commit_inode(struct inode *inode, int how)
 	}
 	return res;
 }
-#else
-static inline int nfs_commit_list(struct inode *inode, struct list_head *head, int how)
-{
-	return 0;
-}
-#endif
 
-long nfs_sync_mapping_wait(struct address_space *mapping, struct writeback_control *wbc, int how)
+static int nfs_commit_unstable_pages(struct inode *inode, struct writeback_control *wbc)
 {
-	struct inode *inode = mapping->host;
-	pgoff_t idx_start, idx_end;
-	unsigned int npages = 0;
-	LIST_HEAD(head);
-	int nocommit = how & FLUSH_NOCOMMIT;
-	long pages, ret;
-
-	/* FIXME */
-	if (wbc->range_cyclic)
-		idx_start = 0;
-	else {
-		idx_start = wbc->range_start >> PAGE_CACHE_SHIFT;
-		idx_end = wbc->range_end >> PAGE_CACHE_SHIFT;
-		if (idx_end > idx_start) {
-			pgoff_t l_npages = 1 + idx_end - idx_start;
-			npages = l_npages;
-			if (sizeof(npages) != sizeof(l_npages) &&
-					(pgoff_t)npages != l_npages)
-				npages = 0;
+	struct nfs_inode *nfsi = NFS_I(inode);
+	int flags = FLUSH_SYNC;
+	int ret = 0;
+
+	/* Don't commit yet if this is a non-blocking flush and there are
+	 * lots of outstanding writes for this mapping.
+	 */
+	if (wbc->sync_mode == WB_SYNC_NONE &&
+	    nfsi->ncommit <= (nfsi->npages >> 1))
+		goto out_mark_dirty;
+
+	if (wbc->nonblocking || wbc->for_background)
+		flags = 0;
+	ret = nfs_commit_inode(inode, flags);
+	if (ret >= 0) {
+		if (wbc->sync_mode == WB_SYNC_NONE) {
+			if (ret < wbc->nr_to_write)
+				wbc->nr_to_write -= ret;
+			else
+				wbc->nr_to_write = 0;
 		}
+		return 0;
 	}
-	how &= ~FLUSH_NOCOMMIT;
-	spin_lock(&inode->i_lock);
-	do {
-		ret = nfs_wait_on_requests_locked(inode, idx_start, npages);
-		if (ret != 0)
-			continue;
-		if (nocommit)
-			break;
-		pages = nfs_scan_commit(inode, &head, idx_start, npages);
-		if (pages == 0)
-			break;
-		if (how & FLUSH_INVALIDATE) {
-			spin_unlock(&inode->i_lock);
-			nfs_cancel_commit_list(&head);
-			ret = pages;
-			spin_lock(&inode->i_lock);
-			continue;
-		}
-		pages += nfs_scan_commit(inode, &head, 0, 0);
-		spin_unlock(&inode->i_lock);
-		ret = nfs_commit_list(inode, &head, how);
-		spin_lock(&inode->i_lock);
-
-	} while (ret >= 0);
-	spin_unlock(&inode->i_lock);
+out_mark_dirty:
+	__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
 	return ret;
 }
-
-static int __nfs_write_mapping(struct address_space *mapping, struct writeback_control *wbc, int how)
+#else
+static int nfs_commit_inode(struct inode *inode, int how)
 {
-	int ret;
-
-	ret = nfs_writepages(mapping, wbc);
-	if (ret < 0)
-		goto out;
-	ret = nfs_sync_mapping_wait(mapping, wbc, how);
-	if (ret < 0)
-		goto out;
 	return 0;
-out:
-	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
-	return ret;
 }
 
-/* Two pass sync: first using WB_SYNC_NONE, then WB_SYNC_ALL */
-static int nfs_write_mapping(struct address_space *mapping, int how)
+static int nfs_commit_unstable_pages(struct inode *inode, struct writeback_control *wbc)
 {
-	struct writeback_control wbc = {
-		.bdi = mapping->backing_dev_info,
-		.sync_mode = WB_SYNC_ALL,
-		.nr_to_write = LONG_MAX,
-		.range_start = 0,
-		.range_end = LLONG_MAX,
-	};
+	return 0;
+}
+#endif
 
-	return __nfs_write_mapping(mapping, &wbc, how);
+int nfs_write_inode(struct inode *inode, struct writeback_control *wbc)
+{
+	return nfs_commit_unstable_pages(inode, wbc);
 }
 
 /*
@@ -1499,37 +1415,26 @@ static int nfs_write_mapping(struct address_space *mapping, int how)
  */
 int nfs_wb_all(struct inode *inode)
 {
-	return nfs_write_mapping(inode->i_mapping, 0);
-}
+	struct writeback_control wbc = {
+		.sync_mode = WB_SYNC_ALL,
+		.nr_to_write = LONG_MAX,
+		.range_start = 0,
+		.range_end = LLONG_MAX,
+	};
 
-int nfs_wb_nocommit(struct inode *inode)
-{
-	return nfs_write_mapping(inode->i_mapping, FLUSH_NOCOMMIT);
+	return sync_inode(inode, &wbc);
 }
 
 int nfs_wb_page_cancel(struct inode *inode, struct page *page)
 {
 	struct nfs_page *req;
-	loff_t range_start = page_offset(page);
-	loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1);
-	struct writeback_control wbc = {
-		.bdi = page->mapping->backing_dev_info,
-		.sync_mode = WB_SYNC_ALL,
-		.nr_to_write = LONG_MAX,
-		.range_start = range_start,
-		.range_end = range_end,
-	};
 	int ret = 0;
 
 	BUG_ON(!PageLocked(page));
 	for (;;) {
 		req = nfs_page_find_request(page);
 		if (req == NULL)
-			goto out;
-		if (test_bit(PG_CLEAN, &req->wb_flags)) {
-			nfs_release_request(req);
 			break;
-		}
 		if (nfs_lock_request_dontget(req)) {
 			nfs_inode_remove_request(req);
 			/*
@@ -1543,54 +1448,54 @@ int nfs_wb_page_cancel(struct inode *inode, struct page *page)
 		ret = nfs_wait_on_request(req);
 		nfs_release_request(req);
 		if (ret < 0)
-			goto out;
+			break;
 	}
-	if (!PagePrivate(page))
-		return 0;
-	ret = nfs_sync_mapping_wait(page->mapping, &wbc, FLUSH_INVALIDATE);
-out:
 	return ret;
 }
 
-static int nfs_wb_page_priority(struct inode *inode, struct page *page,
-				int how)
+/*
+ * Write back all requests on one page - we do this before reading it.
+ */
+int nfs_wb_page(struct inode *inode, struct page *page)
 {
 	loff_t range_start = page_offset(page);
 	loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1);
 	struct writeback_control wbc = {
-		.bdi = page->mapping->backing_dev_info,
 		.sync_mode = WB_SYNC_ALL,
-		.nr_to_write = LONG_MAX,
+		.nr_to_write = 0,
 		.range_start = range_start,
 		.range_end = range_end,
 	};
+	struct nfs_page *req;
+	int need_commit;
 	int ret;
 
-	do {
+	while(PagePrivate(page)) {
 		if (clear_page_dirty_for_io(page)) {
 			ret = nfs_writepage_locked(page, &wbc);
 			if (ret < 0)
 				goto out_error;
-		} else if (!PagePrivate(page))
+		}
+		req = nfs_find_and_lock_request(page);
+		if (!req)
 			break;
-		ret = nfs_sync_mapping_wait(page->mapping, &wbc, how);
-		if (ret < 0)
+		if (IS_ERR(req)) {
+			ret = PTR_ERR(req);
 			goto out_error;
-	} while (PagePrivate(page));
+		}
+		need_commit = test_bit(PG_CLEAN, &req->wb_flags);
+		nfs_clear_page_tag_locked(req);
+		if (need_commit) {
+			ret = nfs_commit_inode(inode, FLUSH_SYNC);
+			if (ret < 0)
+				goto out_error;
+		}
+	}
 	return 0;
 out_error:
-	__mark_inode_dirty(inode, I_DIRTY_PAGES);
 	return ret;
 }
 
-/*
- * Write back all requests on one page - we do this before reading it.
- */
-int nfs_wb_page(struct inode *inode, struct page* page)
-{
-	return nfs_wb_page_priority(inode, page, FLUSH_STABLE);
-}
-
 #ifdef CONFIG_MIGRATION
 int nfs_migrate_page(struct address_space *mapping, struct page *newpage,
 		struct page *page)