nfs4state.c

/*
 *  fs/nfs/nfs4state.c
 *
 *  Client-side XDR for NFSv4.
 *
 *  Copyright (c) 2002 The Regents of the University of Michigan.
 *  All rights reserved.
 *
 *  Kendrick Smith <kmsmith@umich.edu>
 *
 *  Redistribution and use in source and binary forms, with or without
 *  modification, are permitted provided that the following conditions
 *  are met:
 *
 *  1. Redistributions of source code must retain the above copyright
 *     notice, this list of conditions and the following disclaimer.
 *  2. Redistributions in binary form must reproduce the above copyright
 *     notice, this list of conditions and the following disclaimer in the
 *     documentation and/or other materials provided with the distribution.
 *  3. Neither the name of the University nor the names of its
 *     contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission.
 *
 *  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
 *  WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
 *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 *  DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 *  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 *  CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 *  SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
 *  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
 *  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
 *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
 *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 *
 * Implementation of the NFSv4 state model.  For the time being,
 * this is minimal, but will be made much more complex in a
 * subsequent patch.
 */

#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/fs.h>
#include <linux/nfs_fs.h>
#include <linux/nfs_idmap.h>
#include <linux/kthread.h>
#include <linux/module.h>
#include <linux/random.h>
#include <linux/ratelimit.h>
#include <linux/workqueue.h>
#include <linux/bitops.h>
#include <linux/jiffies.h>

#include <linux/sunrpc/clnt.h>

#include "nfs4_fs.h"
#include "callback.h"
#include "delegation.h"
#include "internal.h"
#include "pnfs.h"
#include "netns.h"

#define NFSDBG_FACILITY		NFSDBG_STATE

#define OPENOWNER_POOL_SIZE	8

const nfs4_stateid zero_stateid;
static DEFINE_MUTEX(nfs_clid_init_mutex);
static LIST_HEAD(nfs4_clientid_list);

int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
{
	struct nfs4_setclientid_res clid = {
		.clientid = clp->cl_clientid,
		.confirm = clp->cl_confirm,
	};
	unsigned short port;
	int status;
	struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id);

	if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state))
		goto do_confirm;
	port = nn->nfs_callback_tcpport;
	if (clp->cl_addr.ss_family == AF_INET6)
		port = nn->nfs_callback_tcpport6;

	status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid);
	if (status != 0)
		goto out;
	clp->cl_clientid = clid.clientid;
	clp->cl_confirm = clid.confirm;
	set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
do_confirm:
	status = nfs4_proc_setclientid_confirm(clp, &clid, cred);
	if (status != 0)
		goto out;
	clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
	nfs4_schedule_state_renewal(clp);
out:
	return status;
}

/**
 * nfs40_discover_server_trunking - Detect server IP address trunking (mv0)
 *
 * @clp: nfs_client under test
 * @result: OUT: found nfs_client, or clp
 * @cred: credential to use for trunking test
 *
 * Returns zero, a negative errno, or a negative NFS4ERR status.
 * If zero is returned, an nfs_client pointer is planted in
 * "result".
 *
 * Note: The returned client may not yet be marked ready.
 */
int nfs40_discover_server_trunking(struct nfs_client *clp,
				   struct nfs_client **result,
				   struct rpc_cred *cred)
{
	struct nfs4_setclientid_res clid = {
		.clientid = clp->cl_clientid,
		.confirm = clp->cl_confirm,
	};
	struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id);
	unsigned short port;
	int status;

	port = nn->nfs_callback_tcpport;
	if (clp->cl_addr.ss_family == AF_INET6)
		port = nn->nfs_callback_tcpport6;

	status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid);
	if (status != 0)
		goto out;
	clp->cl_clientid = clid.clientid;
	clp->cl_confirm = clid.confirm;

	status = nfs40_walk_client_list(clp, result, cred);
	switch (status) {
	case -NFS4ERR_STALE_CLIENTID:
		set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
	case 0:
		/* Sustain the lease, even if it's empty.  If the clientid4
		 * goes stale it's of no use for trunking discovery. */
		nfs4_schedule_state_renewal(*result);
		break;
	}

out:
	return status;
}

struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp)
{
	struct rpc_cred *cred = NULL;

	if (clp->cl_machine_cred != NULL)
		cred = get_rpccred(clp->cl_machine_cred);
	return cred;
}

static void nfs4_clear_machine_cred(struct nfs_client *clp)
{
	struct rpc_cred *cred;

	spin_lock(&clp->cl_lock);
	cred = clp->cl_machine_cred;
	clp->cl_machine_cred = NULL;
	spin_unlock(&clp->cl_lock);
	if (cred != NULL)
		put_rpccred(cred);
}

static struct rpc_cred *
nfs4_get_renew_cred_server_locked(struct nfs_server *server)
{
	struct rpc_cred *cred = NULL;
	struct nfs4_state_owner *sp;
	struct rb_node *pos;

	for (pos = rb_first(&server->state_owners);
	     pos != NULL;
	     pos = rb_next(pos)) {
		sp = rb_entry(pos, struct nfs4_state_owner, so_server_node);
		if (list_empty(&sp->so_states))
			continue;
		cred = get_rpccred(sp->so_cred);
		break;
	}
	return cred;
}

/**
 * nfs4_get_renew_cred_locked - Acquire credential for a renew operation
 * @clp: client state handle
 *
 * Returns an rpc_cred with reference count bumped, or NULL.
 * Caller must hold clp->cl_lock.
 */
struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp)
{
	struct rpc_cred *cred = NULL;
	struct nfs_server *server;

	/* Use machine credentials if available */
	cred = nfs4_get_machine_cred_locked(clp);
	if (cred != NULL)
		goto out;

	rcu_read_lock();
	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
		cred = nfs4_get_renew_cred_server_locked(server);
		if (cred != NULL)
			break;
	}
	rcu_read_unlock();

out:
	return cred;
}

#if defined(CONFIG_NFS_V4_1)

static int nfs41_setup_state_renewal(struct nfs_client *clp)
{
	int status;
	struct nfs_fsinfo fsinfo;

	if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) {
		nfs4_schedule_state_renewal(clp);
		return 0;
	}

	status = nfs4_proc_get_lease_time(clp, &fsinfo);
	if (status == 0) {
		/* Update lease time and schedule renewal */
		spin_lock(&clp->cl_lock);
		clp->cl_lease_time = fsinfo.lease_time * HZ;
		clp->cl_last_renewal = jiffies;
		spin_unlock(&clp->cl_lock);

		nfs4_schedule_state_renewal(clp);
	}

	return status;
}

/*
 * Back channel returns NFS4ERR_DELAY for new requests when
 * NFS4_SESSION_DRAINING is set so there is no work to be done when draining
 * is ended.
 */
static void nfs4_end_drain_session(struct nfs_client *clp)
{
	struct nfs4_session *ses = clp->cl_session;
	struct nfs4_slot_table *tbl;
	unsigned int i;

	if (ses == NULL)
		return;
	tbl = &ses->fc_slot_table;
	if (test_and_clear_bit(NFS4_SESSION_DRAINING, &ses->session_state)) {
		spin_lock(&tbl->slot_tbl_lock);
		for (i = 0; i <= tbl->max_slotid; i++) {
			if (rpc_wake_up_first(&tbl->slot_tbl_waitq,
						nfs4_set_task_privileged,
						NULL) == NULL)
				break;
		}
		spin_unlock(&tbl->slot_tbl_lock);
	}
}

/*
 * Signal state manager thread if session fore channel is drained
 */
void nfs4_session_drain_complete(struct nfs4_session *session,
		struct nfs4_slot_table *tbl)
{
	if (nfs4_session_draining(session))
		complete(&tbl->complete);
}

static int nfs4_wait_on_slot_tbl(struct nfs4_slot_table *tbl)
{
	spin_lock(&tbl->slot_tbl_lock);
	if (tbl->highest_used_slotid != NFS4_NO_SLOT) {
		INIT_COMPLETION(tbl->complete);
		spin_unlock(&tbl->slot_tbl_lock);
		return wait_for_completion_interruptible(&tbl->complete);
	}
	spin_unlock(&tbl->slot_tbl_lock);
	return 0;
}

static int nfs4_begin_drain_session(struct nfs_client *clp)
{
	struct nfs4_session *ses = clp->cl_session;
	int ret = 0;

	set_bit(NFS4_SESSION_DRAINING, &ses->session_state);
	/* back channel */
	ret = nfs4_wait_on_slot_tbl(&ses->bc_slot_table);
	if (ret)
		return ret;
	/* fore channel */
	return nfs4_wait_on_slot_tbl(&ses->fc_slot_table);
}

static void nfs41_finish_session_reset(struct nfs_client *clp)
{
	clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
	clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state);
	/* create_session negotiated new slot table */
	clear_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state);
	nfs41_setup_state_renewal(clp);
}

int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
{
	int status;

	if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state))
		goto do_confirm;
	nfs4_begin_drain_session(clp);
	status = nfs4_proc_exchange_id(clp, cred);
	if (status != 0)
		goto out;
	set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
do_confirm:
	status = nfs4_proc_create_session(clp, cred);
	if (status != 0)
		goto out;
	nfs41_finish_session_reset(clp);
	nfs_mark_client_ready(clp, NFS_CS_READY);
out:
	return status;
}

/**
 * nfs41_discover_server_trunking - Detect server IP address trunking (mv1)
 *
 * @clp: nfs_client under test
 * @result: OUT: found nfs_client, or clp
 * @cred: credential to use for trunking test
 *
 * Returns NFS4_OK, a negative errno, or a negative NFS4ERR status.
 * If NFS4_OK is returned, an nfs_client pointer is planted in
 * "result".
 *
 * Note: The returned client may not yet be marked ready.
 */
int nfs41_discover_server_trunking(struct nfs_client *clp,
				   struct nfs_client **result,
				   struct rpc_cred *cred)
{
	int status;

	status = nfs4_proc_exchange_id(clp, cred);
	if (status != NFS4_OK)
		return status;
	set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);

	return nfs41_walk_client_list(clp, result, cred);
}

struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp)
{
	struct rpc_cred *cred;

	spin_lock(&clp->cl_lock);
	cred = nfs4_get_machine_cred_locked(clp);
	spin_unlock(&clp->cl_lock);
	return cred;
}

#endif /* CONFIG_NFS_V4_1 */

static struct rpc_cred *
nfs4_get_setclientid_cred_server(struct nfs_server *server)
{
	struct nfs_client *clp = server->nfs_client;
	struct rpc_cred *cred = NULL;
	struct nfs4_state_owner *sp;
	struct rb_node *pos;

	spin_lock(&clp->cl_lock);
	pos = rb_first(&server->state_owners);
	if (pos != NULL) {
		sp = rb_entry(pos, struct nfs4_state_owner, so_server_node);
		cred = get_rpccred(sp->so_cred);
	}
	spin_unlock(&clp->cl_lock);
	return cred;
}

/**
 * nfs4_get_setclientid_cred - Acquire credential for a setclientid operation
 * @clp: client state handle
 *
 * Returns an rpc_cred with reference count bumped, or NULL.
 */
struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
{
	struct nfs_server *server;
	struct rpc_cred *cred;

	spin_lock(&clp->cl_lock);
	cred = nfs4_get_machine_cred_locked(clp);
	spin_unlock(&clp->cl_lock);
	if (cred != NULL)
		goto out;

	rcu_read_lock();
	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
		cred = nfs4_get_setclientid_cred_server(server);
		if (cred != NULL)
			break;
	}
	rcu_read_unlock();

out:
	return cred;
}

static struct nfs4_state_owner *
nfs4_find_state_owner_locked(struct nfs_server *server, struct rpc_cred *cred)
{
	struct rb_node **p = &server->state_owners.rb_node,
		       *parent = NULL;
	struct nfs4_state_owner *sp;

	while (*p != NULL) {
		parent = *p;
		sp = rb_entry(parent, struct nfs4_state_owner, so_server_node);

		if (cred < sp->so_cred)
			p = &parent->rb_left;
		else if (cred > sp->so_cred)
			p = &parent->rb_right;
		else {
			if (!list_empty(&sp->so_lru))
				list_del_init(&sp->so_lru);
			atomic_inc(&sp->so_count);
			return sp;
		}
	}
	return NULL;
}

static struct nfs4_state_owner *
nfs4_insert_state_owner_locked(struct nfs4_state_owner *new)
{
	struct nfs_server *server = new->so_server;
	struct rb_node **p = &server->state_owners.rb_node,
		       *parent = NULL;
	struct nfs4_state_owner *sp;
	int err;

	while (*p != NULL) {
		parent = *p;
		sp = rb_entry(parent, struct nfs4_state_owner, so_server_node);

		if (new->so_cred < sp->so_cred)
			p = &parent->rb_left;
		else if (new->so_cred > sp->so_cred)
			p = &parent->rb_right;
		else {
			if (!list_empty(&sp->so_lru))
				list_del_init(&sp->so_lru);
			atomic_inc(&sp->so_count);
			return sp;
		}
	}
	err = ida_get_new(&server->openowner_id, &new->so_seqid.owner_id);
	if (err)
		return ERR_PTR(err);
	rb_link_node(&new->so_server_node, parent, p);
	rb_insert_color(&new->so_server_node, &server->state_owners);
	return new;
}

static void
nfs4_remove_state_owner_locked(struct nfs4_state_owner *sp)
{
	struct nfs_server *server = sp->so_server;

	if (!RB_EMPTY_NODE(&sp->so_server_node))
		rb_erase(&sp->so_server_node, &server->state_owners);
	ida_remove(&server->openowner_id, sp->so_seqid.owner_id);
}

static void
nfs4_init_seqid_counter(struct nfs_seqid_counter *sc)
{
	sc->create_time = ktime_get();
	sc->flags = 0;
	sc->counter = 0;
	spin_lock_init(&sc->lock);
	INIT_LIST_HEAD(&sc->list);
	rpc_init_wait_queue(&sc->wait, "Seqid_waitqueue");
}

static void
nfs4_destroy_seqid_counter(struct nfs_seqid_counter *sc)
{
	rpc_destroy_wait_queue(&sc->wait);
}

/*
 * nfs4_alloc_state_owner(): this is called on the OPEN or CREATE path to
 * create a new state_owner.
 *
 */
static struct nfs4_state_owner *
nfs4_alloc_state_owner(struct nfs_server *server,
		struct rpc_cred *cred,
		gfp_t gfp_flags)
{
	struct nfs4_state_owner *sp;

	sp = kzalloc(sizeof(*sp), gfp_flags);
	if (!sp)
		return NULL;
	sp->so_server = server;
	sp->so_cred = get_rpccred(cred);
	spin_lock_init(&sp->so_lock);
	INIT_LIST_HEAD(&sp->so_states);
	nfs4_init_seqid_counter(&sp->so_seqid);
	atomic_set(&sp->so_count, 1);
	INIT_LIST_HEAD(&sp->so_lru);
	return sp;
}

static void
nfs4_drop_state_owner(struct nfs4_state_owner *sp)
{
	struct rb_node *rb_node = &sp->so_server_node;

	if (!RB_EMPTY_NODE(rb_node)) {
		struct nfs_server *server = sp->so_server;
		struct nfs_client *clp = server->nfs_client;

		spin_lock(&clp->cl_lock);
		if (!RB_EMPTY_NODE(rb_node)) {
			rb_erase(rb_node, &server->state_owners);
			RB_CLEAR_NODE(rb_node);
		}
		spin_unlock(&clp->cl_lock);
	}
}

static void nfs4_free_state_owner(struct nfs4_state_owner *sp)
{
	nfs4_destroy_seqid_counter(&sp->so_seqid);
	put_rpccred(sp->so_cred);
	kfree(sp);
}

static void nfs4_gc_state_owners(struct nfs_server *server)
{
	struct nfs_client *clp = server->nfs_client;
	struct nfs4_state_owner *sp, *tmp;
	unsigned long time_min, time_max;
	LIST_HEAD(doomed);

	spin_lock(&clp->cl_lock);
	time_max = jiffies;
	time_min = (long)time_max - (long)clp->cl_lease_time;
	list_for_each_entry_safe(sp, tmp, &server->state_owners_lru, so_lru) {
		/* NB: LRU is sorted so that oldest is at the head */
		if (time_in_range(sp->so_expires, time_min, time_max))
			break;
		list_move(&sp->so_lru, &doomed);
		nfs4_remove_state_owner_locked(sp);
	}
	spin_unlock(&clp->cl_lock);

	list_for_each_entry_safe(sp, tmp, &doomed, so_lru) {
		list_del(&sp->so_lru);
		nfs4_free_state_owner(sp);
	}
}

/**
 * nfs4_get_state_owner - Look up a state owner given a credential
 * @server: nfs_server to search
 * @cred: RPC credential to match
 *
 * Returns a pointer to an instantiated nfs4_state_owner struct, or NULL.
 */
struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server,
					      struct rpc_cred *cred,
					      gfp_t gfp_flags)
{
	struct nfs_client *clp = server->nfs_client;
	struct nfs4_state_owner *sp, *new;

	spin_lock(&clp->cl_lock);
	sp = nfs4_find_state_owner_locked(server, cred);
	spin_unlock(&clp->cl_lock);
	if (sp != NULL)
		goto out;
	new = nfs4_alloc_state_owner(server, cred, gfp_flags);
	if (new == NULL)
		goto out;
	do {
		if (ida_pre_get(&server->openowner_id, gfp_flags) == 0)
			break;
		spin_lock(&clp->cl_lock);
		sp = nfs4_insert_state_owner_locked(new);
		spin_unlock(&clp->cl_lock);
	} while (sp == ERR_PTR(-EAGAIN));
	if (sp != new)
		nfs4_free_state_owner(new);
out:
	nfs4_gc_state_owners(server);
	return sp;
}

/**
 * nfs4_put_state_owner - Release a nfs4_state_owner
 * @sp: state owner data to release
 *
 * Note that we keep released state owners on an LRU
 * list.
 * This caches valid state owners so that they can be
 * reused, to avoid the OPEN_CONFIRM on minor version 0.
 * It also pins the uniquifier of dropped state owners for
 * a while, to ensure that those state owner names are
 * never reused.
 */
void nfs4_put_state_owner(struct nfs4_state_owner *sp)
{
	struct nfs_server *server = sp->so_server;
	struct nfs_client *clp = server->nfs_client;

	if (!atomic_dec_and_lock(&sp->so_count, &clp->cl_lock))
		return;

	sp->so_expires = jiffies;
	list_add_tail(&sp->so_lru, &server->state_owners_lru);
	spin_unlock(&clp->cl_lock);
}

/**
 * nfs4_purge_state_owners - Release all cached state owners
 * @server: nfs_server with cached state owners to release
 *
 * Called at umount time.  Remaining state owners will be on
 * the LRU with ref count of zero.
 */
void nfs4_purge_state_owners(struct nfs_server *server)
{
	struct nfs_client *clp = server->nfs_client;
	struct nfs4_state_owner *sp, *tmp;
	LIST_HEAD(doomed);

	spin_lock(&clp->cl_lock);
	list_for_each_entry_safe(sp, tmp, &server->state_owners_lru, so_lru) {
		list_move(&sp->so_lru, &doomed);
		nfs4_remove_state_owner_locked(sp);
	}
	spin_unlock(&clp->cl_lock);

	list_for_each_entry_safe(sp, tmp, &doomed, so_lru) {
		list_del(&sp->so_lru);
		nfs4_free_state_owner(sp);
	}
}

static struct nfs4_state *
nfs4_alloc_open_state(void)
{
	struct nfs4_state *state;

	state = kzalloc(sizeof(*state), GFP_NOFS);
	if (!state)
		return NULL;
	atomic_set(&state->count, 1);
	INIT_LIST_HEAD(&state->lock_states);
	spin_lock_init(&state->state_lock);
	seqlock_init(&state->seqlock);
	return state;
}

void
nfs4_state_set_mode_locked(struct nfs4_state *state, fmode_t fmode)
{
	if (state->state == fmode)
		return;
	/* NB! List reordering - see the reclaim code for why.  */
	if ((fmode & FMODE_WRITE) != (state->state & FMODE_WRITE)) {
		if (fmode & FMODE_WRITE)
			list_move(&state->open_states, &state->owner->so_states);
		else
			list_move_tail(&state->open_states, &state->owner->so_states);
	}
	state->state = fmode;
}

static struct nfs4_state *
__nfs4_find_state_byowner(struct inode *inode, struct nfs4_state_owner *owner)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	struct nfs4_state *state;

	list_for_each_entry(state, &nfsi->open_states, inode_states) {
		if (state->owner != owner)
			continue;
		if (atomic_inc_not_zero(&state->count))
			return state;
	}
	return NULL;
}

static void
nfs4_free_open_state(struct nfs4_state *state)
{
	kfree(state);
}

struct nfs4_state *
nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner)
{
	struct nfs4_state *state, *new;
	struct nfs_inode *nfsi = NFS_I(inode);

	spin_lock(&inode->i_lock);
	state = __nfs4_find_state_byowner(inode, owner);
	spin_unlock(&inode->i_lock);
	if (state)
		goto out;
	new = nfs4_alloc_open_state();
	spin_lock(&owner->so_lock);
	spin_lock(&inode->i_lock);
	state = __nfs4_find_state_byowner(inode, owner);
	if (state == NULL && new != NULL) {
		state = new;
		state->owner = owner;
		atomic_inc(&owner->so_count);
		list_add(&state->inode_states, &nfsi->open_states);
		ihold(inode);
		state->inode = inode;
		spin_unlock(&inode->i_lock);
		/* Note: The reclaim code dictates that we add stateless
		 * and read-only stateids to the end of the list */
		list_add_tail(&state->open_states, &owner->so_states);
		spin_unlock(&owner->so_lock);
	} else {
		spin_unlock(&inode->i_lock);
		spin_unlock(&owner->so_lock);
		if (new)
			nfs4_free_open_state(new);
	}
out:
	return state;
}

void nfs4_put_open_state(struct nfs4_state *state)
{
	struct inode *inode = state->inode;
	struct nfs4_state_owner *owner = state->owner;

	if (!atomic_dec_and_lock(&state->count, &owner->so_lock))
		return;
	spin_lock(&inode->i_lock);
	list_del(&state->inode_states);
	list_del(&state->open_states);
	spin_unlock(&inode->i_lock);
	spin_unlock(&owner->so_lock);
	iput(inode);
	nfs4_free_open_state(state);
	nfs4_put_state_owner(owner);
}

/*
 * Close the current file.
 */
static void __nfs4_close(struct nfs4_state *state,
		fmode_t fmode, gfp_t gfp_mask, int wait)
{
	struct nfs4_state_owner *owner = state->owner;
	int call_close = 0;
	fmode_t newstate;

	atomic_inc(&owner->so_count);
	/* Protect against nfs4_find_state() */
	spin_lock(&owner->so_lock);
	switch (fmode & (FMODE_READ | FMODE_WRITE)) {
		case FMODE_READ:
			state->n_rdonly--;
			break;
		case FMODE_WRITE:
			state->n_wronly--;
			break;
		case FMODE_READ|FMODE_WRITE:
			state->n_rdwr--;
	}
	newstate = FMODE_READ|FMODE_WRITE;
	if (state->n_rdwr == 0) {
		if (state->n_rdonly == 0) {
			newstate &= ~FMODE_READ;
			call_close |= test_bit(NFS_O_RDONLY_STATE, &state->flags);
			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
		}
		if (state->n_wronly == 0) {
			newstate &= ~FMODE_WRITE;
			call_close |= test_bit(NFS_O_WRONLY_STATE, &state->flags);
			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
		}
		if (newstate == 0)
			clear_bit(NFS_DELEGATED_STATE, &state->flags);
	}
	nfs4_state_set_mode_locked(state, newstate);
	spin_unlock(&owner->so_lock);

	if (!call_close) {
		nfs4_put_open_state(state);
		nfs4_put_state_owner(owner);
	} else
		nfs4_do_close(state, gfp_mask, wait);
}

void nfs4_close_state(struct nfs4_state *state, fmode_t fmode)
{
	__nfs4_close(state, fmode, GFP_NOFS, 0);
}

void nfs4_close_sync(struct nfs4_state *state, fmode_t fmode)
{
	__nfs4_close(state, fmode, GFP_KERNEL, 1);
}

/*
 * Search the state->lock_states for an existing lock_owner
 * that is compatible with current->files
 */
static struct nfs4_lock_state *
__nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner, pid_t fl_pid, unsigned int type)
{
	struct nfs4_lock_state *pos;
	list_for_each_entry(pos, &state->lock_states, ls_locks) {
		if (type != NFS4_ANY_LOCK_TYPE && pos->ls_owner.lo_type != type)
			continue;
		switch (pos->ls_owner.lo_type) {
		case NFS4_POSIX_LOCK_TYPE:
			if (pos->ls_owner.lo_u.posix_owner != fl_owner)
				continue;
			break;
		case NFS4_FLOCK_LOCK_TYPE:
			if (pos->ls_owner.lo_u.flock_owner != fl_pid)
				continue;
		}
		atomic_inc(&pos->ls_count);
		return pos;
	}
	return NULL;
}

/*
 * Return a compatible lock_state. If no initialized lock_state structure
 * exists, return an uninitialized one.
 *
 */
static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner, pid_t fl_pid, unsigned int type)
{
	struct nfs4_lock_state *lsp;
	struct nfs_server *server = state->owner->so_server;

	lsp = kzalloc(sizeof(*lsp), GFP_NOFS);
	if (lsp == NULL)
		return NULL;
	nfs4_init_seqid_counter(&lsp->ls_seqid);
	atomic_set(&lsp->ls_count, 1);
	lsp->ls_state = state;
	lsp->ls_owner.lo_type = type;
	switch (lsp->ls_owner.lo_type) {
	case NFS4_FLOCK_LOCK_TYPE:
		lsp->ls_owner.lo_u.flock_owner = fl_pid;
		break;
	case NFS4_POSIX_LOCK_TYPE:
		lsp->ls_owner.lo_u.posix_owner = fl_owner;
		break;
	default:
		goto out_free;
	}
	lsp->ls_seqid.owner_id = ida_simple_get(&server->lockowner_id, 0, 0, GFP_NOFS);
	if (lsp->ls_seqid.owner_id < 0)
		goto out_free;
	INIT_LIST_HEAD(&lsp->ls_locks);
	return lsp;
out_free:
	kfree(lsp);
	return NULL;
}

void nfs4_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp)
{
	ida_simple_remove(&server->lockowner_id, lsp->ls_seqid.owner_id);
	nfs4_destroy_seqid_counter(&lsp->ls_seqid);
	kfree(lsp);
}

/*
 * Return a compatible lock_state. If no initialized lock_state structure
 * exists, return an uninitialized one.
 *
 */
static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner, pid_t pid, unsigned int type)
{
	struct nfs4_lock_state *lsp, *new = NULL;
	
	for(;;) {
		spin_lock(&state->state_lock);
		lsp = __nfs4_find_lock_state(state, owner, pid, type);
		if (lsp != NULL)
			break;
		if (new != NULL) {
			list_add(&new->ls_locks, &state->lock_states);
			set_bit(LK_STATE_IN_USE, &state->flags);
			lsp = new;
			new = NULL;
			break;
		}
		spin_unlock(&state->state_lock);
		new = nfs4_alloc_lock_state(state, owner, pid, type);
		if (new == NULL)
			return NULL;
	}
	spin_unlock(&state->state_lock);
	if (new != NULL)
		nfs4_free_lock_state(state->owner->so_server, new);
	return lsp;
}

/*
 * Release reference to lock_state, and free it if we see that
 * it is no longer in use
 */
void nfs4_put_lock_state(struct nfs4_lock_state *lsp)
{
	struct nfs4_state *state;

	if (lsp == NULL)
		return;
	state = lsp->ls_state;
	if (!atomic_dec_and_lock(&lsp->ls_count, &state->state_lock))
		return;
	list_del(&lsp->ls_locks);
	if (list_empty(&state->lock_states))
		clear_bit(LK_STATE_IN_USE, &state->flags);
	spin_unlock(&state->state_lock);
	if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags)) {
		if (nfs4_release_lockowner(lsp) == 0)
			return;
	}
	nfs4_free_lock_state(lsp->ls_state->owner->so_server, lsp);
}

static void nfs4_fl_copy_lock(struct file_lock *dst, struct file_lock *src)
{
	struct nfs4_lock_state *lsp = src->fl_u.nfs4_fl.owner;

	dst->fl_u.nfs4_fl.owner = lsp;
	atomic_inc(&lsp->ls_count);
}

static void nfs4_fl_release_lock(struct file_lock *fl)
{
	nfs4_put_lock_state(fl->fl_u.nfs4_fl.owner);
}

static const struct file_lock_operations nfs4_fl_lock_ops = {
	.fl_copy_lock = nfs4_fl_copy_lock,
	.fl_release_private = nfs4_fl_release_lock,
};

int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl)
{
	struct nfs4_lock_state *lsp;

	if (fl->fl_ops != NULL)
		return 0;
	if (fl->fl_flags & FL_POSIX)
		lsp = nfs4_get_lock_state(state, fl->fl_owner, 0, NFS4_POSIX_LOCK_TYPE);
	else if (fl->fl_flags & FL_FLOCK)
		lsp = nfs4_get_lock_state(state, NULL, fl->fl_pid,
				NFS4_FLOCK_LOCK_TYPE);
	else
		return -EINVAL;
	if (lsp == NULL)
		return -ENOMEM;
	fl->fl_u.nfs4_fl.owner = lsp;
	fl->fl_ops = &nfs4_fl_lock_ops;
	return 0;
}

static bool nfs4_copy_lock_stateid(nfs4_stateid *dst, struct nfs4_state *state,
		const struct nfs_lockowner *lockowner)