/*
 *  fs/nfs/nfs4proc.c
 *
 *  Client-side procedure declarations for NFSv4.
 *
 *  Copyright (c) 2002 The Regents of the University of Michigan.
 *  All rights reserved.
 *
 *  Kendrick Smith <kmsmith@umich.edu>
 *  Andy Adamson   <andros@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.
 */

#include <linux/mm.h>
#include <linux/delay.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/ratelimit.h>
#include <linux/printk.h>
#include <linux/slab.h>
#include <linux/sunrpc/clnt.h>
#include <linux/sunrpc/gss_api.h>
#include <linux/nfs.h>
#include <linux/nfs4.h>
#include <linux/nfs_fs.h>
#include <linux/nfs_page.h>
#include <linux/nfs_mount.h>
#include <linux/namei.h>
#include <linux/mount.h>
#include <linux/module.h>
#include <linux/nfs_idmap.h>
#include <linux/sunrpc/bc_xprt.h>
#include <linux/xattr.h>
#include <linux/utsname.h>
#include <linux/freezer.h>

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

#define NFSDBG_FACILITY		NFSDBG_PROC

#define NFS4_POLL_RETRY_MIN	(HZ/10)
#define NFS4_POLL_RETRY_MAX	(15*HZ)

#define NFS4_MAX_LOOP_ON_RECOVER (10)

static unsigned short max_session_slots = NFS4_DEF_SLOT_TABLE_SIZE;

struct nfs4_opendata;
static int _nfs4_proc_open(struct nfs4_opendata *data);
static int _nfs4_recover_proc_open(struct nfs4_opendata *data);
static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *, struct nfs4_state *);
static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr);
static int nfs4_proc_getattr(struct nfs_server *, struct nfs_fh *, struct nfs_fattr *);
static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr);
static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
			    struct nfs_fattr *fattr, struct iattr *sattr,
			    struct nfs4_state *state);
#ifdef CONFIG_NFS_V4_1
static int nfs41_test_stateid(struct nfs_server *, nfs4_stateid *);
static int nfs41_free_stateid(struct nfs_server *, nfs4_stateid *);
#endif
/* Prevent leaks of NFSv4 errors into userland */
static int nfs4_map_errors(int err)
{
	if (err >= -1000)
		return err;
	switch (err) {
	case -NFS4ERR_RESOURCE:
		return -EREMOTEIO;
	case -NFS4ERR_WRONGSEC:
		return -EPERM;
	case -NFS4ERR_BADOWNER:
	case -NFS4ERR_BADNAME:
		return -EINVAL;
	case -NFS4ERR_SHARE_DENIED:
		return -EACCES;
	default:
		dprintk("%s could not handle NFSv4 error %d\n",
				__func__, -err);
		break;
	}
	return -EIO;
}

/*
 * This is our standard bitmap for GETATTR requests.
 */
const u32 nfs4_fattr_bitmap[2] = {
	FATTR4_WORD0_TYPE
	| FATTR4_WORD0_CHANGE
	| FATTR4_WORD0_SIZE
	| FATTR4_WORD0_FSID
	| FATTR4_WORD0_FILEID,
	FATTR4_WORD1_MODE
	| FATTR4_WORD1_NUMLINKS
	| FATTR4_WORD1_OWNER
	| FATTR4_WORD1_OWNER_GROUP
	| FATTR4_WORD1_RAWDEV
	| FATTR4_WORD1_SPACE_USED
	| FATTR4_WORD1_TIME_ACCESS
	| FATTR4_WORD1_TIME_METADATA
	| FATTR4_WORD1_TIME_MODIFY
};

const u32 nfs4_statfs_bitmap[2] = {
	FATTR4_WORD0_FILES_AVAIL
	| FATTR4_WORD0_FILES_FREE
	| FATTR4_WORD0_FILES_TOTAL,
	FATTR4_WORD1_SPACE_AVAIL
	| FATTR4_WORD1_SPACE_FREE
	| FATTR4_WORD1_SPACE_TOTAL
};

const u32 nfs4_pathconf_bitmap[2] = {
	FATTR4_WORD0_MAXLINK
	| FATTR4_WORD0_MAXNAME,
	0
};

const u32 nfs4_fsinfo_bitmap[3] = { FATTR4_WORD0_MAXFILESIZE
			| FATTR4_WORD0_MAXREAD
			| FATTR4_WORD0_MAXWRITE
			| FATTR4_WORD0_LEASE_TIME,
			FATTR4_WORD1_TIME_DELTA
			| FATTR4_WORD1_FS_LAYOUT_TYPES,
			FATTR4_WORD2_LAYOUT_BLKSIZE
};

const u32 nfs4_fs_locations_bitmap[2] = {
	FATTR4_WORD0_TYPE
	| FATTR4_WORD0_CHANGE
	| FATTR4_WORD0_SIZE
	| FATTR4_WORD0_FSID
	| FATTR4_WORD0_FILEID
	| FATTR4_WORD0_FS_LOCATIONS,
	FATTR4_WORD1_MODE
	| FATTR4_WORD1_NUMLINKS
	| FATTR4_WORD1_OWNER
	| FATTR4_WORD1_OWNER_GROUP
	| FATTR4_WORD1_RAWDEV
	| FATTR4_WORD1_SPACE_USED
	| FATTR4_WORD1_TIME_ACCESS
	| FATTR4_WORD1_TIME_METADATA
	| FATTR4_WORD1_TIME_MODIFY
	| FATTR4_WORD1_MOUNTED_ON_FILEID
};

static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dentry,
		struct nfs4_readdir_arg *readdir)
{
	__be32 *start, *p;

	BUG_ON(readdir->count < 80);
	if (cookie > 2) {
		readdir->cookie = cookie;
		memcpy(&readdir->verifier, verifier, sizeof(readdir->verifier));
		return;
	}

	readdir->cookie = 0;
	memset(&readdir->verifier, 0, sizeof(readdir->verifier));
	if (cookie == 2)
		return;
	
	/*
	 * NFSv4 servers do not return entries for '.' and '..'
	 * Therefore, we fake these entries here.  We let '.'
	 * have cookie 0 and '..' have cookie 1.  Note that
	 * when talking to the server, we always send cookie 0
	 * instead of 1 or 2.
	 */
	start = p = kmap_atomic(*readdir->pages);
	
	if (cookie == 0) {
		*p++ = xdr_one;                                  /* next */
		*p++ = xdr_zero;                   /* cookie, first word */
		*p++ = xdr_one;                   /* cookie, second word */
		*p++ = xdr_one;                             /* entry len */
		memcpy(p, ".\0\0\0", 4);                        /* entry */
		p++;
		*p++ = xdr_one;                         /* bitmap length */
		*p++ = htonl(FATTR4_WORD0_FILEID);             /* bitmap */
		*p++ = htonl(8);              /* attribute buffer length */
		p = xdr_encode_hyper(p, NFS_FILEID(dentry->d_inode));
	}
	
	*p++ = xdr_one;                                  /* next */
	*p++ = xdr_zero;                   /* cookie, first word */
	*p++ = xdr_two;                   /* cookie, second word */
	*p++ = xdr_two;                             /* entry len */
	memcpy(p, "..\0\0", 4);                         /* entry */
	p++;
	*p++ = xdr_one;                         /* bitmap length */
	*p++ = htonl(FATTR4_WORD0_FILEID);             /* bitmap */
	*p++ = htonl(8);              /* attribute buffer length */
	p = xdr_encode_hyper(p, NFS_FILEID(dentry->d_parent->d_inode));

	readdir->pgbase = (char *)p - (char *)start;
	readdir->count -= readdir->pgbase;
	kunmap_atomic(start);
}

static int nfs4_wait_clnt_recover(struct nfs_client *clp)
{
	int res;

	might_sleep();

	res = wait_on_bit(&clp->cl_state, NFS4CLNT_MANAGER_RUNNING,
			nfs_wait_bit_killable, TASK_KILLABLE);
	return res;
}

static int nfs4_delay(struct rpc_clnt *clnt, long *timeout)
{
	int res = 0;

	might_sleep();

	if (*timeout <= 0)
		*timeout = NFS4_POLL_RETRY_MIN;
	if (*timeout > NFS4_POLL_RETRY_MAX)
		*timeout = NFS4_POLL_RETRY_MAX;
	freezable_schedule_timeout_killable(*timeout);
	if (fatal_signal_pending(current))
		res = -ERESTARTSYS;
	*timeout <<= 1;
	return res;
}

/* This is the error handling routine for processes that are allowed
 * to sleep.
 */
static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
{
	struct nfs_client *clp = server->nfs_client;
	struct nfs4_state *state = exception->state;
	struct inode *inode = exception->inode;
	int ret = errorcode;

	exception->retry = 0;
	switch(errorcode) {
		case 0:
			return 0;
		case -NFS4ERR_OPENMODE:
			if (inode && nfs_have_delegation(inode, FMODE_READ)) {
				nfs_inode_return_delegation(inode);
				exception->retry = 1;
				return 0;
			}
			if (state == NULL)
				break;
			nfs4_schedule_stateid_recovery(server, state);
			goto wait_on_recovery;
		case -NFS4ERR_DELEG_REVOKED:
		case -NFS4ERR_ADMIN_REVOKED:
		case -NFS4ERR_BAD_STATEID:
			if (state == NULL)
				break;
			nfs_remove_bad_delegation(state->inode);
			nfs4_schedule_stateid_recovery(server, state);
			goto wait_on_recovery;
		case -NFS4ERR_EXPIRED:
			if (state != NULL)
				nfs4_schedule_stateid_recovery(server, state);
		case -NFS4ERR_STALE_STATEID:
		case -NFS4ERR_STALE_CLIENTID:
			nfs4_schedule_lease_recovery(clp);
			goto wait_on_recovery;
#if defined(CONFIG_NFS_V4_1)
		case -NFS4ERR_BADSESSION:
		case -NFS4ERR_BADSLOT:
		case -NFS4ERR_BAD_HIGH_SLOT:
		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
		case -NFS4ERR_DEADSESSION:
		case -NFS4ERR_SEQ_FALSE_RETRY:
		case -NFS4ERR_SEQ_MISORDERED:
			dprintk("%s ERROR: %d Reset session\n", __func__,
				errorcode);
			nfs4_schedule_session_recovery(clp->cl_session, errorcode);
			exception->retry = 1;
			break;
#endif /* defined(CONFIG_NFS_V4_1) */
		case -NFS4ERR_FILE_OPEN:
			if (exception->timeout > HZ) {
				/* We have retried a decent amount, time to
				 * fail
				 */
				ret = -EBUSY;
				break;
			}
		case -NFS4ERR_GRACE:
		case -NFS4ERR_DELAY:
		case -EKEYEXPIRED:
			ret = nfs4_delay(server->client, &exception->timeout);
			if (ret != 0)
				break;
		case -NFS4ERR_RETRY_UNCACHED_REP:
		case -NFS4ERR_OLD_STATEID:
			exception->retry = 1;
			break;
		case -NFS4ERR_BADOWNER:
			/* The following works around a Linux server bug! */
		case -NFS4ERR_BADNAME:
			if (server->caps & NFS_CAP_UIDGID_NOMAP) {
				server->caps &= ~NFS_CAP_UIDGID_NOMAP;
				exception->retry = 1;
				printk(KERN_WARNING "NFS: v4 server %s "
						"does not accept raw "
						"uid/gids. "
						"Reenabling the idmapper.\n",
						server->nfs_client->cl_hostname);
			}
	}
	/* We failed to handle the error */
	return nfs4_map_errors(ret);
wait_on_recovery:
	ret = nfs4_wait_clnt_recover(clp);
	if (ret == 0)
		exception->retry = 1;
	return ret;
}


static void do_renew_lease(struct nfs_client *clp, unsigned long timestamp)
{
	spin_lock(&clp->cl_lock);
	if (time_before(clp->cl_last_renewal,timestamp))
		clp->cl_last_renewal = timestamp;
	spin_unlock(&clp->cl_lock);
}

static void renew_lease(const struct nfs_server *server, unsigned long timestamp)
{
	do_renew_lease(server->nfs_client, timestamp);
}

#if defined(CONFIG_NFS_V4_1)

/*
 * nfs4_free_slot - free a slot and efficiently update slot table.
 *
 * freeing a slot is trivially done by clearing its respective bit
 * in the bitmap.
 * If the freed slotid equals highest_used_slotid we want to update it
 * so that the server would be able to size down the slot table if needed,
 * otherwise we know that the highest_used_slotid is still in use.
 * When updating highest_used_slotid there may be "holes" in the bitmap
 * so we need to scan down from highest_used_slotid to 0 looking for the now
 * highest slotid in use.
 * If none found, highest_used_slotid is set to NFS4_NO_SLOT.
 *
 * Must be called while holding tbl->slot_tbl_lock
 */
static void
nfs4_free_slot(struct nfs4_slot_table *tbl, u32 slotid)
{
	BUG_ON(slotid >= NFS4_MAX_SLOT_TABLE);
	/* clear used bit in bitmap */
	__clear_bit(slotid, tbl->used_slots);

	/* update highest_used_slotid when it is freed */
	if (slotid == tbl->highest_used_slotid) {
		slotid = find_last_bit(tbl->used_slots, tbl->max_slots);
		if (slotid < tbl->max_slots)
			tbl->highest_used_slotid = slotid;
		else
			tbl->highest_used_slotid = NFS4_NO_SLOT;
	}
	dprintk("%s: slotid %u highest_used_slotid %d\n", __func__,
		slotid, tbl->highest_used_slotid);
}

bool nfs4_set_task_privileged(struct rpc_task *task, void *dummy)
{
	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
	return true;
}

/*
 * Signal state manager thread if session fore channel is drained
 */
static void nfs4_check_drain_fc_complete(struct nfs4_session *ses)
{
	if (!test_bit(NFS4_SESSION_DRAINING, &ses->session_state)) {
		rpc_wake_up_first(&ses->fc_slot_table.slot_tbl_waitq,
				nfs4_set_task_privileged, NULL);
		return;
	}

	if (ses->fc_slot_table.highest_used_slotid != NFS4_NO_SLOT)
		return;

	dprintk("%s COMPLETE: Session Fore Channel Drained\n", __func__);
	complete(&ses->fc_slot_table.complete);
}

/*
 * Signal state manager thread if session back channel is drained
 */
void nfs4_check_drain_bc_complete(struct nfs4_session *ses)
{
	if (!test_bit(NFS4_SESSION_DRAINING, &ses->session_state) ||
	    ses->bc_slot_table.highest_used_slotid != NFS4_NO_SLOT)
		return;
	dprintk("%s COMPLETE: Session Back Channel Drained\n", __func__);
	complete(&ses->bc_slot_table.complete);
}

static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res)
{
	struct nfs4_slot_table *tbl;

	tbl = &res->sr_session->fc_slot_table;
	if (!res->sr_slot) {
		/* just wake up the next guy waiting since
		 * we may have not consumed a slot after all */
		dprintk("%s: No slot\n", __func__);
		return;
	}

	spin_lock(&tbl->slot_tbl_lock);
	nfs4_free_slot(tbl, res->sr_slot - tbl->slots);
	nfs4_check_drain_fc_complete(res->sr_session);
	spin_unlock(&tbl->slot_tbl_lock);
	res->sr_slot = NULL;
}

static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res)
{
	unsigned long timestamp;
	struct nfs_client *clp;

	/*
	 * sr_status remains 1 if an RPC level error occurred. The server
	 * may or may not have processed the sequence operation..
	 * Proceed as if the server received and processed the sequence
	 * operation.
	 */
	if (res->sr_status == 1)
		res->sr_status = NFS_OK;

	/* don't increment the sequence number if the task wasn't sent */
	if (!RPC_WAS_SENT(task))
		goto out;

	/* Check the SEQUENCE operation status */
	switch (res->sr_status) {
	case 0:
		/* Update the slot's sequence and clientid lease timer */
		++res->sr_slot->seq_nr;
		timestamp = res->sr_renewal_time;
		clp = res->sr_session->clp;
		do_renew_lease(clp, timestamp);
		/* Check sequence flags */
		if (res->sr_status_flags != 0)
			nfs4_schedule_lease_recovery(clp);
		break;
	case -NFS4ERR_DELAY:
		/* The server detected a resend of the RPC call and
		 * returned NFS4ERR_DELAY as per Section 2.10.6.2
		 * of RFC5661.
		 */
		dprintk("%s: slot=%td seq=%d: Operation in progress\n",
			__func__,
			res->sr_slot - res->sr_session->fc_slot_table.slots,
			res->sr_slot->seq_nr);
		goto out_retry;
	default:
		/* Just update the slot sequence no. */
		++res->sr_slot->seq_nr;
	}
out:
	/* The session may be reset by one of the error handlers. */
	dprintk("%s: Error %d free the slot \n", __func__, res->sr_status);
	nfs41_sequence_free_slot(res);
	return 1;
out_retry:
	if (!rpc_restart_call(task))
		goto out;
	rpc_delay(task, NFS4_POLL_RETRY_MAX);
	return 0;
}

static int nfs4_sequence_done(struct rpc_task *task,
			       struct nfs4_sequence_res *res)
{
	if (res->sr_session == NULL)
		return 1;
	return nfs41_sequence_done(task, res);
}

/*
 * nfs4_find_slot - efficiently look for a free slot
 *
 * nfs4_find_slot looks for an unset bit in the used_slots bitmap.
 * If found, we mark the slot as used, update the highest_used_slotid,
 * and respectively set up the sequence operation args.
 * The slot number is returned if found, or NFS4_NO_SLOT otherwise.
 *
 * Note: must be called with under the slot_tbl_lock.
 */
static u32
nfs4_find_slot(struct nfs4_slot_table *tbl)
{
	u32 slotid;
	u32 ret_id = NFS4_NO_SLOT;

	dprintk("--> %s used_slots=%04lx highest_used=%u max_slots=%u\n",
		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
		tbl->max_slots);
	slotid = find_first_zero_bit(tbl->used_slots, tbl->max_slots);
	if (slotid >= tbl->max_slots)
		goto out;
	__set_bit(slotid, tbl->used_slots);
	if (slotid > tbl->highest_used_slotid ||
			tbl->highest_used_slotid == NFS4_NO_SLOT)
		tbl->highest_used_slotid = slotid;
	ret_id = slotid;
out:
	dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n",
		__func__, tbl->used_slots[0], tbl->highest_used_slotid, ret_id);
	return ret_id;
}

static void nfs41_init_sequence(struct nfs4_sequence_args *args,
		struct nfs4_sequence_res *res, int cache_reply)
{
	args->sa_session = NULL;
	args->sa_cache_this = 0;
	if (cache_reply)
		args->sa_cache_this = 1;
	res->sr_session = NULL;
	res->sr_slot = NULL;
}

int nfs41_setup_sequence(struct nfs4_session *session,
				struct nfs4_sequence_args *args,
				struct nfs4_sequence_res *res,
				struct rpc_task *task)
{
	struct nfs4_slot *slot;
	struct nfs4_slot_table *tbl;
	u32 slotid;

	dprintk("--> %s\n", __func__);
	/* slot already allocated? */
	if (res->sr_slot != NULL)
		return 0;

	tbl = &session->fc_slot_table;

	spin_lock(&tbl->slot_tbl_lock);
	if (test_bit(NFS4_SESSION_DRAINING, &session->session_state) &&
	    !rpc_task_has_priority(task, RPC_PRIORITY_PRIVILEGED)) {
		/* The state manager will wait until the slot table is empty */
		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
		spin_unlock(&tbl->slot_tbl_lock);
		dprintk("%s session is draining\n", __func__);
		return -EAGAIN;
	}

	if (!rpc_queue_empty(&tbl->slot_tbl_waitq) &&
	    !rpc_task_has_priority(task, RPC_PRIORITY_PRIVILEGED)) {
		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
		spin_unlock(&tbl->slot_tbl_lock);
		dprintk("%s enforce FIFO order\n", __func__);
		return -EAGAIN;
	}

	slotid = nfs4_find_slot(tbl);
	if (slotid == NFS4_NO_SLOT) {
		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
		spin_unlock(&tbl->slot_tbl_lock);
		dprintk("<-- %s: no free slots\n", __func__);
		return -EAGAIN;
	}
	spin_unlock(&tbl->slot_tbl_lock);

	rpc_task_set_priority(task, RPC_PRIORITY_NORMAL);
	slot = tbl->slots + slotid;
	args->sa_session = session;
	args->sa_slotid = slotid;

	dprintk("<-- %s slotid=%d seqid=%d\n", __func__, slotid, slot->seq_nr);

	res->sr_session = session;
	res->sr_slot = slot;
	res->sr_renewal_time = jiffies;
	res->sr_status_flags = 0;
	/*
	 * sr_status is only set in decode_sequence, and so will remain
	 * set to 1 if an rpc level failure occurs.
	 */
	res->sr_status = 1;
	return 0;
}
EXPORT_SYMBOL_GPL(nfs41_setup_sequence);

int nfs4_setup_sequence(const struct nfs_server *server,
			struct nfs4_sequence_args *args,
			struct nfs4_sequence_res *res,
			struct rpc_task *task)
{
	struct nfs4_session *session = nfs4_get_session(server);
	int ret = 0;

	if (session == NULL)
		goto out;

	dprintk("--> %s clp %p session %p sr_slot %td\n",
		__func__, session->clp, session, res->sr_slot ?
			res->sr_slot - session->fc_slot_table.slots : -1);

	ret = nfs41_setup_sequence(session, args, res, task);
out:
	dprintk("<-- %s status=%d\n", __func__, ret);
	return ret;
}

struct nfs41_call_sync_data {
	const struct nfs_server *seq_server;
	struct nfs4_sequence_args *seq_args;
	struct nfs4_sequence_res *seq_res;
};

static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata)
{
	struct nfs41_call_sync_data *data = calldata;

	dprintk("--> %s data->seq_server %p\n", __func__, data->seq_server);

	if (nfs4_setup_sequence(data->seq_server, data->seq_args,
				data->seq_res, task))
		return;
	rpc_call_start(task);
}

static void nfs41_call_priv_sync_prepare(struct rpc_task *task, void *calldata)
{
	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
	nfs41_call_sync_prepare(task, calldata);
}

static void nfs41_call_sync_done(struct rpc_task *task, void *calldata)
{
	struct nfs41_call_sync_data *data = calldata;

	nfs41_sequence_done(task, data->seq_res);
}

static const struct rpc_call_ops nfs41_call_sync_ops = {
	.rpc_call_prepare = nfs41_call_sync_prepare,
	.rpc_call_done = nfs41_call_sync_done,
};

static const struct rpc_call_ops nfs41_call_priv_sync_ops = {
	.rpc_call_prepare = nfs41_call_priv_sync_prepare,
	.rpc_call_done = nfs41_call_sync_done,
};

static int nfs4_call_sync_sequence(struct rpc_clnt *clnt,
				   struct nfs_server *server,
				   struct rpc_message *msg,
				   struct nfs4_sequence_args *args,
				   struct nfs4_sequence_res *res,
				   int privileged)
{
	int ret;
	struct rpc_task *task;
	struct nfs41_call_sync_data data = {
		.seq_server = server,
		.seq_args = args,
		.seq_res = res,
	};
	struct rpc_task_setup task_setup = {
		.rpc_client = clnt,
		.rpc_message = msg,
		.callback_ops = &nfs41_call_sync_ops,
		.callback_data = &data
	};

	if (privileged)
		task_setup.callback_ops = &nfs41_call_priv_sync_ops;
	task = rpc_run_task(&task_setup);
	if (IS_ERR(task))
		ret = PTR_ERR(task);
	else {
		ret = task->tk_status;
		rpc_put_task(task);
	}
	return ret;
}

int _nfs4_call_sync_session(struct rpc_clnt *clnt,
			    struct nfs_server *server,
			    struct rpc_message *msg,
			    struct nfs4_sequence_args *args,
			    struct nfs4_sequence_res *res,
			    int cache_reply)
{
	nfs41_init_sequence(args, res, cache_reply);
	return nfs4_call_sync_sequence(clnt, server, msg, args, res, 0);
}

#else
static inline
void nfs41_init_sequence(struct nfs4_sequence_args *args,
		struct nfs4_sequence_res *res, int cache_reply)
{
}

static int nfs4_sequence_done(struct rpc_task *task,
			       struct nfs4_sequence_res *res)
{
	return 1;
}
#endif /* CONFIG_NFS_V4_1 */

int _nfs4_call_sync(struct rpc_clnt *clnt,
		    struct nfs_server *server,
		    struct rpc_message *msg,
		    struct nfs4_sequence_args *args,
		    struct nfs4_sequence_res *res,
		    int cache_reply)
{
	nfs41_init_sequence(args, res, cache_reply);
	return rpc_call_sync(clnt, msg, 0);
}

static inline
int nfs4_call_sync(struct rpc_clnt *clnt,
		   struct nfs_server *server,
		   struct rpc_message *msg,
		   struct nfs4_sequence_args *args,
		   struct nfs4_sequence_res *res,
		   int cache_reply)
{
	return server->nfs_client->cl_mvops->call_sync(clnt, server, msg,
						args, res, cache_reply);
}

static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
{
	struct nfs_inode *nfsi = NFS_I(dir);

	spin_lock(&dir->i_lock);
	nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
	if (!cinfo->atomic || cinfo->before != dir->i_version)
		nfs_force_lookup_revalidate(dir);
	dir->i_version = cinfo->after;
	spin_unlock(&dir->i_lock);
}

struct nfs4_opendata {
	struct kref kref;
	struct nfs_openargs o_arg;
	struct nfs_openres o_res;
	struct nfs_open_confirmargs c_arg;
	struct nfs_open_confirmres c_res;
	struct nfs4_string owner_name;
	struct nfs4_string group_name;
	struct nfs_fattr f_attr;
	struct dentry *dir;
	struct dentry *dentry;
	struct nfs4_state_owner *owner;
	struct nfs4_state *state;
	struct iattr attrs;
	unsigned long timestamp;
	unsigned int rpc_done : 1;
	int rpc_status;
	int cancelled;
};


static void nfs4_init_opendata_res(struct nfs4_opendata *p)
{
	p->o_res.f_attr = &p->f_attr;
	p->o_res.seqid = p->o_arg.seqid;
	p->c_res.seqid = p->c_arg.seqid;
	p->o_res.server = p->o_arg.server;
	nfs_fattr_init(&p->f_attr);
	nfs_fattr_init_names(&p->f_attr, &p->owner_name, &p->group_name);
}

static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry,
		struct nfs4_state_owner *sp, fmode_t fmode, int flags,
		const struct iattr *attrs,
		gfp_t gfp_mask)
{
	struct dentry *parent = dget_parent(dentry);
	struct inode *dir = parent->d_inode;
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs4_opendata *p;

	p = kzalloc(sizeof(*p), gfp_mask);
	if (p == NULL)
		goto err;
	p->o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid, gfp_mask);
	if (p->o_arg.seqid == NULL)
		goto err_free;
	nfs_sb_active(dentry->d_sb);
	p->dentry = dget(dentry);
	p->dir = parent;
	p->owner = sp;
	atomic_inc(&sp->so_count);
	p->o_arg.fh = NFS_FH(dir);
	p->o_arg.open_flags = flags;
	p->o_arg.fmode = fmode & (FMODE_READ|FMODE_WRITE);
	p->o_arg.clientid = server->nfs_client->cl_clientid;
	p->o_arg.id.create_time = ktime_to_ns(sp->so_seqid.create_time);
	p->o_arg.id.uniquifier = sp->so_seqid.owner_id;
	p->o_arg.name = &dentry->d_name;
	p->o_arg.server = server;
	p->o_arg.bitmask = server->attr_bitmask;
	p->o_arg.claim = NFS4_OPEN_CLAIM_NULL;
	if (attrs != NULL && attrs->ia_valid != 0) {
		__be32 verf[2];

		p->o_arg.u.attrs = &p->attrs;
		memcpy(&p->attrs, attrs, sizeof(p->attrs));

		verf[0] = jiffies;
		verf[1] = current->pid;
		memcpy(p->o_arg.u.verifier.data, verf,
				sizeof(p->o_arg.u.verifier.data));
	}
	p->c_arg.fh = &p->o_res.fh;
	p->c_arg.stateid = &p->o_res.stateid;
	p->c_arg.seqid = p->o_arg.seqid;
	nfs4_init_opendata_res(p);
	kref_init(&p->kref);
	return p;
err_free:
	kfree(p);
err:
	dput(parent);
	return NULL;
}

static void nfs4_opendata_free(struct kref *kref)
{
	struct nfs4_opendata *p = container_of(kref,
			struct nfs4_opendata, kref);
	struct super_block *sb = p->dentry->d_sb;

	nfs_free_seqid(p->o_arg.seqid);
	if (p->state != NULL)
		nfs4_put_open_state(p->state);
	nfs4_put_state_owner(p->owner);
	dput(p->dir);
	dput(p->dentry);
	nfs_sb_deactive(sb);
	nfs_fattr_free_names(&p->f_attr);
	kfree(p);
}

static void nfs4_opendata_put(struct nfs4_opendata *p)
{
	if (p != NULL)
		kref_put(&p->kref, nfs4_opendata_free);
}

static int nfs4_wait_for_completion_rpc_task(struct rpc_task *task)
{
	int ret;

	ret = rpc_wait_for_completion_task(task);
	return ret;
}

static int can_open_cached(struct nfs4_state *state, fmode_t mode, int open_mode)
{
	int ret = 0;

	if (open_mode & (O_EXCL|O_TRUNC))
		goto out;
	switch (mode & (FMODE_READ|FMODE_WRITE)) {
		case FMODE_READ:
			ret |= test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0
				&& state->n_rdonly != 0;
			break;
		case FMODE_WRITE:
			ret |= test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0
				&& state->n_wronly != 0;
			break;
		case FMODE_READ|FMODE_WRITE:
			ret |= test_bit(NFS_O_RDWR_STATE, &state->flags) != 0
				&& state->n_rdwr != 0;
	}
out:
	return ret;
}

static int can_open_delegated(struct nfs_delegation *delegation, fmode_t fmode)
{
	if (delegation == NULL)
		return 0;
	if ((delegation->type & fmode) != fmode)
		return 0;
	if (test_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags))
		return 0;
	nfs_mark_delegation_referenced(delegation);
	return 1;
}

static void update_open_stateflags(struct nfs4_state *state, fmode_t fmode)
{
	switch (fmode) {
		case FMODE_WRITE:
			state->n_wronly++;
			break;
		case FMODE_READ:
			state->n_rdonly++;
			break;
		case FMODE_READ|FMODE_WRITE:
			state->n_rdwr++;
	}
	nfs4_state_set_mode_locked(state, state->state | fmode);
}

static void nfs_set_open_stateid_locked(struct nfs4_state *state, nfs4_stateid *stateid, fmode_t fmode)
{
	if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0)
		nfs4_stateid_copy(&state->stateid, stateid);
	nfs4_stateid_copy(&state->open_stateid, stateid);
	switch (fmode) {
		case FMODE_READ:
			set_bit(NFS_O_RDONLY_STATE, &state->flags);
			break;
		case FMODE_WRITE:
			set_bit(NFS_O_WRONLY_STATE, &state->flags);
			break;
		case FMODE_READ|FMODE_WRITE:
			set_bit(NFS_O_RDWR_STATE, &state->flags);
	}
}

static void nfs_set_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid, fmode_t fmode)
{
	write_seqlock(&state->seqlock);
	nfs_set_open_stateid_locked(state, stateid, fmode);
	write_sequnlock(&state->seqlock);
}

static void __update_open_stateid(struct nfs4_state *state, nfs4_stateid *open_stateid, const nfs4_stateid *deleg_stateid, fmode_t fmode)
{
	/*
	 * Protect the call to nfs4_state_set_mode_locked and
	 * serialise the stateid update
	 */
	write_seqlock(&state->seqlock);
	if (deleg_stateid != NULL) {
		nfs4_stateid_copy(&state->stateid, deleg_stateid);
		set_bit(NFS_DELEGATED_STATE, &state->flags);
	}
	if (open_stateid != NULL)
		nfs_set_open_stateid_locked(state, open_stateid, fmode);
	write_sequnlock(&state->seqlock);
	spin_lock(&state->owner->so_lock);
	update_open_stateflags(state, fmode);
	spin_unlock(&state->owner->so_lock);
}

static int update_open_stateid(struct nfs4_state *state, nfs4_stateid *open_stateid, nfs4_stateid *delegation, fmode_t fmode)
{
	struct nfs_inode *nfsi = NFS_I(state->inode);
	struct nfs_delegation *deleg_cur;
	int ret = 0;

	fmode &= (FMODE_READ|FMODE_WRITE);

	rcu_read_lock();
	deleg_cur = rcu_dereference(nfsi->delegation);
	if (deleg_cur == NULL)
		goto no_delegation;

	spin_lock(&deleg_cur->lock);
	if (nfsi->delegation != deleg_cur ||
	    (deleg_cur->type & fmode) != fmode)
		goto no_delegation_unlock;

	if (delegation == NULL)
		delegation = &deleg_cur->stateid;
	else if (!nfs4_stateid_match(&deleg_cur->stateid, delegation))
		goto no_delegation_unlock;

	nfs_mark_delegation_referenced(deleg_cur);
	__update_open_stateid(state, open_stateid, &deleg_cur->stateid, fmode);
	ret = 1;
no_delegation_unlock:
	spin_unlock(&deleg_cur->lock);
no_delegation:
	rcu_read_unlock();

	if (!ret && open_stateid != NULL) {
		__update_open_stateid(state, open_stateid, NULL, fmode);
		ret = 1;
	}

	return ret;
}


static void nfs4_return_incompatible_delegation(struct inode *inode, fmode_t fmode)
{
	struct nfs_delegation *delegation;

	rcu_read_lock();
	delegation = rcu_dereference(NFS_I(inode)->delegation);
	if (delegation == NULL || (delegation->type & fmode) == fmode) {
		rcu_read_unlock();
		return;
	}
	rcu_read_unlock();
	nfs_inode_return_delegation(inode);
}

static struct nfs4_state *nfs4_try_open_cached(struct nfs4_opendata *opendata)
{
	struct nfs4_state *state = opendata->state;
	struct nfs_inode *nfsi = NFS_I(state->inode);
	struct nfs_delegation *delegation;
	int open_mode = opendata->o_arg.open_flags & (O_EXCL|O_TRUNC);
	fmode_t fmode = opendata->o_arg.fmode;
	nfs4_stateid stateid;
	int ret = -EAGAIN;

	for (;;) {
		if (can_open_cached(state, fmode, open_mode)) {
			spin_lock(&state->owner->so_lock);
			if (can_open_cached(state, fmode, open_mode)) {
				update_open_stateflags(state, fmode);
				spin_unlock(&state->owner->so_lock);
				goto out_return_state;
			}
			spin_unlock(&state->owner->so_lock);
		}
		rcu_read_lock();
		delegation = rcu_dereference(nfsi->delegation);
		if (!can_open_delegated(delegation, fmode)) {
			rcu_read_unlock();
			break;
		}
		/* Save the delegation */
		nfs4_stateid_copy(&stateid, &delegation->stateid);
		rcu_read_unlock();
		ret = nfs_may_open(state->inode, state->owner->so_cred, open_mode);
		if (ret != 0)
			goto out;
		ret = -EAGAIN;

		/* Try to update the stateid using the delegation */
		if (update_open_stateid(state, NULL, &stateid, fmode))
			goto out_return_state;
	}
out:
	return ERR_PTR(ret);
out_return_state:
	atomic_inc(&state->count);
	return state;
}

static struct nfs4_state *nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data)
{
	struct inode *inode;
	struct nfs4_state *state = NULL;
	struct nfs_delegation *delegation;
	int ret;

	if (!data->rpc_done) {
		state = nfs4_try_open_cached(data);
		goto out;
	}

	ret = -EAGAIN;
	if (!(data->f_attr.valid & NFS_ATTR_FATTR))
		goto err;
	inode = nfs_fhget(data->dir->d_sb, &data->o_res.fh, &data->f_attr);
	ret = PTR_ERR(inode);
	if (IS_ERR(inode))
		goto err;
	ret = -ENOMEM;
	state = nfs4_get_open_state(inode, data->owner);
	if (state == NULL)
		goto err_put_inode;
	if (data->o_res.delegation_type != 0) {
		struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
		int delegation_flags = 0;

		rcu_read_lock();
		delegation = rcu_dereference(NFS_I(inode)->delegation);
		if (delegation)
			delegation_flags = delegation->flags;
		rcu_read_unlock();
		if (data->o_arg.claim == NFS4_OPEN_CLAIM_DELEGATE_CUR) {
			pr_err_ratelimited("NFS: Broken NFSv4 server %s is "
					"returning a delegation for "
					"OPEN(CLAIM_DELEGATE_CUR)\n",
					clp->cl_hostname);
		} else if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0)
			nfs_inode_set_delegation(state->inode,
					data->owner->so_cred,
					&data->o_res);
		else
			nfs_inode_reclaim_delegation(state->inode,
					data->owner->so_cred,
					&data->o_res);
	}

	update_open_stateid(state, &data->o_res.stateid, NULL,
			data->o_arg.fmode);
	iput(inode);
out:
	return state;
err_put_inode:
	iput(inode);
err:
	return ERR_PTR(ret);
}

static struct nfs_open_context *nfs4_state_find_open_context(struct nfs4_state *state)
{
	struct nfs_inode *nfsi = NFS_I(state->inode);
	struct nfs_open_context *ctx;

	spin_lock(&state->inode->i_lock);
	list_for_each_entry(ctx, &nfsi->open_files, list) {
		if (ctx->state != state)
			continue;
		get_nfs_open_context(ctx);
		spin_unlock(&state->inode->i_lock);
		return ctx;
	}
	spin_unlock(&state->inode->i_lock);
	return ERR_PTR(-ENOENT);
}

static struct nfs4_opendata *nfs4_open_recoverdata_alloc(struct nfs_open_context *ctx, struct nfs4_state *state)
{
	struct nfs4_opendata *opendata;

	opendata = nfs4_opendata_alloc(ctx->dentry, state->owner, 0, 0, NULL, GFP_NOFS);
	if (opendata == NULL)
		return ERR_PTR(-ENOMEM);
	opendata->state = state;
	atomic_inc(&state->count);
	return opendata;
}

static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, fmode_t fmode, struct nfs4_state **res)
{
	struct nfs4_state *newstate;
	int ret;

	opendata->o_arg.open_flags = 0;
	opendata->o_arg.fmode = fmode;
	memset(&opendata->o_res, 0, sizeof(opendata->o_res));
	memset(&opendata->c_res, 0, sizeof(opendata->c_res));
	nfs4_init_opendata_res(opendata);
	ret = _nfs4_recover_proc_open(opendata);
	if (ret != 0)
		return ret; 
	newstate = nfs4_opendata_to_nfs4_state(opendata);
	if (IS_ERR(newstate))
		return PTR_ERR(newstate);
	nfs4_close_state(newstate, fmode);
	*res = newstate;
	return 0;
}

static int nfs4_open_recover(struct nfs4_opendata *opendata, struct nfs4_state *state)
{
	struct nfs4_state *newstate;
	int ret;

	/* memory barrier prior to reading state->n_* */
	clear_bit(NFS_DELEGATED_STATE, &state->flags);
	smp_rmb();
	if (state->n_rdwr != 0) {
		clear_bit(NFS_O_RDWR_STATE, &state->flags);
		ret = nfs4_open_recover_helper(opendata, FMODE_READ|FMODE_WRITE, &newstate);
		if (ret != 0)
			return ret;
		if (newstate != state)
			return -ESTALE;
	}
	if (state->n_wronly != 0) {
		clear_bit(NFS_O_WRONLY_STATE, &state->flags);
		ret = nfs4_open_recover_helper(opendata, FMODE_WRITE, &newstate);
		if (ret != 0)
			return ret;
		if (newstate != state)
			return -ESTALE;
	}
	if (state->n_rdonly != 0) {
		clear_bit(NFS_O_RDONLY_STATE, &state->flags);
		ret = nfs4_open_recover_helper(opendata, FMODE_READ, &newstate);
		if (ret != 0)
			return ret;
		if (newstate != state)
			return -ESTALE;
	}
	/*
	 * We may have performed cached opens for all three recoveries.
	 * Check if we need to update the current stateid.
	 */
	if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0 &&
	    !nfs4_stateid_match(&state->stateid, &state->open_stateid)) {
		write_seqlock(&state->seqlock);
		if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0)
			nfs4_stateid_copy(&state->stateid, &state->open_stateid);
		write_sequnlock(&state->seqlock);
	}
	return 0;
}

/*
 * OPEN_RECLAIM:
 * 	reclaim state on the server after a reboot.
 */
static int _nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state *state)
{
	struct nfs_delegation *delegation;
	struct nfs4_opendata *opendata;
	fmode_t delegation_type = 0;
	int status;

	opendata = nfs4_open_recoverdata_alloc(ctx, state);
	if (IS_ERR(opendata))
		return PTR_ERR(opendata);
	opendata->o_arg.claim = NFS4_OPEN_CLAIM_PREVIOUS;
	opendata->o_arg.fh = NFS_FH(state->inode);
	rcu_read_lock();
	delegation = rcu_dereference(NFS_I(state->inode)->delegation);
	if (delegation != NULL && test_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags) != 0)
		delegation_type = delegation->type;
	rcu_read_unlock();
	opendata->o_arg.u.delegation_type = delegation_type;
	status = nfs4_open_recover(opendata, state);
	nfs4_opendata_put(opendata);
	return status;
}

static int nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state *state)
{
	struct nfs_server *server = NFS_SERVER(state->inode);
	struct nfs4_exception exception = { };
	int err;
	do {
		err = _nfs4_do_open_reclaim(ctx, state);
		if (err != -NFS4ERR_DELAY)
			break;
		nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
	return err;
}

static int nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *state)
{
	struct nfs_open_context *ctx;
	int ret;

	ctx = nfs4_state_find_open_context(state);
	if (IS_ERR(ctx))
		return PTR_ERR(ctx);
	ret = nfs4_do_open_reclaim(ctx, state);
	put_nfs_open_context(ctx);
	return ret;
}

static int _nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid)
{
	struct nfs4_opendata *opendata;
	int ret;

	opendata = nfs4_open_recoverdata_alloc(ctx, state);
	if (IS_ERR(opendata))
		return PTR_ERR(opendata);
	opendata->o_arg.claim = NFS4_OPEN_CLAIM_DELEGATE_CUR;
	nfs4_stateid_copy(&opendata->o_arg.u.delegation, stateid);
	ret = nfs4_open_recover(opendata, state);
	nfs4_opendata_put(opendata);
	return ret;
}

int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid)
{
	struct nfs4_exception exception = { };
	struct nfs_server *server = NFS_SERVER(state->inode);
	int err;
	do {
		err = _nfs4_open_delegation_recall(ctx, state, stateid);
		switch (err) {
			case 0:
			case -ENOENT:
			case -ESTALE:
				goto out;
			case -NFS4ERR_BADSESSION:
			case -NFS4ERR_BADSLOT:
			case -NFS4ERR_BAD_HIGH_SLOT:
			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
			case -NFS4ERR_DEADSESSION:
				nfs4_schedule_session_recovery(server->nfs_client->cl_session, err);
				goto out;
			case -NFS4ERR_STALE_CLIENTID:
			case -NFS4ERR_STALE_STATEID:
			case -NFS4ERR_EXPIRED:
				/* Don't recall a delegation if it was lost */
				nfs4_schedule_lease_recovery(server->nfs_client);
				goto out;
			case -ERESTARTSYS:
				/*
				 * The show must go on: exit, but mark the
				 * stateid as needing recovery.
				 */
			case -NFS4ERR_DELEG_REVOKED:
			case -NFS4ERR_ADMIN_REVOKED:
			case -NFS4ERR_BAD_STATEID:
				nfs_inode_find_state_and_recover(state->inode,
						stateid);
				nfs4_schedule_stateid_recovery(server, state);
			case -EKEYEXPIRED:
				/*
				 * User RPCSEC_GSS context has expired.
				 * We cannot recover this stateid now, so
				 * skip it and allow recovery thread to
				 * proceed.
				 */
			case -ENOMEM:
				err = 0;
				goto out;
		}
		err = nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
out:
	return err;
}

static void nfs4_open_confirm_done(struct rpc_task *task, void *calldata)
{
	struct nfs4_opendata *data = calldata;

	data->rpc_status = task->tk_status;
	if (data->rpc_status == 0) {
		nfs4_stateid_copy(&data->o_res.stateid, &data->c_res.stateid);
		nfs_confirm_seqid(&data->owner->so_seqid, 0);
		renew_lease(data->o_res.server, data->timestamp);
		data->rpc_done = 1;
	}
}

static void nfs4_open_confirm_release(void *calldata)
{
	struct nfs4_opendata *data = calldata;
	struct nfs4_state *state = NULL;

	/* If this request hasn't been cancelled, do nothing */
	if (data->cancelled == 0)
		goto out_free;
	/* In case of error, no cleanup! */
	if (!data->rpc_done)
		goto out_free;
	state = nfs4_opendata_to_nfs4_state(data);
	if (!IS_ERR(state))
		nfs4_close_state(state, data->o_arg.fmode);
out_free:
	nfs4_opendata_put(data);
}

static const struct rpc_call_ops nfs4_open_confirm_ops = {
	.rpc_call_done = nfs4_open_confirm_done,
	.rpc_release = nfs4_open_confirm_release,
};

/*
 * Note: On error, nfs4_proc_open_confirm will free the struct nfs4_opendata
 */
static int _nfs4_proc_open_confirm(struct nfs4_opendata *data)
{
	struct nfs_server *server = NFS_SERVER(data->dir->d_inode);
	struct rpc_task *task;
	struct  rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_CONFIRM],
		.rpc_argp = &data->c_arg,
		.rpc_resp = &data->c_res,
		.rpc_cred = data->owner->so_cred,
	};
	struct rpc_task_setup task_setup_data = {
		.rpc_client = server->client,
		.rpc_message = &msg,
		.callback_ops = &nfs4_open_confirm_ops,
		.callback_data = data,
		.workqueue = nfsiod_workqueue,
		.flags = RPC_TASK_ASYNC,
	};
	int status;

	kref_get(&data->kref);
	data->rpc_done = 0;
	data->rpc_status = 0;
	data->timestamp = jiffies;
	task = rpc_run_task(&task_setup_data);
	if (IS_ERR(task))
		return PTR_ERR(task);
	status = nfs4_wait_for_completion_rpc_task(task);
	if (status != 0) {
		data->cancelled = 1;
		smp_wmb();
	} else
		status = data->rpc_status;
	rpc_put_task(task);
	return status;
}

static void nfs4_open_prepare(struct rpc_task *task, void *calldata)
{
	struct nfs4_opendata *data = calldata;
	struct nfs4_state_owner *sp = data->owner;

	if (nfs_wait_on_sequence(data->o_arg.seqid, task) != 0)
		return;
	/*
	 * Check if we still need to send an OPEN call, or if we can use
	 * a delegation instead.
	 */
	if (data->state != NULL) {
		struct nfs_delegation *delegation;

		if (can_open_cached(data->state, data->o_arg.fmode, data->o_arg.open_flags))
			goto out_no_action;
		rcu_read_lock();
		delegation = rcu_dereference(NFS_I(data->state->inode)->delegation);
		if (data->o_arg.claim != NFS4_OPEN_CLAIM_DELEGATE_CUR &&
		    can_open_delegated(delegation, data->o_arg.fmode))
			goto unlock_no_action;
		rcu_read_unlock();
	}
	/* Update client id. */
	data->o_arg.clientid = sp->so_server->nfs_client->cl_clientid;
	if (data->o_arg.claim == NFS4_OPEN_CLAIM_PREVIOUS) {
		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR];
		nfs_copy_fh(&data->o_res.fh, data->o_arg.fh);
	}
	data->timestamp = jiffies;
	if (nfs4_setup_sequence(data->o_arg.server,
				&data->o_arg.seq_args,
				&data->o_res.seq_res, task))
		return;
	rpc_call_start(task);
	return;
unlock_no_action:
	rcu_read_unlock();
out_no_action:
	task->tk_action = NULL;

}

static void nfs4_recover_open_prepare(struct rpc_task *task, void *calldata)
{
	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
	nfs4_open_prepare(task, calldata);
}

static void nfs4_open_done(struct rpc_task *task, void *calldata)
{
	struct nfs4_opendata *data = calldata;

	data->rpc_status = task->tk_status;

	if (!nfs4_sequence_done(task, &data->o_res.seq_res))
		return;

	if (task->tk_status == 0) {
		switch (data->o_res.f_attr->mode & S_IFMT) {
			case S_IFREG:
				break;
			case S_IFLNK:
				data->rpc_status = -ELOOP;
				break;
			case S_IFDIR:
				data->rpc_status = -EISDIR;
				break;
			default:
				data->rpc_status = -ENOTDIR;
		}
		renew_lease(data->o_res.server, data->timestamp);
		if (!(data->o_res.rflags & NFS4_OPEN_RESULT_CONFIRM))
			nfs_confirm_seqid(&data->owner->so_seqid, 0);
	}
	data->rpc_done = 1;
}

static void nfs4_open_release(void *calldata)
{
	struct nfs4_opendata *data = calldata;
	struct nfs4_state *state = NULL;

	/* If this request hasn't been cancelled, do nothing */
	if (data->cancelled == 0)
		goto out_free;
	/* In case of error, no cleanup! */
	if (data->rpc_status != 0 || !data->rpc_done)
		goto out_free;
	/* In case we need an open_confirm, no cleanup! */
	if (data->o_res.rflags & NFS4_OPEN_RESULT_CONFIRM)
		goto out_free;
	state = nfs4_opendata_to_nfs4_state(data);
	if (!IS_ERR(state))
		nfs4_close_state(state, data->o_arg.fmode);
out_free:
	nfs4_opendata_put(data);
}

static const struct rpc_call_ops nfs4_open_ops = {
	.rpc_call_prepare = nfs4_open_prepare,
	.rpc_call_done = nfs4_open_done,
	.rpc_release = nfs4_open_release,
};

static const struct rpc_call_ops nfs4_recover_open_ops = {
	.rpc_call_prepare = nfs4_recover_open_prepare,
	.rpc_call_done = nfs4_open_done,
	.rpc_release = nfs4_open_release,
};

static int nfs4_run_open_task(struct nfs4_opendata *data, int isrecover)
{
	struct inode *dir = data->dir->d_inode;
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs_openargs *o_arg = &data->o_arg;
	struct nfs_openres *o_res = &data->o_res;
	struct rpc_task *task;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN],
		.rpc_argp = o_arg,
		.rpc_resp = o_res,
		.rpc_cred = data->owner->so_cred,
	};
	struct rpc_task_setup task_setup_data = {
		.rpc_client = server->client,
		.rpc_message = &msg,
		.callback_ops = &nfs4_open_ops,
		.callback_data = data,
		.workqueue = nfsiod_workqueue,
		.flags = RPC_TASK_ASYNC,
	};
	int status;

	nfs41_init_sequence(&o_arg->seq_args, &o_res->seq_res, 1);
	kref_get(&data->kref);
	data->rpc_done = 0;
	data->rpc_status = 0;
	data->cancelled = 0;
	if (isrecover)
		task_setup_data.callback_ops = &nfs4_recover_open_ops;
	task = rpc_run_task(&task_setup_data);
        if (IS_ERR(task))
                return PTR_ERR(task);
        status = nfs4_wait_for_completion_rpc_task(task);
        if (status != 0) {
                data->cancelled = 1;
                smp_wmb();
        } else
                status = data->rpc_status;
        rpc_put_task(task);

	return status;
}

static int _nfs4_recover_proc_open(struct nfs4_opendata *data)
{
	struct inode *dir = data->dir->d_inode;
	struct nfs_openres *o_res = &data->o_res;
        int status;

	status = nfs4_run_open_task(data, 1);
	if (status != 0 || !data->rpc_done)
		return status;

	nfs_fattr_map_and_free_names(NFS_SERVER(dir), &data->f_attr);

	if (o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) {
		status = _nfs4_proc_open_confirm(data);
		if (status != 0)
			return status;
	}

	return status;
}

/*
 * Note: On error, nfs4_proc_open will free the struct nfs4_opendata
 */
static int _nfs4_proc_open(struct nfs4_opendata *data)
{
	struct inode *dir = data->dir->d_inode;
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs_openargs *o_arg = &data->o_arg;
	struct nfs_openres *o_res = &data->o_res;
	int status;

	status = nfs4_run_open_task(data, 0);
	if (!data->rpc_done)
		return status;
	if (status != 0) {
		if (status == -NFS4ERR_BADNAME &&
				!(o_arg->open_flags & O_CREAT))
			return -ENOENT;
		return status;
	}

	nfs_fattr_map_and_free_names(server, &data->f_attr);

	if (o_arg->open_flags & O_CREAT)
		update_changeattr(dir, &o_res->cinfo);
	if ((o_res->rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) == 0)
		server->caps &= ~NFS_CAP_POSIX_LOCK;
	if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) {
		status = _nfs4_proc_open_confirm(data);
		if (status != 0)
			return status;
	}
	if (!(o_res->f_attr->valid & NFS_ATTR_FATTR))
		_nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr);
	return 0;
}

static int nfs4_client_recover_expired_lease(struct nfs_client *clp)
{
	unsigned int loop;
	int ret;

	for (loop = NFS4_MAX_LOOP_ON_RECOVER; loop != 0; loop--) {
		ret = nfs4_wait_clnt_recover(clp);
		if (ret != 0)
			break;
		if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) &&
		    !test_bit(NFS4CLNT_CHECK_LEASE,&clp->cl_state))
			break;
		nfs4_schedule_state_manager(clp);
		ret = -EIO;
	}
	return ret;
}

static int nfs4_recover_expired_lease(struct nfs_server *server)
{
	return nfs4_client_recover_expired_lease(server->nfs_client);
}

/*
 * OPEN_EXPIRED:
 * 	reclaim state on the server after a network partition.
 * 	Assumes caller holds the appropriate lock
 */
static int _nfs4_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state)
{
	struct nfs4_opendata *opendata;
	int ret;

	opendata = nfs4_open_recoverdata_alloc(ctx, state);
	if (IS_ERR(opendata))
		return PTR_ERR(opendata);
	ret = nfs4_open_recover(opendata, state);
	if (ret == -ESTALE)
		d_drop(ctx->dentry);
	nfs4_opendata_put(opendata);
	return ret;
}

static int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state)
{
	struct nfs_server *server = NFS_SERVER(state->inode);
	struct nfs4_exception exception = { };
	int err;

	do {
		err = _nfs4_open_expired(ctx, state);
		switch (err) {
		default:
			goto out;
		case -NFS4ERR_GRACE:
		case -NFS4ERR_DELAY:
			nfs4_handle_exception(server, err, &exception);
			err = 0;
		}
	} while (exception.retry);
out:
	return err;
}

static int nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state)
{
	struct nfs_open_context *ctx;
	int ret;

	ctx = nfs4_state_find_open_context(state);
	if (IS_ERR(ctx))
		return PTR_ERR(ctx);
	ret = nfs4_do_open_expired(ctx, state);
	put_nfs_open_context(ctx);
	return ret;
}

#if defined(CONFIG_NFS_V4_1)
static int nfs41_check_expired_stateid(struct nfs4_state *state, nfs4_stateid *stateid, unsigned int flags)
{
	int status = NFS_OK;
	struct nfs_server *server = NFS_SERVER(state->inode);

	if (state->flags & flags) {
		status = nfs41_test_stateid(server, stateid);
		if (status != NFS_OK) {
			nfs41_free_stateid(server, stateid);
			state->flags &= ~flags;
		}
	}
	return status;
}

static int nfs41_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state)
{
	int deleg_status, open_status;
	int deleg_flags = 1 << NFS_DELEGATED_STATE;
	int open_flags = (1 << NFS_O_RDONLY_STATE) | (1 << NFS_O_WRONLY_STATE) | (1 << NFS_O_RDWR_STATE);

	deleg_status = nfs41_check_expired_stateid(state, &state->stateid, deleg_flags);
	open_status = nfs41_check_expired_stateid(state,  &state->open_stateid, open_flags);

	if ((deleg_status == NFS_OK) && (open_status == NFS_OK))
		return NFS_OK;
	return nfs4_open_expired(sp, state);
}
#endif

/*
 * on an EXCLUSIVE create, the server should send back a bitmask with FATTR4-*
 * fields corresponding to attributes that were used to store the verifier.
 * Make sure we clobber those fields in the later setattr call
 */
static inline void nfs4_exclusive_attrset(struct nfs4_opendata *opendata, struct iattr *sattr)
{
	if ((opendata->o_res.attrset[1] & FATTR4_WORD1_TIME_ACCESS) &&
	    !(sattr->ia_valid & ATTR_ATIME_SET))
		sattr->ia_valid |= ATTR_ATIME;

	if ((opendata->o_res.attrset[1] & FATTR4_WORD1_TIME_MODIFY) &&
	    !(sattr->ia_valid & ATTR_MTIME_SET))
		sattr->ia_valid |= ATTR_MTIME;
}

/*
 * Returns a referenced nfs4_state
 */
static int _nfs4_do_open(struct inode *dir,
			struct dentry *dentry,
			fmode_t fmode,
			int flags,
			struct iattr *sattr,
			struct rpc_cred *cred,
			struct nfs4_state **res,
			struct nfs4_threshold **ctx_th)
{
	struct nfs4_state_owner  *sp;
	struct nfs4_state     *state = NULL;
	struct nfs_server       *server = NFS_SERVER(dir);
	struct nfs4_opendata *opendata;
	int status;

	/* Protect against reboot recovery conflicts */
	status = -ENOMEM;
	sp = nfs4_get_state_owner(server, cred, GFP_KERNEL);
	if (sp == NULL) {
		dprintk("nfs4_do_open: nfs4_get_state_owner failed!\n");
		goto out_err;
	}
	status = nfs4_recover_expired_lease(server);
	if (status != 0)
		goto err_put_state_owner;
	if (dentry->d_inode != NULL)
		nfs4_return_incompatible_delegation(dentry->d_inode, fmode);
	status = -ENOMEM;
	opendata = nfs4_opendata_alloc(dentry, sp, fmode, flags, sattr, GFP_KERNEL);
	if (opendata == NULL)
		goto err_put_state_owner;

	if (ctx_th && server->attr_bitmask[2] & FATTR4_WORD2_MDSTHRESHOLD) {
		opendata->f_attr.mdsthreshold = pnfs_mdsthreshold_alloc();
		if (!opendata->f_attr.mdsthreshold)
			goto err_opendata_put;
	}
	if (dentry->d_inode != NULL)
		opendata->state = nfs4_get_open_state(dentry->d_inode, sp);

	status = _nfs4_proc_open(opendata);
	if (status != 0)
		goto err_opendata_put;

	state = nfs4_opendata_to_nfs4_state(opendata);
	status = PTR_ERR(state);
	if (IS_ERR(state))
		goto err_opendata_put;
	if (server->caps & NFS_CAP_POSIX_LOCK)
		set_bit(NFS_STATE_POSIX_LOCKS, &state->flags);

	if (opendata->o_arg.open_flags & O_EXCL) {
		nfs4_exclusive_attrset(opendata, sattr);

		nfs_fattr_init(opendata->o_res.f_attr);
		status = nfs4_do_setattr(state->inode, cred,
				opendata->o_res.f_attr, sattr,
				state);
		if (status == 0)
			nfs_setattr_update_inode(state->inode, sattr);
		nfs_post_op_update_inode(state->inode, opendata->o_res.f_attr);
	}

	if (pnfs_use_threshold(ctx_th, opendata->f_attr.mdsthreshold, server))
		*ctx_th = opendata->f_attr.mdsthreshold;
	else
		kfree(opendata->f_attr.mdsthreshold);
	opendata->f_attr.mdsthreshold = NULL;

	nfs4_opendata_put(opendata);
	nfs4_put_state_owner(sp);
	*res = state;
	return 0;
err_opendata_put:
	kfree(opendata->f_attr.mdsthreshold);
	nfs4_opendata_put(opendata);
err_put_state_owner:
	nfs4_put_state_owner(sp);
out_err:
	*res = NULL;
	return status;
}


static struct nfs4_state *nfs4_do_open(struct inode *dir,
					struct dentry *dentry,
					fmode_t fmode,
					int flags,
					struct iattr *sattr,
					struct rpc_cred *cred,
					struct nfs4_threshold **ctx_th)
{
	struct nfs4_exception exception = { };
	struct nfs4_state *res;
	int status;

	do {
		status = _nfs4_do_open(dir, dentry, fmode, flags, sattr, cred,
				       &res, ctx_th);
		if (status == 0)
			break;
		/* NOTE: BAD_SEQID means the server and client disagree about the
		 * book-keeping w.r.t. state-changing operations
		 * (OPEN/CLOSE/LOCK/LOCKU...)
		 * It is actually a sign of a bug on the client or on the server.
		 *
		 * If we receive a BAD_SEQID error in the particular case of
		 * doing an OPEN, we assume that nfs_increment_open_seqid() will
		 * have unhashed the old state_owner for us, and that we can
		 * therefore safely retry using a new one. We should still warn
		 * the user though...
		 */
		if (status == -NFS4ERR_BAD_SEQID) {
			pr_warn_ratelimited("NFS: v4 server %s "
					" returned a bad sequence-id error!\n",
					NFS_SERVER(dir)->nfs_client->cl_hostname);
			exception.retry = 1;
			continue;
		}
		/*
		 * BAD_STATEID on OPEN means that the server cancelled our
		 * state before it received the OPEN_CONFIRM.
		 * Recover by retrying the request as per the discussion
		 * on Page 181 of RFC3530.
		 */
		if (status == -NFS4ERR_BAD_STATEID) {
			exception.retry = 1;
			continue;
		}
		if (status == -EAGAIN) {
			/* We must have found a delegation */
			exception.retry = 1;
			continue;
		}
		res = ERR_PTR(nfs4_handle_exception(NFS_SERVER(dir),
					status, &exception));
	} while (exception.retry);
	return res;
}

static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
			    struct nfs_fattr *fattr, struct iattr *sattr,
			    struct nfs4_state *state)
{
	struct nfs_server *server = NFS_SERVER(inode);
        struct nfs_setattrargs  arg = {
                .fh             = NFS_FH(inode),
                .iap            = sattr,
		.server		= server,
		.bitmask = server->attr_bitmask,
        };
        struct nfs_setattrres  res = {
		.fattr		= fattr,
		.server		= server,
        };
        struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_SETATTR],
		.rpc_argp	= &arg,
		.rpc_resp	= &res,
		.rpc_cred	= cred,
        };
	unsigned long timestamp = jiffies;
	int status;

	nfs_fattr_init(fattr);

	if (state != NULL) {
		nfs4_select_rw_stateid(&arg.stateid, state, FMODE_WRITE,
				current->files, current->tgid);
	} else if (nfs4_copy_delegation_stateid(&arg.stateid, inode,
				FMODE_WRITE)) {
		/* Use that stateid */
	} else
		nfs4_stateid_copy(&arg.stateid, &zero_stateid);

	status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1);
	if (status == 0 && state != NULL)
		renew_lease(server, timestamp);
	return status;
}

static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
			   struct nfs_fattr *fattr, struct iattr *sattr,
			   struct nfs4_state *state)
{
	struct nfs_server *server = NFS_SERVER(inode);
	struct nfs4_exception exception = {
		.state = state,
		.inode = inode,
	};
	int err;
	do {
		err = _nfs4_do_setattr(inode, cred, fattr, sattr, state);
		switch (err) {
		case -NFS4ERR_OPENMODE:
			if (state && !(state->state & FMODE_WRITE)) {
				err = -EBADF;
				if (sattr->ia_valid & ATTR_OPEN)
					err = -EACCES;
				goto out;
			}
		}
		err = nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
out:
	return err;
}

struct nfs4_closedata {
	struct inode *inode;
	struct nfs4_state *state;
	struct nfs_closeargs arg;
	struct nfs_closeres res;
	struct nfs_fattr fattr;
	unsigned long timestamp;
	bool roc;
	u32 roc_barrier;
};

static void nfs4_free_closedata(void *data)
{
	struct nfs4_closedata *calldata = data;
	struct nfs4_state_owner *sp = calldata->state->owner;
	struct super_block *sb = calldata->state->inode->i_sb;

	if (calldata->roc)
		pnfs_roc_release(calldata->state->inode);
	nfs4_put_open_state(calldata->state);
	nfs_free_seqid(calldata->arg.seqid);
	nfs4_put_state_owner(sp);
	nfs_sb_deactive(sb);
	kfree(calldata);
}

static void nfs4_close_clear_stateid_flags(struct nfs4_state *state,
		fmode_t fmode)
{
	spin_lock(&state->owner->so_lock);
	if (!(fmode & FMODE_READ))
		clear_bit(NFS_O_RDONLY_STATE, &state->flags);
	if (!(fmode & FMODE_WRITE))
		clear_bit(NFS_O_WRONLY_STATE, &state->flags);
	clear_bit(NFS_O_RDWR_STATE, &state->flags);
	spin_unlock(&state->owner->so_lock);
}

static void nfs4_close_done(struct rpc_task *task, void *data)
{
	struct nfs4_closedata *calldata = data;
	struct nfs4_state *state = calldata->state;
	struct nfs_server *server = NFS_SERVER(calldata->inode);

	dprintk("%s: begin!\n", __func__);
	if (!nfs4_sequence_done(task, &calldata->res.seq_res))
		return;
        /* hmm. we are done with the inode, and in the process of freeing
	 * the state_owner. we keep this around to process errors
	 */
	switch (task->tk_status) {
		case 0:
			if (calldata->roc)
				pnfs_roc_set_barrier(state->inode,
						     calldata->roc_barrier);
			nfs_set_open_stateid(state, &calldata->res.stateid, 0);
			renew_lease(server, calldata->timestamp);
			nfs4_close_clear_stateid_flags(state,
					calldata->arg.fmode);
			break;
		case -NFS4ERR_STALE_STATEID:
		case -NFS4ERR_OLD_STATEID:
		case -NFS4ERR_BAD_STATEID:
		case -NFS4ERR_EXPIRED:
			if (calldata->arg.fmode == 0)
				break;
		default:
			if (nfs4_async_handle_error(task, server, state) == -EAGAIN)
				rpc_restart_call_prepare(task);
	}
	nfs_release_seqid(calldata->arg.seqid);
	nfs_refresh_inode(calldata->inode, calldata->res.fattr);
	dprintk("%s: done, ret = %d!\n", __func__, task->tk_status);
}

static void nfs4_close_prepare(struct rpc_task *task, void *data)
{
	struct nfs4_closedata *calldata = data;
	struct nfs4_state *state = calldata->state;
	int call_close = 0;

	dprintk("%s: begin!\n", __func__);
	if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0)
		return;

	task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
	calldata->arg.fmode = FMODE_READ|FMODE_WRITE;
	spin_lock(&state->owner->so_lock);
	/* Calculate the change in open mode */
	if (state->n_rdwr == 0) {
		if (state->n_rdonly == 0) {
			call_close |= test_bit(NFS_O_RDONLY_STATE, &state->flags);
			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
			calldata->arg.fmode &= ~FMODE_READ;
		}
		if (state->n_wronly == 0) {
			call_close |= test_bit(NFS_O_WRONLY_STATE, &state->flags);
			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
			calldata->arg.fmode &= ~FMODE_WRITE;
		}
	}
	spin_unlock(&state->owner->so_lock);

	if (!call_close) {
		/* Note: exit _without_ calling nfs4_close_done */
		task->tk_action = NULL;
		goto out;
	}

	if (calldata->arg.fmode == 0) {
		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE];
		if (calldata->roc &&
		    pnfs_roc_drain(calldata->inode, &calldata->roc_barrier)) {
			rpc_sleep_on(&NFS_SERVER(calldata->inode)->roc_rpcwaitq,
				     task, NULL);
			goto out;
		}
	}

	nfs_fattr_init(calldata->res.fattr);
	calldata->timestamp = jiffies;
	if (nfs4_setup_sequence(NFS_SERVER(calldata->inode),
				&calldata->arg.seq_args,
				&calldata->res.seq_res,
				task))
		goto out;
	rpc_call_start(task);
out:
	dprintk("%s: done!\n", __func__);
}

static const struct rpc_call_ops nfs4_close_ops = {
	.rpc_call_prepare = nfs4_close_prepare,
	.rpc_call_done = nfs4_close_done,
	.rpc_release = nfs4_free_closedata,
};

/* 
 * It is possible for data to be read/written from a mem-mapped file 
 * after the sys_close call (which hits the vfs layer as a flush).
 * This means that we can't safely call nfsv4 close on a file until 
 * the inode is cleared. This in turn means that we are not good
 * NFSv4 citizens - we do not indicate to the server to update the file's 
 * share state even when we are done with one of the three share 
 * stateid's in the inode.
 *
 * NOTE: Caller must be holding the sp->so_owner semaphore!
 */
int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait, bool roc)
{
	struct nfs_server *server = NFS_SERVER(state->inode);
	struct nfs4_closedata *calldata;
	struct nfs4_state_owner *sp = state->owner;
	struct rpc_task *task;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE],
		.rpc_cred = state->owner->so_cred,
	};
	struct rpc_task_setup task_setup_data = {
		.rpc_client = server->client,
		.rpc_message = &msg,
		.callback_ops = &nfs4_close_ops,
		.workqueue = nfsiod_workqueue,
		.flags = RPC_TASK_ASYNC,
	};
	int status = -ENOMEM;

	calldata = kzalloc(sizeof(*calldata), gfp_mask);
	if (calldata == NULL)
		goto out;
	nfs41_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 1);
	calldata->inode = state->inode;
	calldata->state = state;
	calldata->arg.fh = NFS_FH(state->inode);
	calldata->arg.stateid = &state->open_stateid;
	/* Serialization for the sequence id */
	calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid, gfp_mask);
	if (calldata->arg.seqid == NULL)
		goto out_free_calldata;
	calldata->arg.fmode = 0;
	calldata->arg.bitmask = server->cache_consistency_bitmask;
	calldata->res.fattr = &calldata->fattr;
	calldata->res.seqid = calldata->arg.seqid;
	calldata->res.server = server;
	calldata->roc = roc;
	nfs_sb_active(calldata->inode->i_sb);

	msg.rpc_argp = &calldata->arg;
	msg.rpc_resp = &calldata->res;
	task_setup_data.callback_data = calldata;
	task = rpc_run_task(&task_setup_data);
	if (IS_ERR(task))
		return PTR_ERR(task);
	status = 0;
	if (wait)
		status = rpc_wait_for_completion_task(task);
	rpc_put_task(task);
	return status;
out_free_calldata:
	kfree(calldata);
out:
	if (roc)
		pnfs_roc_release(state->inode);
	nfs4_put_open_state(state);
	nfs4_put_state_owner(sp);
	return status;
}

static struct inode *
nfs4_atomic_open(struct inode *dir, struct nfs_open_context *ctx, int open_flags, struct iattr *attr)
{
	struct nfs4_state *state;

	/* Protect against concurrent sillydeletes */
	state = nfs4_do_open(dir, ctx->dentry, ctx->mode, open_flags, attr,
			     ctx->cred, &ctx->mdsthreshold);
	if (IS_ERR(state))
		return ERR_CAST(state);
	ctx->state = state;
	return igrab(state->inode);
}

static void nfs4_close_context(struct nfs_open_context *ctx, int is_sync)
{
	if (ctx->state == NULL)
		return;
	if (is_sync)
		nfs4_close_sync(ctx->state, ctx->mode);
	else
		nfs4_close_state(ctx->state, ctx->mode);
}

static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle)
{
	struct nfs4_server_caps_arg args = {
		.fhandle = fhandle,
	};
	struct nfs4_server_caps_res res = {};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SERVER_CAPS],
		.rpc_argp = &args,
		.rpc_resp = &res,
	};
	int status;

	status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
	if (status == 0) {
		memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask));
		server->caps &= ~(NFS_CAP_ACLS|NFS_CAP_HARDLINKS|
				NFS_CAP_SYMLINKS|NFS_CAP_FILEID|
				NFS_CAP_MODE|NFS_CAP_NLINK|NFS_CAP_OWNER|
				NFS_CAP_OWNER_GROUP|NFS_CAP_ATIME|
				NFS_CAP_CTIME|NFS_CAP_MTIME);
		if (res.attr_bitmask[0] & FATTR4_WORD0_ACL)
			server->caps |= NFS_CAP_ACLS;
		if (res.has_links != 0)
			server->caps |= NFS_CAP_HARDLINKS;
		if (res.has_symlinks != 0)
			server->caps |= NFS_CAP_SYMLINKS;
		if (res.attr_bitmask[0] & FATTR4_WORD0_FILEID)
			server->caps |= NFS_CAP_FILEID;
		if (res.attr_bitmask[1] & FATTR4_WORD1_MODE)
			server->caps |= NFS_CAP_MODE;
		if (res.attr_bitmask[1] & FATTR4_WORD1_NUMLINKS)
			server->caps |= NFS_CAP_NLINK;
		if (res.attr_bitmask[1] & FATTR4_WORD1_OWNER)
			server->caps |= NFS_CAP_OWNER;
		if (res.attr_bitmask[1] & FATTR4_WORD1_OWNER_GROUP)
			server->caps |= NFS_CAP_OWNER_GROUP;
		if (res.attr_bitmask[1] & FATTR4_WORD1_TIME_ACCESS)
			server->caps |= NFS_CAP_ATIME;
		if (res.attr_bitmask[1] & FATTR4_WORD1_TIME_METADATA)
			server->caps |= NFS_CAP_CTIME;
		if (res.attr_bitmask[1] & FATTR4_WORD1_TIME_MODIFY)
			server->caps |= NFS_CAP_MTIME;

		memcpy(server->cache_consistency_bitmask, res.attr_bitmask, sizeof(server->cache_consistency_bitmask));
		server->cache_consistency_bitmask[0] &= FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE;
		server->cache_consistency_bitmask[1] &= FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
		server->acl_bitmask = res.acl_bitmask;
		server->fh_expire_type = res.fh_expire_type;
	}

	return status;
}

int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(server,
				_nfs4_server_capabilities(server, fhandle),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
		struct nfs_fsinfo *info)
{
	struct nfs4_lookup_root_arg args = {
		.bitmask = nfs4_fattr_bitmap,
	};
	struct nfs4_lookup_res res = {
		.server = server,
		.fattr = info->fattr,
		.fh = fhandle,
	};
	stru