nfs4proc.c

	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;
	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 (delegation == NULL ||
		    !can_open_delegated(delegation, fmode)) {
			rcu_read_unlock();
			break;
		}
		/* Save the delegation */
		memcpy(stateid.data, delegation->stateid.data, sizeof(stateid.data));
		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) {
		int delegation_flags = 0;

		rcu_read_lock();
		delegation = rcu_dereference(NFS_I(inode)->delegation);
		if (delegation)
			delegation_flags = delegation->flags;
		rcu_read_unlock();
		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->path, 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(&opendata->path, 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 &&
	    memcmp(state->stateid.data, state->open_stateid.data, sizeof(state->stateid.data)) != 0) {
		write_seqlock(&state->seqlock);
		if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0)
			memcpy(state->stateid.data, state->open_stateid.data, sizeof(state->stateid.data));
		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;
	memcpy(opendata->o_arg.u.delegation.data, stateid->data,
			sizeof(opendata->o_arg.u.delegation.data));
	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);
				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_ADMIN_REVOKED:
			case -NFS4ERR_BAD_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) {
		memcpy(data->o_res.stateid.data, data->c_res.stateid.data,
				sizeof(data->o_res.stateid.data));
		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(&data->path, 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 (delegation != NULL &&
		    test_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags) == 0) {
			rcu_read_unlock();
			goto out_no_action;
		}
		rcu_read_unlock();
	}
	/* Update sequence id. */
	data->o_arg.id = sp->so_owner_id.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, 1, task))
		return;
	rpc_call_start(task);
	return;
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(&data->path, 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;

	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_refresh_inode(dir, o_res->dir_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 (status != 0 || !data->rpc_done)
		return status;

	if (o_arg->open_flags & O_CREAT) {
		update_changeattr(dir, &o_res->cinfo);
		nfs_post_op_update_inode(dir, o_res->dir_attr);
	} else
		nfs_refresh_inode(dir, o_res->dir_attr);
	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->path.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;
}

/*
 * 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 path *path, fmode_t fmode, int flags, struct iattr *sattr, struct rpc_cred *cred, struct nfs4_state **res)
{
	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;
	if (!(sp = nfs4_get_state_owner(server, cred))) {
		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 (path->dentry->d_inode != NULL)
		nfs4_return_incompatible_delegation(path->dentry->d_inode, fmode);
	status = -ENOMEM;
	opendata = nfs4_opendata_alloc(path, sp, fmode, flags, sattr, GFP_KERNEL);
	if (opendata == NULL)
		goto err_put_state_owner;

	if (path->dentry->d_inode != NULL)
		opendata->state = nfs4_get_open_state(path->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);
	}
	nfs4_opendata_put(opendata);
	nfs4_put_state_owner(sp);
	*res = state;
	return 0;
err_opendata_put:
	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 path *path, fmode_t fmode, int flags, struct iattr *sattr, struct rpc_cred *cred)
{
	struct nfs4_exception exception = { };
	struct nfs4_state *res;
	int status;

	do {
		status = _nfs4_do_open(dir, path, fmode, flags, sattr, cred, &res);
		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) {
			printk(KERN_WARNING "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 (nfs4_copy_delegation_stateid(&arg.stateid, inode)) {
		/* Use that stateid */
	} else if (state != NULL) {
		nfs4_copy_stateid(&arg.stateid, state, current->files, current->tgid);
	} else
		memcpy(&arg.stateid, &zero_stateid, sizeof(arg.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 = { };
	int err;
	do {
		err = nfs4_handle_exception(server,
				_nfs4_do_setattr(inode, cred, fattr, sattr, state),
				&exception);
	} while (exception.retry);
	return err;
}

struct nfs4_closedata {
	struct path path;
	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;

	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);
	path_put(&calldata->path);
	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);

	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);
}

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;

	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;
		return;
	}

	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);
			return;
		}
	}

	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,
				1, task))
		return;
	rpc_call_start(task);
}

static const struct rpc_call_ops nfs4_close_ops = {
	.rpc_call_prepare = nfs4_close_prepare,
	.rpc_call_done = nfs4_close_done,