inode.c

void nfs_file_set_open_context(struct file *filp, struct nfs_open_context *ctx)
{
	struct inode *inode = filp->f_dentry->d_inode;
	struct nfs_inode *nfsi = NFS_I(inode);

	filp->private_data = get_nfs_open_context(ctx);
	spin_lock(&inode->i_lock);
	list_add(&ctx->list, &nfsi->open_files);
	spin_unlock(&inode->i_lock);
}

struct nfs_open_context *nfs_find_open_context(struct inode *inode, int mode)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	struct nfs_open_context *pos, *ctx = NULL;

	spin_lock(&inode->i_lock);
	list_for_each_entry(pos, &nfsi->open_files, list) {
		if ((pos->mode & mode) == mode) {
			ctx = get_nfs_open_context(pos);
			break;
		}
	}
	spin_unlock(&inode->i_lock);
	return ctx;
}

void nfs_file_clear_open_context(struct file *filp)
{
	struct inode *inode = filp->f_dentry->d_inode;
	struct nfs_open_context *ctx = (struct nfs_open_context *)filp->private_data;

	if (ctx) {
		filp->private_data = NULL;
		spin_lock(&inode->i_lock);
		list_move_tail(&ctx->list, &NFS_I(inode)->open_files);
		spin_unlock(&inode->i_lock);
		put_nfs_open_context(ctx);
	}
}

/*
 * These allocate and release file read/write context information.
 */
int nfs_open(struct inode *inode, struct file *filp)
{
	struct nfs_open_context *ctx;
	struct rpc_cred *cred;

	cred = rpcauth_lookupcred(NFS_CLIENT(inode)->cl_auth, 0);
	if (IS_ERR(cred))
		return PTR_ERR(cred);
	ctx = alloc_nfs_open_context(filp->f_dentry, cred);
	put_rpccred(cred);
	if (ctx == NULL)
		return -ENOMEM;
	ctx->mode = filp->f_mode;
	nfs_file_set_open_context(filp, ctx);
	put_nfs_open_context(ctx);
	if ((filp->f_mode & FMODE_WRITE) != 0)
		nfs_begin_data_update(inode);
	return 0;
}

int nfs_release(struct inode *inode, struct file *filp)
{
	if ((filp->f_mode & FMODE_WRITE) != 0)
		nfs_end_data_update(inode);
	nfs_file_clear_open_context(filp);
	return 0;
}

/*
 * This function is called whenever some part of NFS notices that
 * the cached attributes have to be refreshed.
 */
int
__nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
{
	int		 status = -ESTALE;
	struct nfs_fattr fattr;
	struct nfs_inode *nfsi = NFS_I(inode);
	unsigned long verifier;
	unsigned long cache_validity;

	dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n",
		inode->i_sb->s_id, (long long)NFS_FILEID(inode));

	lock_kernel();
	if (!inode || is_bad_inode(inode))
 		goto out_nowait;
	if (NFS_STALE(inode))
 		goto out_nowait;

	status = nfs_wait_on_inode(inode);
	if (status < 0)
		goto out;
	if (NFS_STALE(inode)) {
		status = -ESTALE;
		/* Do we trust the cached ESTALE? */
		if (NFS_ATTRTIMEO(inode) != 0) {
			if (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ATIME)) {
				/* no */
			} else
				goto out;
		}
	}

	/* Protect against RPC races by saving the change attribute */
	verifier = nfs_save_change_attribute(inode);
	status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), &fattr);
	if (status != 0) {
		dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) getattr failed, error=%d\n",
			 inode->i_sb->s_id,
			 (long long)NFS_FILEID(inode), status);
		if (status == -ESTALE) {
			nfs_zap_caches(inode);
			if (!S_ISDIR(inode->i_mode))
				set_bit(NFS_INO_STALE, &NFS_FLAGS(inode));
		}
		goto out;
	}

	spin_lock(&inode->i_lock);
	status = nfs_update_inode(inode, &fattr, verifier);
	if (status) {
		spin_unlock(&inode->i_lock);
		dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) refresh failed, error=%d\n",
			 inode->i_sb->s_id,
			 (long long)NFS_FILEID(inode), status);
		goto out;
	}
	cache_validity = nfsi->cache_validity;
	nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;

	/*
	 * We may need to keep the attributes marked as invalid if
	 * we raced with nfs_end_attr_update().
	 */
	if (time_after_eq(verifier, nfsi->cache_change_attribute))
		nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
	spin_unlock(&inode->i_lock);

	nfs_revalidate_mapping(inode, inode->i_mapping);

	if (cache_validity & NFS_INO_INVALID_ACL)
		nfs_zap_acl_cache(inode);

	dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n",
		inode->i_sb->s_id,
		(long long)NFS_FILEID(inode));

 out:
	nfs_wake_up_inode(inode);

 out_nowait:
	unlock_kernel();
	return status;
}

int nfs_attribute_timeout(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	if (nfs_have_delegation(inode, FMODE_READ))
		return 0;
	return time_after(jiffies, nfsi->read_cache_jiffies+nfsi->attrtimeo);
}

/**
 * nfs_revalidate_inode - Revalidate the inode attributes
 * @server - pointer to nfs_server struct
 * @inode - pointer to inode struct
 *
 * Updates inode attribute information by retrieving the data from the server.
 */
int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
{
	if (!(NFS_I(inode)->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA))
			&& !nfs_attribute_timeout(inode))
		return NFS_STALE(inode) ? -ESTALE : 0;
	return __nfs_revalidate_inode(server, inode);
}

/**
 * nfs_revalidate_mapping - Revalidate the pagecache
 * @inode - pointer to host inode
 * @mapping - pointer to mapping
 */
void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	if (nfsi->cache_validity & NFS_INO_INVALID_DATA) {
		if (S_ISREG(inode->i_mode)) {
			if (filemap_fdatawrite(mapping) == 0)
				filemap_fdatawait(mapping);
			nfs_wb_all(inode);
		}
		invalidate_inode_pages2(mapping);

		spin_lock(&inode->i_lock);
		nfsi->cache_validity &= ~NFS_INO_INVALID_DATA;
		if (S_ISDIR(inode->i_mode)) {
			memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
			/* This ensures we revalidate child dentries */
			nfsi->cache_change_attribute = jiffies;
		}
		spin_unlock(&inode->i_lock);

		dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n",
				inode->i_sb->s_id,
				(long long)NFS_FILEID(inode));
	}
}

/**
 * nfs_begin_data_update
 * @inode - pointer to inode
 * Declare that a set of operations will update file data on the server
 */
void nfs_begin_data_update(struct inode *inode)
{
	atomic_inc(&NFS_I(inode)->data_updates);
}

/**
 * nfs_end_data_update
 * @inode - pointer to inode
 * Declare end of the operations that will update file data
 * This will mark the inode as immediately needing revalidation
 * of its attribute cache.
 */
void nfs_end_data_update(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	if (!nfs_have_delegation(inode, FMODE_READ)) {
		/* Mark the attribute cache for revalidation */
		spin_lock(&inode->i_lock);
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
		/* Directories and symlinks: invalidate page cache too */
		if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
			nfsi->cache_validity |= NFS_INO_INVALID_DATA;
		spin_unlock(&inode->i_lock);
	}
	nfsi->cache_change_attribute = jiffies;
	atomic_dec(&nfsi->data_updates);
}

/**
 * nfs_check_inode_attributes - verify consistency of the inode attribute cache
 * @inode - pointer to inode
 * @fattr - updated attributes
 *
 * Verifies the attribute cache. If we have just changed the attributes,
 * so that fattr carries weak cache consistency data, then it may
 * also update the ctime/mtime/change_attribute.
 */
static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fattr)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	loff_t cur_size, new_isize;
	int data_unstable;


	/* Are we in the process of updating data on the server? */
	data_unstable = nfs_caches_unstable(inode);

	if (fattr->valid & NFS_ATTR_FATTR_V4) {
		if ((fattr->valid & NFS_ATTR_PRE_CHANGE) != 0
				&& nfsi->change_attr == fattr->pre_change_attr)
			nfsi->change_attr = fattr->change_attr;
		if (nfsi->change_attr != fattr->change_attr) {
			nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
			if (!data_unstable)
				nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE;
		}
	}

	if ((fattr->valid & NFS_ATTR_FATTR) == 0) {
		spin_unlock(&inode->i_lock);
		return 0;
	}

	/* Has the inode gone and changed behind our back? */
	if (nfsi->fileid != fattr->fileid
			|| (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) {
		spin_unlock(&inode->i_lock);
		return -EIO;
	}

	cur_size = i_size_read(inode);
 	new_isize = nfs_size_to_loff_t(fattr->size);

	/* If we have atomic WCC data, we may update some attributes */
	if ((fattr->valid & NFS_ATTR_WCC) != 0) {
		if (timespec_equal(&inode->i_ctime, &fattr->pre_ctime))
			memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime));
		if (timespec_equal(&inode->i_mtime, &fattr->pre_mtime))
			memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime));
	}

	/* Verify a few of the more important attributes */
	if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) {
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
		if (!data_unstable)
			nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE;
	}
	if (cur_size != new_isize) {
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
		if (nfsi->npages == 0)
			nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE;
	}

	/* Have any file permissions changed? */
	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)
			|| inode->i_uid != fattr->uid
			|| inode->i_gid != fattr->gid)
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL;

	/* Has the link count changed? */
	if (inode->i_nlink != fattr->nlink)
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR;

	if (!timespec_equal(&inode->i_atime, &fattr->atime))
		nfsi->cache_validity |= NFS_INO_INVALID_ATIME;

	nfsi->read_cache_jiffies = fattr->time_start;
	return 0;
}

/**
 * nfs_refresh_inode - try to update the inode attribute cache
 * @inode - pointer to inode
 * @fattr - updated attributes
 *
 * Check that an RPC call that returned attributes has not overlapped with
 * other recent updates of the inode metadata, then decide whether it is
 * safe to do a full update of the inode attributes, or whether just to
 * call nfs_check_inode_attributes.
 */
int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	int status;

	if ((fattr->valid & NFS_ATTR_FATTR) == 0)
		return 0;
	spin_lock(&inode->i_lock);
	nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;
	if (nfs_verify_change_attribute(inode, fattr->time_start))
		nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
	if (time_after(fattr->time_start, nfsi->last_updated))
		status = nfs_update_inode(inode, fattr, fattr->time_start);
	else
		status = nfs_check_inode_attributes(inode, fattr);

	spin_unlock(&inode->i_lock);
	return status;
}

/*
 * Many nfs protocol calls return the new file attributes after
 * an operation.  Here we update the inode to reflect the state
 * of the server's inode.
 *
 * This is a bit tricky because we have to make sure all dirty pages
 * have been sent off to the server before calling invalidate_inode_pages.
 * To make sure no other process adds more write requests while we try
 * our best to flush them, we make them sleep during the attribute refresh.
 *
 * A very similar scenario holds for the dir cache.
 */
static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsigned long verifier)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	loff_t cur_isize, new_isize;
	unsigned int	invalid = 0;
	int data_unstable;

	dfprintk(VFS, "NFS: %s(%s/%ld ct=%d info=0x%x)\n",
			__FUNCTION__, inode->i_sb->s_id, inode->i_ino,
			atomic_read(&inode->i_count), fattr->valid);

	if ((fattr->valid & NFS_ATTR_FATTR) == 0)
		return 0;

	if (nfsi->fileid != fattr->fileid) {
		printk(KERN_ERR "%s: inode number mismatch\n"
		       "expected (%s/0x%Lx), got (%s/0x%Lx)\n",
		       __FUNCTION__,
		       inode->i_sb->s_id, (long long)nfsi->fileid,
		       inode->i_sb->s_id, (long long)fattr->fileid);
		goto out_err;
	}

	/*
	 * Make sure the inode's type hasn't changed.
	 */
	if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT))
		goto out_changed;

	/*
	 * Update the read time so we don't revalidate too often.
	 */
	nfsi->read_cache_jiffies = fattr->time_start;
	nfsi->last_updated = jiffies;

	/* Are we racing with known updates of the metadata on the server? */
	data_unstable = ! (nfs_verify_change_attribute(inode, verifier) ||
		(nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE));

	/* Check if our cached file size is stale */
 	new_isize = nfs_size_to_loff_t(fattr->size);
	cur_isize = i_size_read(inode);
	if (new_isize != cur_isize) {
		/* Do we perhaps have any outstanding writes? */
		if (nfsi->npages == 0) {
			/* No, but did we race with nfs_end_data_update()? */
			if (time_after_eq(verifier,  nfsi->cache_change_attribute)) {
				inode->i_size = new_isize;
				invalid |= NFS_INO_INVALID_DATA;
			}
			invalid |= NFS_INO_INVALID_ATTR;
		} else if (new_isize > cur_isize) {
			inode->i_size = new_isize;
			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
		}
		dprintk("NFS: isize change on server for file %s/%ld\n",
				inode->i_sb->s_id, inode->i_ino);
	}

	/* Check if the mtime agrees */
	if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) {
		memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime));
		dprintk("NFS: mtime change on server for file %s/%ld\n",
				inode->i_sb->s_id, inode->i_ino);
		if (!data_unstable)
			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
	}

	if ((fattr->valid & NFS_ATTR_FATTR_V4)
	    && nfsi->change_attr != fattr->change_attr) {
		dprintk("NFS: change_attr change on server for file %s/%ld\n",
		       inode->i_sb->s_id, inode->i_ino);
		nfsi->change_attr = fattr->change_attr;
		if (!data_unstable)
			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
	}

	/* If ctime has changed we should definitely clear access+acl caches */
	if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) {
		if (!data_unstable)
			invalid |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
		memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime));
	}
	memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime));

	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) ||
	    inode->i_uid != fattr->uid ||
	    inode->i_gid != fattr->gid)
		invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;

	inode->i_mode = fattr->mode;
	inode->i_nlink = fattr->nlink;
	inode->i_uid = fattr->uid;
	inode->i_gid = fattr->gid;

	if (fattr->valid & (NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4)) {
		/*
		 * report the blocks in 512byte units
		 */
		inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
		inode->i_blksize = inode->i_sb->s_blocksize;
 	} else {
 		inode->i_blocks = fattr->du.nfs2.blocks;
 		inode->i_blksize = fattr->du.nfs2.blocksize;
 	}

	/* Update attrtimeo value if we're out of the unstable period */
	if (invalid & NFS_INO_INVALID_ATTR) {
		nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
		nfsi->attrtimeo_timestamp = jiffies;
	} else if (time_after(jiffies, nfsi->attrtimeo_timestamp+nfsi->attrtimeo)) {
		if ((nfsi->attrtimeo <<= 1) > NFS_MAXATTRTIMEO(inode))
			nfsi->attrtimeo = NFS_MAXATTRTIMEO(inode);
		nfsi->attrtimeo_timestamp = jiffies;
	}
	/* Don't invalidate the data if we were to blame */
	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)
				|| S_ISLNK(inode->i_mode)))
		invalid &= ~NFS_INO_INVALID_DATA;
	if (!nfs_have_delegation(inode, FMODE_READ))
		nfsi->cache_validity |= invalid;

	return 0;
 out_changed:
	/*
	 * Big trouble! The inode has become a different object.
	 */
#ifdef NFS_PARANOIA
	printk(KERN_DEBUG "%s: inode %ld mode changed, %07o to %07o\n",
			__FUNCTION__, inode->i_ino, inode->i_mode, fattr->mode);
#endif
	/*
	 * No need to worry about unhashing the dentry, as the
	 * lookup validation will know that the inode is bad.
	 * (But we fall through to invalidate the caches.)
	 */
	nfs_invalidate_inode(inode);
 out_err:
	set_bit(NFS_INO_STALE, &NFS_FLAGS(inode));
	return -ESTALE;
}

/*
 * File system information
 */

static int nfs_set_super(struct super_block *s, void *data)
{
	s->s_fs_info = data;
	return set_anon_super(s, data);
}
 
static int nfs_compare_super(struct super_block *sb, void *data)
{
	struct nfs_server *server = data;
	struct nfs_server *old = NFS_SB(sb);

	if (old->addr.sin_addr.s_addr != server->addr.sin_addr.s_addr)
		return 0;
	if (old->addr.sin_port != server->addr.sin_port)
		return 0;
	return !nfs_compare_fh(&old->fh, &server->fh);
}

static struct super_block *nfs_get_sb(struct file_system_type *fs_type,
	int flags, const char *dev_name, void *raw_data)
{
	int error;
	struct nfs_server *server = NULL;
	struct super_block *s;
	struct nfs_fh *root;
	struct nfs_mount_data *data = raw_data;

	s = ERR_PTR(-EINVAL);
	if (data == NULL) {
		dprintk("%s: missing data argument\n", __FUNCTION__);
		goto out_err;
	}
	if (data->version <= 0 || data->version > NFS_MOUNT_VERSION) {
		dprintk("%s: bad mount version\n", __FUNCTION__);
		goto out_err;
	}
	switch (data->version) {
		case 1:
			data->namlen = 0;
		case 2:
			data->bsize  = 0;
		case 3:
			if (data->flags & NFS_MOUNT_VER3) {
				dprintk("%s: mount structure version %d does not support NFSv3\n",
						__FUNCTION__,
						data->version);
				goto out_err;
			}
			data->root.size = NFS2_FHSIZE;
			memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE);
		case 4:
			if (data->flags & NFS_MOUNT_SECFLAVOUR) {
				dprintk("%s: mount structure version %d does not support strong security\n",
						__FUNCTION__,
						data->version);
				goto out_err;
			}
		case 5:
			memset(data->context, 0, sizeof(data->context));
	}
#ifndef CONFIG_NFS_V3
	/* If NFSv3 is not compiled in, return -EPROTONOSUPPORT */
	s = ERR_PTR(-EPROTONOSUPPORT);
	if (data->flags & NFS_MOUNT_VER3) {
		dprintk("%s: NFSv3 not compiled into kernel\n", __FUNCTION__);
		goto out_err;
	}
#endif /* CONFIG_NFS_V3 */

	s = ERR_PTR(-ENOMEM);
	server = kmalloc(sizeof(struct nfs_server), GFP_KERNEL);
	if (!server)
		goto out_err;
	memset(server, 0, sizeof(struct nfs_server));
	/* Zero out the NFS state stuff */
	init_nfsv4_state(server);
	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);

	root = &server->fh;
	if (data->flags & NFS_MOUNT_VER3)
		root->size = data->root.size;
	else
		root->size = NFS2_FHSIZE;
	s = ERR_PTR(-EINVAL);
	if (root->size > sizeof(root->data)) {
		dprintk("%s: invalid root filehandle\n", __FUNCTION__);
		goto out_err;
	}
	memcpy(root->data, data->root.data, root->size);

	/* We now require that the mount process passes the remote address */
	memcpy(&server->addr, &data->addr, sizeof(server->addr));
	if (server->addr.sin_addr.s_addr == INADDR_ANY) {
		dprintk("%s: mount program didn't pass remote address!\n",
				__FUNCTION__);
		goto out_err;
	}

	/* Fire up rpciod if not yet running */
	s = ERR_PTR(rpciod_up());
	if (IS_ERR(s)) {
		dprintk("%s: couldn't start rpciod! Error = %ld\n",
				__FUNCTION__, PTR_ERR(s));
		goto out_err;
	}

	s = sget(fs_type, nfs_compare_super, nfs_set_super, server);
	if (IS_ERR(s) || s->s_root)
		goto out_rpciod_down;

	s->s_flags = flags;

	error = nfs_fill_super(s, data, flags & MS_VERBOSE ? 1 : 0);
	if (error) {
		up_write(&s->s_umount);
		deactivate_super(s);
		return ERR_PTR(error);
	}
	s->s_flags |= MS_ACTIVE;
	return s;
out_rpciod_down:
	rpciod_down();
out_err:
	kfree(server);
	return s;
}

static void nfs_kill_super(struct super_block *s)
{
	struct nfs_server *server = NFS_SB(s);

	kill_anon_super(s);

	if (!IS_ERR(server->client))
		rpc_shutdown_client(server->client);
	if (!IS_ERR(server->client_sys))
		rpc_shutdown_client(server->client_sys);
	if (!IS_ERR(server->client_acl))
		rpc_shutdown_client(server->client_acl);

	if (!(server->flags & NFS_MOUNT_NONLM))
		lockd_down();	/* release rpc.lockd */

	rpciod_down();		/* release rpciod */

	if (server->hostname != NULL)
		kfree(server->hostname);
	kfree(server);
}

static struct file_system_type nfs_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "nfs",
	.get_sb		= nfs_get_sb,
	.kill_sb	= nfs_kill_super,
	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
};

#ifdef CONFIG_NFS_V4

static void nfs4_clear_inode(struct inode *);


static struct super_operations nfs4_sops = { 
	.alloc_inode	= nfs_alloc_inode,
	.destroy_inode	= nfs_destroy_inode,
	.write_inode	= nfs_write_inode,
	.delete_inode	= nfs_delete_inode,
	.statfs		= nfs_statfs,
	.clear_inode	= nfs4_clear_inode,
	.umount_begin	= nfs_umount_begin,
	.show_options	= nfs_show_options,
};

/*
 * Clean out any remaining NFSv4 state that might be left over due
 * to open() calls that passed nfs_atomic_lookup, but failed to call
 * nfs_open().
 */
static void nfs4_clear_inode(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	/* If we are holding a delegation, return it! */
	nfs_inode_return_delegation(inode);
	/* First call standard NFS clear_inode() code */
	nfs_clear_inode(inode);
	/* Now clear out any remaining state */
	while (!list_empty(&nfsi->open_states)) {
		struct nfs4_state *state;
		
		state = list_entry(nfsi->open_states.next,
				struct nfs4_state,
				inode_states);
		dprintk("%s(%s/%Ld): found unclaimed NFSv4 state %p\n",
				__FUNCTION__,
				inode->i_sb->s_id,
				(long long)NFS_FILEID(inode),
				state);
		BUG_ON(atomic_read(&state->count) != 1);
		nfs4_close_state(state, state->state);
	}
}


static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, int silent)
{
	struct nfs_server *server;
	struct nfs4_client *clp = NULL;
	struct rpc_xprt *xprt = NULL;
	struct rpc_clnt *clnt = NULL;
	struct rpc_timeout timeparms;
	rpc_authflavor_t authflavour;
	int err = -EIO;

	sb->s_blocksize_bits = 0;
	sb->s_blocksize = 0;
	server = NFS_SB(sb);
	if (data->rsize != 0)
		server->rsize = nfs_block_size(data->rsize, NULL);
	if (data->wsize != 0)
		server->wsize = nfs_block_size(data->wsize, NULL);
	server->flags = data->flags & NFS_MOUNT_FLAGMASK;
	server->caps = NFS_CAP_ATOMIC_OPEN;

	server->acregmin = data->acregmin*HZ;
	server->acregmax = data->acregmax*HZ;
	server->acdirmin = data->acdirmin*HZ;
	server->acdirmax = data->acdirmax*HZ;

	server->rpc_ops = &nfs_v4_clientops;

	nfs_init_timeout_values(&timeparms, data->proto, data->timeo, data->retrans);

	clp = nfs4_get_client(&server->addr.sin_addr);
	if (!clp) {
		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
		return -EIO;
	}

	/* Now create transport and client */
	authflavour = RPC_AUTH_UNIX;
	if (data->auth_flavourlen != 0) {
		if (data->auth_flavourlen != 1) {
			dprintk("%s: Invalid number of RPC auth flavours %d.\n",
					__FUNCTION__, data->auth_flavourlen);
			err = -EINVAL;
			goto out_fail;
		}
		if (copy_from_user(&authflavour, data->auth_flavours, sizeof(authflavour))) {
			err = -EFAULT;
			goto out_fail;
		}
	}

	down_write(&clp->cl_sem);
	if (IS_ERR(clp->cl_rpcclient)) {
		xprt = xprt_create_proto(data->proto, &server->addr, &timeparms);
		if (IS_ERR(xprt)) {
			up_write(&clp->cl_sem);
			err = PTR_ERR(xprt);
			dprintk("%s: cannot create RPC transport. Error = %d\n",
					__FUNCTION__, err);
			goto out_fail;
		}
		clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
				server->rpc_ops->version, authflavour);
		if (IS_ERR(clnt)) {
			up_write(&clp->cl_sem);
			err = PTR_ERR(clnt);
			dprintk("%s: cannot create RPC client. Error = %d\n",
					__FUNCTION__, err);
			goto out_fail;
		}
		clnt->cl_intr     = 1;
		clnt->cl_softrtry = 1;
		clnt->cl_chatty   = 1;
		clp->cl_rpcclient = clnt;
		clp->cl_cred = rpcauth_lookupcred(clnt->cl_auth, 0);
		if (IS_ERR(clp->cl_cred)) {
			up_write(&clp->cl_sem);
			err = PTR_ERR(clp->cl_cred);
			clp->cl_cred = NULL;
			goto out_fail;
		}
		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
		nfs_idmap_new(clp);
	}
	if (list_empty(&clp->cl_superblocks)) {
		err = nfs4_init_client(clp);
		if (err != 0) {
			up_write(&clp->cl_sem);
			goto out_fail;
		}
	}
	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
	clnt = rpc_clone_client(clp->cl_rpcclient);
	if (!IS_ERR(clnt))
			server->nfs4_state = clp;
	up_write(&clp->cl_sem);
	clp = NULL;

	if (IS_ERR(clnt)) {
		err = PTR_ERR(clnt);
		dprintk("%s: cannot create RPC client. Error = %d\n",
				__FUNCTION__, err);
		return err;
	}

	server->client    = clnt;

	if (server->nfs4_state->cl_idmap == NULL) {
		dprintk("%s: failed to create idmapper.\n", __FUNCTION__);
		return -ENOMEM;
	}

	if (clnt->cl_auth->au_flavor != authflavour) {
		struct rpc_auth *auth;

		auth = rpcauth_create(authflavour, clnt);
		if (IS_ERR(auth)) {
			dprintk("%s: couldn't create credcache!\n", __FUNCTION__);
			return PTR_ERR(auth);
		}
	}

	sb->s_time_gran = 1;

	sb->s_op = &nfs4_sops;
	err = nfs_sb_init(sb, authflavour);
	if (err == 0)
		return 0;
out_fail:
	if (clp)
		nfs4_put_client(clp);
	return err;
}

static int nfs4_compare_super(struct super_block *sb, void *data)
{
	struct nfs_server *server = data;
	struct nfs_server *old = NFS_SB(sb);

	if (strcmp(server->hostname, old->hostname) != 0)
		return 0;
	if (strcmp(server->mnt_path, old->mnt_path) != 0)
		return 0;
	return 1;
}

static void *
nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen)
{
	void *p = NULL;

	if (!src->len)
		return ERR_PTR(-EINVAL);
	if (src->len < maxlen)
		maxlen = src->len;
	if (dst == NULL) {
		p = dst = kmalloc(maxlen + 1, GFP_KERNEL);
		if (p == NULL)
			return ERR_PTR(-ENOMEM);
	}
	if (copy_from_user(dst, src->data, maxlen)) {
		if (p != NULL)
			kfree(p);
		return ERR_PTR(-EFAULT);
	}
	dst[maxlen] = '\0';
	return dst;
}

static struct super_block *nfs4_get_sb(struct file_system_type *fs_type,
	int flags, const char *dev_name, void *raw_data)
{
	int error;
	struct nfs_server *server;
	struct super_block *s;
	struct nfs4_mount_data *data = raw_data;
	void *p;

	if (data == NULL) {
		dprintk("%s: missing data argument\n", __FUNCTION__);
		return ERR_PTR(-EINVAL);
	}
	if (data->version <= 0 || data->version > NFS4_MOUNT_VERSION) {
		dprintk("%s: bad mount version\n", __FUNCTION__);
		return ERR_PTR(-EINVAL);
	}

	server = kmalloc(sizeof(struct nfs_server), GFP_KERNEL);
	if (!server)
		return ERR_PTR(-ENOMEM);
	memset(server, 0, sizeof(struct nfs_server));
	/* Zero out the NFS state stuff */
	init_nfsv4_state(server);
	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);

	p = nfs_copy_user_string(NULL, &data->hostname, 256);
	if (IS_ERR(p))
		goto out_err;
	server->hostname = p;

	p = nfs_copy_user_string(NULL, &data->mnt_path, 1024);
	if (IS_ERR(p))
		goto out_err;
	server->mnt_path = p;

	p = nfs_copy_user_string(server->ip_addr, &data->client_addr,
			sizeof(server->ip_addr) - 1);
	if (IS_ERR(p))
		goto out_err;

	/* We now require that the mount process passes the remote address */
	if (data->host_addrlen != sizeof(server->addr)) {
		s = ERR_PTR(-EINVAL);
		goto out_free;
	}
	if (copy_from_user(&server->addr, data->host_addr, sizeof(server->addr))) {
		s = ERR_PTR(-EFAULT);
		goto out_free;
	}
	if (server->addr.sin_family != AF_INET ||
	    server->addr.sin_addr.s_addr == INADDR_ANY) {
		dprintk("%s: mount program didn't pass remote IP address!\n",
				__FUNCTION__);
		s = ERR_PTR(-EINVAL);
		goto out_free;
	}

	/* Fire up rpciod if not yet running */
	s = ERR_PTR(rpciod_up());
	if (IS_ERR(s)) {
		dprintk("%s: couldn't start rpciod! Error = %ld\n",
				__FUNCTION__, PTR_ERR(s));
		goto out_free;
	}

	s = sget(fs_type, nfs4_compare_super, nfs_set_super, server);

	if (IS_ERR(s) || s->s_root)
		goto out_free;

	s->s_flags = flags;

	error = nfs4_fill_super(s, data, flags & MS_VERBOSE ? 1 : 0);
	if (error) {
		up_write(&s->s_umount);
		deactivate_super(s);
		return ERR_PTR(error);
	}
	s->s_flags |= MS_ACTIVE;
	return s;
out_err:
	s = (struct super_block *)p;
out_free:
	if (server->mnt_path)
		kfree(server->mnt_path);
	if (server->hostname)
		kfree(server->hostname);
	kfree(server);
	return s;
}

static void nfs4_kill_super(struct super_block *sb)
{
	struct nfs_server *server = NFS_SB(sb);

	nfs_return_all_delegations(sb);
	kill_anon_super(sb);

	nfs4_renewd_prepare_shutdown(server);

	if (server->client != NULL && !IS_ERR(server->client))
		rpc_shutdown_client(server->client);
	rpciod_down();		/* release rpciod */

	destroy_nfsv4_state(server);