kernel/fs/rootfs.cpp

/*
 * Copyright 2002-2009, Axel Dörfler, axeld@pinc-software.de.
 * Distributed under the terms of the MIT License.
 *
 * Copyright 2001-2002, Travis Geiselbrecht. All rights reserved.
 * Distributed under the terms of the NewOS License.
 */


#if FS_SHELL
#	include "fssh_api_wrapper.h"

#	include "hash.h"
#	include "list.h"
#else
#	include <stdio.h>
#	include <stdlib.h>
#	include <string.h>
#	include <sys/stat.h>

#	include <fs_cache.h>
#	include <KernelExport.h>
#	include <NodeMonitor.h>

#	include <debug.h>
#	include <khash.h>
#	include <lock.h>
#	include <util/AutoLock.h>
#	include <vfs.h>
#	include <vm/vm.h>
#endif


#if FS_SHELL
	using namespace FSShell;
#	define user_strlcpy(to, from, len)	(strlcpy(to, from, len), FSSH_B_OK)
#endif


//#define TRACE_ROOTFS
#ifdef TRACE_ROOTFS
#	define TRACE(x) dprintf x
#else
#	define TRACE(x)
#endif


struct rootfs_stream {
	mode_t						type;
	struct stream_dir {
		struct rootfs_vnode*	dir_head;
		struct list				cookies;
		mutex					cookie_lock;
	} dir;
	struct stream_symlink {
		char*					path;
		size_t					length;
	} symlink;
};

struct rootfs_vnode {
	struct rootfs_vnode*		all_next;
	ino_t						id;
	char*						name;
	timespec					modification_time;
	timespec					creation_time;
	uid_t						uid;
	gid_t						gid;
	struct rootfs_vnode*		parent;
	struct rootfs_vnode*		dir_next;
	struct rootfs_stream		stream;
};

struct rootfs {
	fs_volume*					volume;
	dev_t						id;
	rw_lock						lock;
	ino_t						next_vnode_id;
	hash_table*					vnode_list_hash;
	struct rootfs_vnode*		root_vnode;
};

// dircookie, dirs are only types of streams supported by rootfs
struct rootfs_dir_cookie {
	struct list_link			link;
	mutex						lock;
	struct rootfs_vnode*		current;
	int32						iteration_state;
};

// directory iteration states
enum {
	ITERATION_STATE_DOT		= 0,
	ITERATION_STATE_DOT_DOT	= 1,
	ITERATION_STATE_OTHERS	= 2,
	ITERATION_STATE_BEGIN	= ITERATION_STATE_DOT,
};

// extern and in a private namespace only to make forward declaration possible
namespace {
	extern fs_volume_ops sVolumeOps;
	extern fs_vnode_ops sVnodeOps;
}

#define ROOTFS_HASH_SIZE 16


static timespec
current_timespec()
{
	bigtime_t time = real_time_clock_usecs();

	timespec tv;
	tv.tv_sec = time / 1000000;
	tv.tv_nsec = (time % 1000000) * 1000;
	return tv;
}


static uint32
rootfs_vnode_hash_func(void* _v, const void* _key, uint32 range)
{
	struct rootfs_vnode* vnode = (rootfs_vnode*)_v;
	const ino_t* key = (const ino_t*)_key;

	if (vnode != NULL)
		return vnode->id % range;

	return (uint64)*key % range;
}


static int
rootfs_vnode_compare_func(void* _v, const void* _key)
{
	struct rootfs_vnode* v = (rootfs_vnode*)_v;
	const ino_t* key = (const ino_t*)_key;

	if (v->id == *key)
		return 0;

	return -1;
}


static struct rootfs_vnode*
rootfs_create_vnode(struct rootfs* fs, struct rootfs_vnode* parent,
	const char* name, int type)
{
	struct rootfs_vnode* vnode;

	vnode = (rootfs_vnode*)malloc(sizeof(struct rootfs_vnode));
	if (vnode == NULL)
		return NULL;

	memset(vnode, 0, sizeof(struct rootfs_vnode));

	if (name != NULL) {
		vnode->name = strdup(name);
		if (vnode->name == NULL) {
			free(vnode);
			return NULL;
		}
	}

	vnode->id = fs->next_vnode_id++;
	vnode->stream.type = type;
	vnode->creation_time = vnode->modification_time = current_timespec();
	vnode->uid = geteuid();
	vnode->gid = parent ? parent->gid : getegid();
		// inherit group from parent if possible

	if (S_ISDIR(type)) {
		list_init(&vnode->stream.dir.cookies);
		mutex_init(&vnode->stream.dir.cookie_lock, "rootfs dir cookies");
	}

	return vnode;
}


static status_t
rootfs_delete_vnode(struct rootfs* fs, struct rootfs_vnode* v, bool force_delete)
{
	// cant delete it if it's in a directory or is a directory
	// and has children
	if (!force_delete && (v->stream.dir.dir_head != NULL || v->dir_next != NULL))
		return EPERM;

	// remove it from the global hash table
	hash_remove(fs->vnode_list_hash, v);

	if (S_ISDIR(v->stream.type))
		mutex_destroy(&v->stream.dir.cookie_lock);

	free(v->name);
	free(v);

	return 0;
}


/*! Makes sure none of the dircookies point to the vnode passed in. */
static void
update_dir_cookies(struct rootfs_vnode* dir, struct rootfs_vnode* vnode)
{
	struct rootfs_dir_cookie* cookie = NULL;

	while ((cookie = (rootfs_dir_cookie*)list_get_next_item(
			&dir->stream.dir.cookies, cookie)) != NULL) {
		MutexLocker cookieLocker(cookie->lock);
		if (cookie->current == vnode)
			cookie->current = vnode->dir_next;
	}
}


static struct rootfs_vnode*
rootfs_find_in_dir(struct rootfs_vnode* dir, const char* path)
{
	struct rootfs_vnode* vnode;

	if (!strcmp(path, "."))
		return dir;
	if (!strcmp(path, ".."))
		return dir->parent;

	for (vnode = dir->stream.dir.dir_head; vnode; vnode = vnode->dir_next) {
		if (!strcmp(vnode->name, path))
			return vnode;
	}
	return NULL;
}


static status_t
rootfs_insert_in_dir(struct rootfs* fs, struct rootfs_vnode* dir,
	struct rootfs_vnode* vnode)
{
	// make sure the directory stays sorted alphabetically

	struct rootfs_vnode* node = dir->stream.dir.dir_head;
	struct rootfs_vnode* last = NULL;
	while (node != NULL && strcmp(node->name, vnode->name) < 0) {
		last = node;
		node = node->dir_next;
	}
	if (last == NULL) {
		// the new vnode is the first entry in the list
		vnode->dir_next = dir->stream.dir.dir_head;
		dir->stream.dir.dir_head = vnode;
	} else {
		// insert after that node
		vnode->dir_next = last->dir_next;
		last->dir_next = vnode;
	}

	vnode->parent = dir;
	dir->modification_time = current_timespec();

	notify_stat_changed(fs->id, dir->id, B_STAT_MODIFICATION_TIME);
	return B_OK;
}


static status_t
rootfs_remove_from_dir(struct rootfs* fs, struct rootfs_vnode* dir,
	struct rootfs_vnode* removeVnode)
{
	struct rootfs_vnode* vnode;
	struct rootfs_vnode* lastVnode;

	for (vnode = dir->stream.dir.dir_head, lastVnode = NULL; vnode != NULL;
			lastVnode = vnode, vnode = vnode->dir_next) {
		if (vnode == removeVnode) {
			// make sure all dircookies dont point to this vnode
			update_dir_cookies(dir, vnode);

			if (lastVnode)
				lastVnode->dir_next = vnode->dir_next;
			else
				dir->stream.dir.dir_head = vnode->dir_next;
			vnode->dir_next = NULL;

			dir->modification_time = current_timespec();
			notify_stat_changed(fs->id, dir->id, B_STAT_MODIFICATION_TIME);
			return B_OK;
		}
	}
	return B_ENTRY_NOT_FOUND;
}


static bool
rootfs_is_dir_empty(struct rootfs_vnode* dir)
{
	return !dir->stream.dir.dir_head;
}


/*! You must hold the FS write lock when calling this function */
static status_t
remove_node(struct rootfs* fs, struct rootfs_vnode* directory,
	struct rootfs_vnode* vnode)
{
	// schedule this vnode to be removed when it's ref goes to zero

	bool gotNode = (get_vnode(fs->volume, vnode->id, NULL) == B_OK);

	status_t status = B_OK;
	if (gotNode)
		status = remove_vnode(fs->volume, vnode->id);

	if (status == B_OK) {
		rootfs_remove_from_dir(fs, directory, vnode);
		notify_entry_removed(fs->id, directory->id, vnode->name, vnode->id);
	}

	if (gotNode)
		put_vnode(fs->volume, vnode->id);

	return status;
}


static status_t
rootfs_remove(struct rootfs* fs, struct rootfs_vnode* dir, const char* name,
	bool isDirectory)
{
	struct rootfs_vnode* vnode;
	status_t status = B_OK;

	WriteLocker locker(fs->lock);

	vnode = rootfs_find_in_dir(dir, name);
	if (!vnode)
		status = B_ENTRY_NOT_FOUND;
	else if (isDirectory && !S_ISDIR(vnode->stream.type))
		status = B_NOT_A_DIRECTORY;
	else if (!isDirectory && S_ISDIR(vnode->stream.type))
		status = B_IS_A_DIRECTORY;
	else if (isDirectory && !rootfs_is_dir_empty(vnode))
		status = B_DIRECTORY_NOT_EMPTY;

	if (status != B_OK)
		return status;

	entry_cache_remove(fs->volume->id, dir->id, name);

	return remove_node(fs, dir, vnode);
}


//	#pragma mark -


static status_t
rootfs_mount(fs_volume* volume, const char* device, uint32 flags,
	const char* args, ino_t* _rootID)
{
	struct rootfs* fs;
	struct rootfs_vnode* vnode;
	status_t err;

	TRACE(("rootfs_mount: entry\n"));

	fs = (rootfs*)malloc(sizeof(struct rootfs));
	if (fs == NULL)
		return B_NO_MEMORY;

	volume->private_volume = fs;
	volume->ops = &sVolumeOps;
	fs->volume = volume;
	fs->id = volume->id;
	fs->next_vnode_id = 1;

	rw_lock_init(&fs->lock, "rootfs");

	fs->vnode_list_hash = hash_init(ROOTFS_HASH_SIZE,
		offsetof(rootfs_vnode, all_next), &rootfs_vnode_compare_func,
		&rootfs_vnode_hash_func);
	if (fs->vnode_list_hash == NULL) {
		err = B_NO_MEMORY;
		goto err2;
	}

	// create the root vnode
	vnode = rootfs_create_vnode(fs, NULL, ".", S_IFDIR | 0777);
	if (vnode == NULL) {
		err = B_NO_MEMORY;
		goto err3;
	}
	vnode->parent = vnode;

	fs->root_vnode = vnode;
	hash_insert(fs->vnode_list_hash, vnode);
	publish_vnode(volume, vnode->id, vnode, &sVnodeOps, vnode->stream.type, 0);

	*_rootID = vnode->id;

	return B_OK;

err3:
	hash_uninit(fs->vnode_list_hash);
err2:
	rw_lock_destroy(&fs->lock);
	free(fs);

	return err;
}


static status_t
rootfs_unmount(fs_volume* _volume)
{
	struct rootfs* fs = (struct rootfs*)_volume->private_volume;

	TRACE(("rootfs_unmount: entry fs = %p\n", fs));

	// release the reference to the root
	put_vnode(fs->volume, fs->root_vnode->id);

	// delete all of the vnodes
	struct hash_iterator i;
	hash_open(fs->vnode_list_hash, &i);

	while (struct rootfs_vnode* vnode = (struct rootfs_vnode*)
			hash_next(fs->vnode_list_hash, &i)) {
		rootfs_delete_vnode(fs, vnode, true);
	}

	hash_close(fs->vnode_list_hash, &i, false);

	hash_uninit(fs->vnode_list_hash);
	rw_lock_destroy(&fs->lock);
	free(fs);

	return B_OK;
}


static status_t
rootfs_sync(fs_volume* _volume)
{
	TRACE(("rootfs_sync: entry\n"));

	return B_OK;
}


static status_t
rootfs_lookup(fs_volume* _volume, fs_vnode* _dir, const char* name, ino_t* _id)
{
	struct rootfs* fs = (struct rootfs*)_volume->private_volume;
	struct rootfs_vnode* dir = (struct rootfs_vnode*)_dir->private_node;
	struct rootfs_vnode* vnode;

	TRACE(("rootfs_lookup: entry dir %p, name '%s'\n", dir, name));
	if (!S_ISDIR(dir->stream.type))
		return B_NOT_A_DIRECTORY;

	ReadLocker locker(fs->lock);

	// look it up
	vnode = rootfs_find_in_dir(dir, name);
	if (!vnode)
		return B_ENTRY_NOT_FOUND;

	status_t status = get_vnode(fs->volume, vnode->id, NULL);
	if (status != B_OK)
		return status;

	entry_cache_add(fs->volume->id, dir->id, name, vnode->id);

	*_id = vnode->id;
	return B_OK;
}


static status_t
rootfs_get_vnode_name(fs_volume* _volume, fs_vnode* _vnode, char* buffer,
	size_t bufferSize)
{
	struct rootfs_vnode* vnode = (struct rootfs_vnode*)_vnode->private_node;

	TRACE(("rootfs_get_vnode_name: vnode = %p (name = %s)\n", vnode,
		vnode->name));

	strlcpy(buffer, vnode->name, bufferSize);
	return B_OK;
}


static status_t
rootfs_get_vnode(fs_volume* _volume, ino_t id, fs_vnode* _vnode, int* _type,
	uint32* _flags, bool reenter)
{
	struct rootfs* fs = (struct rootfs*)_volume->private_volume;
	struct rootfs_vnode* vnode;

	TRACE(("rootfs_getvnode: asking for vnode %Ld, r %d\n", id, reenter));

	if (!reenter)
		rw_lock_read_lock(&fs->lock);

	vnode = (rootfs_vnode*)hash_lookup(fs->vnode_list_hash, &id);

	if (!reenter)
		rw_lock_read_unlock(&fs->lock);

	TRACE(("rootfs_getnvnode: looked it up at %p\n", vnode));

	if (vnode == NULL)
		return B_ENTRY_NOT_FOUND;

	_vnode->private_node = vnode;
	_vnode->ops = &sVnodeOps;
	*_type = vnode->stream.type;
	*_flags = 0;

	return B_OK;
}


static status_t
rootfs_put_vnode(fs_volume* _volume, fs_vnode* _vnode, bool reenter)
{
#ifdef TRACE_ROOTFS
	struct rootfs_vnode* vnode = (struct rootfs_vnode*)_vnode->private_node;

	TRACE(("rootfs_putvnode: entry on vnode 0x%Lx, r %d\n", vnode->id, reenter));
#endif
	return B_OK; // whatever
}


static status_t
rootfs_remove_vnode(fs_volume* _volume, fs_vnode* _vnode, bool reenter)
{
	struct rootfs* fs = (struct rootfs*)_volume->private_volume;
	struct rootfs_vnode* vnode = (struct rootfs_vnode*)_vnode->private_node;

	TRACE(("rootfs_remove_vnode: remove %p (0x%Lx), r %d\n", vnode, vnode->id,
		reenter));

	if (!reenter)
		rw_lock_write_lock(&fs->lock);

	if (vnode->dir_next) {
		// can't remove node if it's linked to the dir
		panic("rootfs_remove_vnode: vnode %p asked to be removed is present in "
			"dir\n", vnode);
	}

	rootfs_delete_vnode(fs, vnode, false);

	if (!reenter)
		rw_lock_write_unlock(&fs->lock);

	return B_OK;
}


static status_t
rootfs_create(fs_volume* _volume, fs_vnode* _dir, const char* name, int omode,
	int perms, void** _cookie, ino_t* _newID)
{
	return B_BAD_VALUE;
}


static status_t
rootfs_open(fs_volume* _volume, fs_vnode* _v, int oflags, void** _cookie)
{
	// allow to open the file, but it can't be done anything with it

	*_cookie = NULL;
	return B_OK;
}


static status_t
rootfs_close(fs_volume* _volume, fs_vnode* _vnode, void* _cookie)
{
	TRACE(("rootfs_close: entry vnode %p, cookie %p\n", _vnode->private_node,
		_cookie));
	return B_OK;
}


static status_t
rootfs_free_cookie(fs_volume* _volume, fs_vnode* _v, void* _cookie)
{
	return B_OK;
}


static status_t
rootfs_fsync(fs_volume* _volume, fs_vnode* _v)
{
	return B_OK;
}


static status_t
rootfs_read(fs_volume* _volume, fs_vnode* _vnode, void* _cookie,
	off_t pos, void* buffer, size_t* _length)
{
	return EINVAL;
}


static status_t
rootfs_write(fs_volume* _volume, fs_vnode* vnode, void* cookie,
	off_t pos, const void* buffer, size_t* _length)
{
	TRACE(("rootfs_write: vnode %p, cookie %p, pos 0x%Lx , len %#x\n",
		vnode, cookie, pos, (int)*_length));

	return EPERM;
}


static status_t
rootfs_create_dir(fs_volume* _volume, fs_vnode* _dir, const char* name,
	int mode)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* dir = (rootfs_vnode*)_dir->private_node;
	struct rootfs_vnode* vnode;

	TRACE(("rootfs_create_dir: dir %p, name = '%s', perms = %d\n", dir, name,
		mode));

	WriteLocker locker(fs->lock);

	vnode = rootfs_find_in_dir(dir, name);
	if (vnode != NULL)
		return B_FILE_EXISTS;

	TRACE(("rootfs_create: creating new vnode\n"));
	vnode = rootfs_create_vnode(fs, dir, name, S_IFDIR | (mode & S_IUMSK));
	if (vnode == NULL)
		return B_NO_MEMORY;

	rootfs_insert_in_dir(fs, dir, vnode);
	hash_insert(fs->vnode_list_hash, vnode);

	entry_cache_add(fs->volume->id, dir->id, name, vnode->id);
	notify_entry_created(fs->id, dir->id, name, vnode->id);

	return B_OK;
}


static status_t
rootfs_remove_dir(fs_volume* _volume, fs_vnode* _dir, const char* name)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* dir = (rootfs_vnode*)_dir->private_node;

	TRACE(("rootfs_remove_dir: dir %p (0x%Lx), name '%s'\n", dir, dir->id,
		name));

	return rootfs_remove(fs, dir, name, true);
}


static status_t
rootfs_open_dir(fs_volume* _volume, fs_vnode* _v, void** _cookie)
{
	struct rootfs* fs = (struct rootfs*)_volume->private_volume;
	struct rootfs_vnode* vnode = (struct rootfs_vnode*)_v->private_node;
	struct rootfs_dir_cookie* cookie;

	TRACE(("rootfs_open: vnode %p\n", vnode));

	if (!S_ISDIR(vnode->stream.type))
		return B_BAD_VALUE;

	cookie = (rootfs_dir_cookie*)malloc(sizeof(struct rootfs_dir_cookie));
	if (cookie == NULL)
		return B_NO_MEMORY;

	mutex_init(&cookie->lock, "rootfs dir cookie");

	ReadLocker locker(fs->lock);

	cookie->current = vnode->stream.dir.dir_head;
	cookie->iteration_state = ITERATION_STATE_BEGIN;

	mutex_lock(&vnode->stream.dir.cookie_lock);
	list_add_item(&vnode->stream.dir.cookies, cookie);
	mutex_unlock(&vnode->stream.dir.cookie_lock);

	*_cookie = cookie;

	return B_OK;
}


static status_t
rootfs_free_dir_cookie(fs_volume* _volume, fs_vnode* _vnode, void* _cookie)
{
	struct rootfs_dir_cookie* cookie = (rootfs_dir_cookie*)_cookie;
	struct rootfs_vnode* vnode = (rootfs_vnode*)_vnode->private_node;
	struct rootfs* fs = (rootfs*)_volume->private_volume;

	ReadLocker locker(fs->lock);

	mutex_lock(&vnode->stream.dir.cookie_lock);
	list_remove_item(&vnode->stream.dir.cookies, cookie);
	mutex_unlock(&vnode->stream.dir.cookie_lock);

	locker.Unlock();

	mutex_destroy(&cookie->lock);

	free(cookie);
	return B_OK;
}


static status_t
rootfs_read_dir(fs_volume* _volume, fs_vnode* _vnode, void* _cookie,
	struct dirent* dirent, size_t bufferSize, uint32* _num)
{
	struct rootfs_vnode* vnode = (struct rootfs_vnode*)_vnode->private_node;
	struct rootfs_dir_cookie* cookie = (rootfs_dir_cookie*)_cookie;
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* childNode = NULL;
	const char* name = NULL;
	struct rootfs_vnode* nextChildNode = NULL;

	TRACE(("rootfs_read_dir: vnode %p, cookie %p, buffer = %p, bufferSize = %d, "
		"num = %p\n", _vnode, cookie, dirent, (int)bufferSize, _num));

	ReadLocker locker(fs->lock);

	MutexLocker cookieLocker(cookie->lock);
	int nextState = cookie->iteration_state;

	switch (cookie->iteration_state) {
		case ITERATION_STATE_DOT:
			childNode = vnode;
			name = ".";
			nextChildNode = vnode->stream.dir.dir_head;
			nextState = cookie->iteration_state + 1;
			break;
		case ITERATION_STATE_DOT_DOT:
			childNode = vnode->parent;
			name = "..";
			nextChildNode = vnode->stream.dir.dir_head;
			nextState = cookie->iteration_state + 1;
			break;
		default:
			childNode = cookie->current;
			if (childNode) {
				name = childNode->name;
				nextChildNode = childNode->dir_next;
			}
			break;
	}

	if (!childNode) {
		// we're at the end of the directory
		*_num = 0;
		return B_OK;
	}

	dirent->d_dev = fs->id;
	dirent->d_ino = childNode->id;
	dirent->d_reclen = strlen(name) + sizeof(struct dirent);

	if (dirent->d_reclen > bufferSize)
		return ENOBUFS;

	int nameLength = user_strlcpy(dirent->d_name, name,
		bufferSize - sizeof(struct dirent));
	if (nameLength < B_OK)
		return nameLength;

	cookie->current = nextChildNode;
	cookie->iteration_state = nextState;
	*_num = 1;
	return B_OK;
}


static status_t
rootfs_rewind_dir(fs_volume* _volume, fs_vnode* _vnode, void* _cookie)
{
	struct rootfs_dir_cookie* cookie = (rootfs_dir_cookie*)_cookie;
	struct rootfs_vnode* vnode = (rootfs_vnode*)_vnode->private_node;
	struct rootfs* fs = (rootfs*)_volume->private_volume;

	ReadLocker locker(fs->lock);
	MutexLocker cookieLocker(cookie->lock);

	cookie->current = vnode->stream.dir.dir_head;
	cookie->iteration_state = ITERATION_STATE_BEGIN;

	return B_OK;
}


static status_t
rootfs_ioctl(fs_volume* _volume, fs_vnode* _v, void* _cookie, uint32 op,
	void* buffer, size_t length)
{
	TRACE(("rootfs_ioctl: vnode %p, cookie %p, op %d, buf %p, length %d\n",
		_volume, _cookie, (int)op, buffer, (int)length));

	return B_BAD_VALUE;
}


static bool
rootfs_can_page(fs_volume* _volume, fs_vnode* _v, void* cookie)
{
	return false;
}


static status_t
rootfs_read_pages(fs_volume* _volume, fs_vnode* _v, void* cookie, off_t pos,
	const iovec* vecs, size_t count, size_t* _numBytes)
{
	return B_NOT_ALLOWED;
}


static status_t
rootfs_write_pages(fs_volume* _volume, fs_vnode* _v, void* cookie, off_t pos,
	const iovec* vecs, size_t count, size_t* _numBytes)
{
	return B_NOT_ALLOWED;
}


static status_t
rootfs_read_link(fs_volume* _volume, fs_vnode* _link, char* buffer,
	size_t* _bufferSize)
{
	struct rootfs_vnode* link = (rootfs_vnode*)_link->private_node;

	if (!S_ISLNK(link->stream.type))
		return B_BAD_VALUE;

	if (link->stream.symlink.length < *_bufferSize)
		*_bufferSize = link->stream.symlink.length;

	memcpy(buffer, link->stream.symlink.path, *_bufferSize);
	return B_OK;
}


static status_t
rootfs_symlink(fs_volume* _volume, fs_vnode* _dir, const char* name,
	const char* path, int mode)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* dir = (rootfs_vnode*)_dir->private_node;
	struct rootfs_vnode* vnode;

	TRACE(("rootfs_symlink: dir %p, name = '%s', path = %s\n", dir, name, path));

	WriteLocker locker(fs->lock);

	vnode = rootfs_find_in_dir(dir, name);
	if (vnode != NULL)
		return B_FILE_EXISTS;

	TRACE(("rootfs_create: creating new symlink\n"));
	vnode = rootfs_create_vnode(fs, dir, name, S_IFLNK | (mode & S_IUMSK));
	if (vnode == NULL)
		return B_NO_MEMORY;

	rootfs_insert_in_dir(fs, dir, vnode);
	hash_insert(fs->vnode_list_hash, vnode);

	vnode->stream.symlink.path = strdup(path);
	if (vnode->stream.symlink.path == NULL) {
		rootfs_delete_vnode(fs, vnode, false);
		return B_NO_MEMORY;
	}
	vnode->stream.symlink.length = strlen(path);

	entry_cache_add(fs->volume->id, dir->id, name, vnode->id);

	notify_entry_created(fs->id, dir->id, name, vnode->id);

	return B_OK;
}


static status_t
rootfs_unlink(fs_volume* _volume, fs_vnode* _dir, const char* name)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* dir = (rootfs_vnode*)_dir->private_node;

	TRACE(("rootfs_unlink: dir %p (0x%Lx), name '%s'\n", dir, dir->id, name));

	return rootfs_remove(fs, dir, name, false);
}


static status_t
rootfs_rename(fs_volume* _volume, fs_vnode* _fromDir, const char* fromName,
	fs_vnode* _toDir, const char* toName)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* fromDirectory = (rootfs_vnode*)_fromDir->private_node;
	struct rootfs_vnode* toDirectory = (rootfs_vnode*)_toDir->private_node;

	TRACE(("rootfs_rename: from %p (0x%Lx, %s), fromName '%s', to %p "
		"(0x%Lx, %s), toName '%s'\n", fromDirectory, fromDirectory->id,
		fromDirectory->name != NULL ? fromDirectory->name : "NULL",
		fromName, toDirectory, toDirectory->id,
		toDirectory->name != NULL ? toDirectory->name : "NULL",
		toName));

	// Prevent renaming /boot, since that will stop everything from working.
	// TODO: This should be solved differently. Either root should still be
	// able to do this or a mechanism should be introduced that does this
	// at the VFS level, for example by checking nodes for a specific
	// attribute.
	if (fromDirectory->id == 1 && strcmp(fromName, "boot") == 0)
		return EPERM;

	WriteLocker locker(fs->lock);

	struct rootfs_vnode* vnode = rootfs_find_in_dir(fromDirectory, fromName);
	if (vnode == NULL)
		return B_ENTRY_NOT_FOUND;

	// make sure the target is not a subdirectory of us
	struct rootfs_vnode* parent = toDirectory->parent;
	while (parent != NULL && parent != parent->parent) {
		if (parent == vnode)
			return B_BAD_VALUE;

		parent = parent->parent;
	}

	struct rootfs_vnode* targetVnode = rootfs_find_in_dir(toDirectory, toName);
	if (targetVnode != NULL) {
		// target node exists, let's see if it is an empty directory
		if (S_ISDIR(targetVnode->stream.type)
			&& !rootfs_is_dir_empty(targetVnode))
			return B_NAME_IN_USE;

		// so we can cleanly remove it
		entry_cache_remove(fs->volume->id, toDirectory->id, toName);
		remove_node(fs, toDirectory, targetVnode);
	}

	// we try to reuse the existing name buffer if possible
	if (strlen(fromName) >= strlen(toName)) {
		char* nameBuffer = strdup(toName);
		if (nameBuffer == NULL)
			return B_NO_MEMORY;

		free(vnode->name);
		vnode->name = nameBuffer;
	} else {
		// we can just copy it
		strcpy(vnode->name, toName);
	}

	// remove it from the dir
	entry_cache_remove(fs->volume->id, fromDirectory->id, fromName);
	rootfs_remove_from_dir(fs, fromDirectory, vnode);

	// Add it back to the dir with the new name.
	// We need to do this even in the same directory,
	// so that it keeps sorted correctly.
	rootfs_insert_in_dir(fs, toDirectory, vnode);

	entry_cache_add(fs->volume->id, toDirectory->id, toName, vnode->id);

	notify_entry_moved(fs->id, fromDirectory->id, fromName, toDirectory->id,
		toName, vnode->id);

	return B_OK;
}


static status_t
rootfs_read_stat(fs_volume* _volume, fs_vnode* _v, struct stat* stat)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* vnode = (rootfs_vnode*)_v->private_node;

	TRACE(("rootfs_read_stat: vnode %p (0x%Lx), stat %p\n", vnode, vnode->id,
		stat));

	// stream exists, but we know to return size 0, since we can only hold
	// directories
	stat->st_dev = fs->id;
	stat->st_ino = vnode->id;
	if (S_ISLNK(vnode->stream.type))
		stat->st_size = vnode->stream.symlink.length;
	else
		stat->st_size = 0;
	stat->st_mode = vnode->stream.type;

	stat->st_nlink = 1;
	stat->st_blksize = 65536;
	stat->st_blocks = 0;

	stat->st_uid = vnode->uid;
	stat->st_gid = vnode->gid;

	stat->st_atim.tv_sec = real_time_clock();
	stat->st_atim.tv_nsec = 0;
	stat->st_mtim = stat->st_ctim = vnode->modification_time;
	stat->st_crtim = vnode->creation_time;

	return B_OK;
}


static status_t
rootfs_write_stat(fs_volume* _volume, fs_vnode* _vnode, const struct stat* stat,
	uint32 statMask)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* vnode = (rootfs_vnode*)_vnode->private_node;

	TRACE(("rootfs_write_stat: vnode %p (0x%Lx), stat %p\n", vnode, vnode->id,
		stat));

	// we cannot change the size of anything
	if (statMask & B_STAT_SIZE)
		return B_BAD_VALUE;

	WriteLocker locker(fs->lock);

	if ((statMask & B_STAT_MODE) != 0) {
		vnode->stream.type = (vnode->stream.type & ~S_IUMSK)
			| (stat->st_mode & S_IUMSK);
	}

	if ((statMask & B_STAT_UID) != 0)
		vnode->uid = stat->st_uid;
	if ((statMask & B_STAT_GID) != 0)
		vnode->gid = stat->st_gid;

	if ((statMask & B_STAT_MODIFICATION_TIME) != 0)
		vnode->modification_time = stat->st_mtim;
	if ((statMask & B_STAT_CREATION_TIME) != 0)
		vnode->creation_time = stat->st_crtim;

	locker.Unlock();

	notify_stat_changed(fs->id, vnode->id, statMask);
	return B_OK;
}


static status_t
rootfs_create_special_node(fs_volume* _volume, fs_vnode* _dir, const char* name,
	fs_vnode* subVnode, mode_t mode, uint32 flags, fs_vnode* _superVnode,
	ino_t* _nodeID)
{
	struct rootfs* fs = (rootfs*)_volume->private_volume;
	struct rootfs_vnode* dir = (rootfs_vnode*)_dir->private_node;
	struct rootfs_vnode* vnode;

	WriteLocker locker(fs->lock);

	if (name != NULL) {
		vnode = rootfs_find_in_dir(dir, name);
		if (vnode != NULL)
			return B_FILE_EXISTS;
	}

	vnode = rootfs_create_vnode(fs, dir, name, mode);
	if (vnode == NULL)
		return B_NO_MEMORY;

	if (name != NULL)
		rootfs_insert_in_dir(fs, dir, vnode);
	else
		flags |= B_VNODE_PUBLISH_REMOVED;

	hash_insert(fs->vnode_list_hash, vnode);

	_superVnode->private_node = vnode;
	_superVnode->ops = &sVnodeOps;
	*_nodeID = vnode->id;

	if (subVnode == NULL)
		subVnode = _superVnode;

	status_t status = publish_vnode(fs->volume, vnode->id,
		subVnode->private_node, subVnode->ops, mode, flags);
	if (status != B_OK) {
		if (name != NULL)
			rootfs_remove_from_dir(fs, dir, vnode);
		rootfs_delete_vnode(fs, vnode, false);
		return status;
	}

	if (name != NULL) {
		entry_cache_add(fs->volume->id, dir->id, name, vnode->id);
		notify_entry_created(fs->id, dir->id, name, vnode->id);
	}

	return B_OK;
}


static status_t
rootfs_std_ops(int32 op, ...)
{
	switch (op) {
		case B_MODULE_INIT:
			return B_OK;

		case B_MODULE_UNINIT:
			return B_OK;

		default:
			return B_ERROR;
	}
}


namespace {

fs_volume_ops sVolumeOps = {
	&rootfs_unmount,
	NULL,
	NULL,
	&rootfs_sync,
	&rootfs_get_vnode,

	// the other operations are not supported (indices, queries)
	NULL,
};

fs_vnode_ops sVnodeOps = {
	&rootfs_lookup,
	&rootfs_get_vnode_name,

	&rootfs_put_vnode,
	&rootfs_remove_vnode,

	&rootfs_can_page,
	&rootfs_read_pages,
	&rootfs_write_pages,

	NULL,	// io()
	NULL,	// cancel_io()

	NULL,	// get_file_map()

	/* common */
	&rootfs_ioctl,
	NULL,	// fs_set_flags()
	NULL,	// select
	NULL,	// deselect
	&rootfs_fsync,

	&rootfs_read_link,
	&rootfs_symlink,
	NULL,	// fs_link()
	&rootfs_unlink,
	&rootfs_rename,

	NULL,	// fs_access()
	&rootfs_read_stat,
	&rootfs_write_stat,
	NULL,

	/* file */
	&rootfs_create,
	&rootfs_open,
	&rootfs_close,
	&rootfs_free_cookie,
	&rootfs_read,
	&rootfs_write,

	/* directory */
	&rootfs_create_dir,
	&rootfs_remove_dir,
	&rootfs_open_dir,
	&rootfs_close,			// same as for files - it does nothing, anyway
	&rootfs_free_dir_cookie,
	&rootfs_read_dir,
	&rootfs_rewind_dir,

	/* attribute directory operations */
	NULL,	// open_attr_dir
	NULL,	// close_attr_dir
	NULL,	// free_attr_dir_cookie
	NULL,	// read_attr_dir
	NULL,	// rewind_attr_dir

	/* attribute operations */
	NULL,	// create_attr
	NULL,	// open_attr
	NULL,	// close_attr
	NULL,	// free_attr_cookie
	NULL,	// read_attr
	NULL,	// write_attr

	NULL,	// read_attr_stat
	NULL,	// write_attr_stat
	NULL,	// rename_attr
	NULL,	// remove_attr

	/* support for node and FS layers */
	&rootfs_create_special_node,
	NULL,	// get_super_vnode,
};

}	// namespace

file_system_module_info gRootFileSystem = {
	{
		"file_systems/rootfs" B_CURRENT_FS_API_VERSION,
		0,
		rootfs_std_ops,
	},

	"rootfs",				// short_name
	"Root File System",		// pretty_name
	0,						// DDM flags

	NULL,	// identify_partition()
	NULL,	// scan_partition()
	NULL,	// free_identify_partition_cookie()
	NULL,	// free_partition_content_cookie()

	&rootfs_mount,
};