Index: boot/Makefile.common
===================================================================
--- boot/Makefile.common	(revision 08cba4b8a94988e64d6efb79b32a3f28694a26e9)
+++ boot/Makefile.common	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -102,4 +102,5 @@
 	$(USPACE_PATH)/srv/fs/tmpfs/tmpfs \
 	$(USPACE_PATH)/srv/fs/fat/fat \
+	$(USPACE_PATH)/srv/fs/cdfs/cdfs \
 	$(USPACE_PATH)/srv/fs/exfat/exfat \
 	$(USPACE_PATH)/srv/fs/ext2fs/ext2fs \
Index: uspace/Makefile
===================================================================
--- uspace/Makefile	(revision 08cba4b8a94988e64d6efb79b32a3f28694a26e9)
+++ uspace/Makefile	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -81,4 +81,5 @@
 	srv/fs/exfat \
 	srv/fs/fat \
+	srv/fs/cdfs \
 	srv/fs/tmpfs \
 	srv/fs/locfs \
Index: uspace/srv/fs/cdfs/Makefile
===================================================================
--- uspace/srv/fs/cdfs/Makefile	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
+++ uspace/srv/fs/cdfs/Makefile	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -0,0 +1,38 @@
+#
+# Copyright (c) 2010 Jiri Kavalik
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+#
+# - Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# - 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.
+# - The name of the author may not be used to endorse or promote products
+#   derived from this software without specific prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
+#
+
+USPACE_PREFIX = ../../..
+LIBS = $(LIBBLOCK_PREFIX)/libblock.a $(LIBFS_PREFIX)/libfs.a
+EXTRA_CFLAGS += -I$(LIBBLOCK_PREFIX) -I$(LIBFS_PREFIX)
+BINARY = cdfs
+
+SOURCES = \
+	cdfs.c \
+	cdfs_ops.c
+
+include $(USPACE_PREFIX)/Makefile.common
Index: uspace/srv/fs/cdfs/cdfs.c
===================================================================
--- uspace/srv/fs/cdfs/cdfs.c	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
+++ uspace/srv/fs/cdfs/cdfs.c	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -0,0 +1,89 @@
+/*
+ * Copyright (c) 2006 Martin Decky
+ * Copyright (c) 2008 Jakub Jermar
+ * Copyright (c) 2010 Jiri Kavalik
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * - 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.
+ * - The name of the author may not be used to endorse or promote products
+ *   derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
+ */
+
+/** @addtogroup fs
+ * @{
+ */
+
+/**
+ * @file cdfs.c
+ * @brief File system driver for ISO9660 file system.
+ */
+
+#include <ipc/services.h>
+#include <ns.h>
+#include <errno.h>
+#include <stdio.h>
+#include <libfs.h>
+#include "cdfs.h"
+#include "cdfs_ops.h"
+
+#define NAME  "cdfs"
+
+static vfs_info_t cdfs_vfs_info = {
+	.name = NAME,
+	.concurrent_read_write = false,
+	.write_retains_size = false,
+};
+
+int main(int argc, char **argv)
+{
+	printf("%s: HelenOS cdfs file system server\n", NAME);
+	
+	if (!cdfs_init()) {
+		printf("%s: failed to initialize cdfs\n", NAME);
+		return -1;
+	}
+	
+	async_sess_t *vfs_sess = service_connect_blocking(EXCHANGE_SERIALIZE,
+	    SERVICE_VFS, 0, 0);
+	if (!vfs_sess) {
+		printf("%s: Unable to connect to VFS\n", NAME);
+		return -1;
+	}
+	
+	int rc = fs_register(vfs_sess, &cdfs_vfs_info, &cdfs_ops,
+	    &cdfs_libfs_ops);
+	if (rc != EOK) {
+		printf("%s: Failed to register file system (%d)\n", NAME, rc);
+		return rc;
+	}
+	
+	printf("%s: Accepting connections\n", NAME);
+	task_retval(0);
+	async_manager();
+	
+	/* Not reached */
+	return 0;
+}
+
+/**
+ * @}
+ */
Index: uspace/srv/fs/cdfs/cdfs.h
===================================================================
--- uspace/srv/fs/cdfs/cdfs.h	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
+++ uspace/srv/fs/cdfs/cdfs.h	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -0,0 +1,45 @@
+/*
+ * Copyright (c) 2011 Martin Decky
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * - 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.
+ * - The name of the author may not be used to endorse or promote products
+ *   derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
+ */
+
+/** @addtogroup fs
+ * @{
+ */
+
+#ifndef CDFS_CDFS_H_
+#define CDFS_CDFS_H_
+
+#include <libfs.h>
+
+extern vfs_out_ops_t cdfs_ops;
+extern libfs_ops_t cdfs_libfs_ops;
+
+#endif
+
+/**
+ * @}
+ */
Index: uspace/srv/fs/cdfs/cdfs_endian.h
===================================================================
--- uspace/srv/fs/cdfs/cdfs_endian.h	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
+++ uspace/srv/fs/cdfs/cdfs_endian.h	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -0,0 +1,78 @@
+/*
+ * Copyright (c) 2011 Martin Decky
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * - 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.
+ * - The name of the author may not be used to endorse or promote products
+ *   derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
+ */
+
+/** @addtogroup fs
+ * @{
+ */
+
+#ifndef CDFS_CDFS_ENDIAN_H_
+#define CDFS_CDFS_ENDIAN_H_
+
+#include <sys/types.h>
+
+#if !(defined(__BE__) ^ defined(__LE__))
+	#error The architecture must be either big-endian or little-endian.
+#endif
+
+typedef struct {
+	uint16_t le;
+	uint16_t be;
+} __attribute__((packed)) uint16_t_lb;
+
+typedef struct {
+	uint32_t le;
+	uint32_t be;
+} __attribute__((packed)) uint32_t_lb;
+
+static inline uint16_t uint16_lb(uint16_t_lb val)
+{
+#ifdef __LE__
+	return val.le;
+#endif
+
+#ifdef __BE__
+	return val.be;
+#endif
+}
+
+static inline uint32_t uint32_lb(uint32_t_lb val)
+{
+#ifdef __LE__
+	return val.le;
+#endif
+
+#ifdef __BE__
+	return val.be;
+#endif
+}
+
+#endif
+
+/**
+ * @}
+ */
Index: uspace/srv/fs/cdfs/cdfs_ops.c
===================================================================
--- uspace/srv/fs/cdfs/cdfs_ops.c	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
+++ uspace/srv/fs/cdfs/cdfs_ops.c	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -0,0 +1,1017 @@
+/*
+ * Copyright (c) 2011 Martin Decky
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * - 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.
+ * - The name of the author may not be used to endorse or promote products
+ *   derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
+ */
+
+/** @addtogroup fs
+ * @{
+ */
+
+/**
+ * @file cdfs_ops.c
+ * @brief Implementation of VFS operations for the cdfs file system
+ *        server.
+ */
+
+#include <bool.h>
+#include <adt/hash_table.h>
+#include <malloc.h>
+#include <mem.h>
+#include <loc.h>
+#include <libfs.h>
+#include <errno.h>
+#include <libblock.h>
+#include <str.h>
+#include <byteorder.h>
+#include <macros.h>
+#include "cdfs.h"
+#include "cdfs_endian.h"
+#include "cdfs_ops.h"
+
+/** Standard CD-ROM block size */
+#define BLOCK_SIZE  2048
+
+/** Implicit node cache size
+ *
+ * More nodes can be actually cached if the files remain
+ * opended.
+ *
+ */
+#define NODE_CACHE_SIZE  200
+
+#define NODES_BUCKETS  256
+
+#define NODES_KEY_SRVC   0
+#define NODES_KEY_INDEX  1
+
+/** All root nodes have index 0 */
+#define CDFS_SOME_ROOT  0
+
+#define CDFS_NODE(node)  ((node) ? (cdfs_node_t *)(node)->data : NULL)
+#define FS_NODE(node)    ((node) ? (node)->fs_node : NULL)
+
+#define CDFS_STANDARD_IDENT  "CD001"
+
+typedef enum {
+	VOL_DESC_BOOT = 0,
+	VOL_DESC_PRIMARY = 1,
+	VOL_DESC_SUPPLEMENTARY = 2,
+	VOL_DESC_VOL_PARTITION = 3,
+	VOL_DESC_SET_TERMINATOR = 255
+} vol_desc_type_t;
+
+typedef struct {
+	uint8_t system_ident[32];
+	uint8_t ident[32];
+} __attribute__((packed)) cdfs_vol_desc_boot_t;
+
+typedef struct {
+	uint8_t year[4];
+	uint8_t mon[2];
+	uint8_t day[2];
+	
+	uint8_t hour[2];
+	uint8_t min[2];
+	uint8_t sec[2];
+	uint8_t msec[2];
+	
+	uint8_t offset;
+} __attribute__((packed)) cdfs_datetime_t;
+
+typedef struct {
+	uint8_t year;  /**< Since 1900 */
+	uint8_t mon;
+	uint8_t day;
+	
+	uint8_t hour;
+	uint8_t min;
+	uint8_t sec;
+	
+	uint8_t offset;
+} __attribute__((packed)) cdfs_timestamp_t;
+
+typedef enum {
+	DIR_FLAG_DIRECTORY = 2
+} cdfs_dir_flag_t;
+
+typedef struct {
+	uint8_t length;
+	uint8_t ea_length;
+	
+	uint32_t_lb lba;
+	uint32_t_lb size;
+	
+	cdfs_timestamp_t timestamp;
+	uint8_t flags;
+	uint8_t unit_size;
+	uint8_t gap_size;
+	uint16_t_lb sequence_nr;
+	
+	uint8_t name_length;
+	uint8_t name[];
+} __attribute__((packed)) cdfs_dir_t;
+
+typedef struct {
+	uint8_t res0;
+	
+	uint8_t system_ident[32];
+	uint8_t ident[32];
+	
+	uint64_t res1;
+	uint32_t_lb lba_size;
+	
+	uint8_t res2[32];
+	uint16_t_lb set_size;
+	uint16_t_lb sequence_nr;
+	
+	uint16_t_lb block_size;
+	uint32_t_lb path_table_size;
+	
+	uint32_t path_table_lsb;
+	uint32_t opt_path_table_lsb;
+	uint32_t path_table_msb;
+	uint32_t opt_path_table_msb;
+	
+	cdfs_dir_t root_dir;
+	uint8_t pad0;
+	
+	uint8_t set_ident[128];
+	uint8_t publisher_ident[128];
+	uint8_t preparer_ident[128];
+	uint8_t app_ident[128];
+	
+	uint8_t copyright_file_ident[37];
+	uint8_t abstract_file_ident[37];
+	uint8_t biblio_file_ident[37];
+	
+	cdfs_datetime_t creation;
+	cdfs_datetime_t modification;
+	cdfs_datetime_t expiration;
+	cdfs_datetime_t effective;
+	
+	uint8_t fs_version;
+} __attribute__((packed)) cdfs_vol_desc_primary_t;
+
+typedef struct {
+	uint8_t type;
+	uint8_t standard_ident[5];
+	uint8_t version;
+	union {
+		cdfs_vol_desc_boot_t boot;
+		cdfs_vol_desc_primary_t primary;
+	} data;
+} __attribute__((packed)) cdfs_vol_desc_t;
+
+typedef enum {
+	CDFS_NONE,
+	CDFS_FILE,
+	CDFS_DIRECTORY
+} cdfs_dentry_type_t;
+
+typedef struct {
+	link_t link;       /**< Siblings list link */
+	fs_index_t index;  /**< Node index */
+	char *name;        /**< Dentry name */
+} cdfs_dentry_t;
+
+typedef uint32_t cdfs_lba_t;
+
+typedef struct {
+	fs_node_t *fs_node;       /**< FS node */
+	fs_index_t index;         /**< Node index */
+	service_id_t service_id;  /**< Service ID of block device */
+	
+	link_t nh_link;           /**< Nodes hash table link */
+	cdfs_dentry_type_t type;  /**< Dentry type */
+	
+	unsigned int lnkcnt;      /**< Link count */
+	uint32_t size;            /**< File size if type is CDFS_FILE */
+	
+	list_t cs_list;           /**< Child's siblings list */
+	cdfs_lba_t lba;           /**< LBA of data on disk */
+	bool processed;           /**< If all children have been read */
+	unsigned int opened;      /**< Opened count */
+} cdfs_node_t;
+
+/** Shared index of nodes */
+static fs_index_t cdfs_index = 1;
+
+/** Number of currently cached nodes */
+static size_t nodes_cached = 0;
+
+/** Hash table of all cdfs nodes */
+static hash_table_t nodes;
+
+static hash_index_t nodes_hash(unsigned long key[])
+{
+	return key[NODES_KEY_INDEX] % NODES_BUCKETS;
+}
+
+static int nodes_compare(unsigned long key[], hash_count_t keys, link_t *item)
+{
+	cdfs_node_t *node =
+	    hash_table_get_instance(item, cdfs_node_t, nh_link);
+	
+	switch (keys) {
+	case 1:
+		return (node->service_id == key[NODES_KEY_SRVC]);
+	case 2:
+		return ((node->service_id == key[NODES_KEY_SRVC]) &&
+		    (node->index == key[NODES_KEY_INDEX]));
+	default:
+		assert((keys == 1) || (keys == 2));
+	}
+	
+	return 0;
+}
+
+static void nodes_remove_callback(link_t *item)
+{
+	cdfs_node_t *node =
+	    hash_table_get_instance(item, cdfs_node_t, nh_link);
+	
+	assert(node->type == CDFS_DIRECTORY);
+	
+	link_t *link;
+	while ((link = list_first(&node->cs_list)) != NULL) {
+		cdfs_dentry_t *dentry = list_get_instance(link, cdfs_dentry_t, link);
+		list_remove(&dentry->link);
+		free(dentry);
+	}
+	
+	free(node->fs_node);
+	free(node);
+}
+
+/** Nodes hash table operations */
+static hash_table_operations_t nodes_ops = {
+	.hash = nodes_hash,
+	.compare = nodes_compare,
+	.remove_callback = nodes_remove_callback
+};
+
+static int cdfs_node_get(fs_node_t **rfn, service_id_t service_id,
+    fs_index_t index)
+{
+	unsigned long key[] = {
+		[NODES_KEY_SRVC] = service_id,
+		[NODES_KEY_INDEX] = index
+	};
+	
+	link_t *link = hash_table_find(&nodes, key);
+	if (link) {
+		cdfs_node_t *node =
+		    hash_table_get_instance(link, cdfs_node_t, nh_link);
+		
+		*rfn = FS_NODE(node);
+	} else
+		*rfn = NULL;
+	
+	return EOK;
+}
+
+static int cdfs_root_get(fs_node_t **rfn, service_id_t service_id)
+{
+	return cdfs_node_get(rfn, service_id, CDFS_SOME_ROOT);
+}
+
+static void cdfs_node_initialize(cdfs_node_t *node)
+{
+	node->fs_node = NULL;
+	node->index = 0;
+	node->service_id = 0;
+	node->type = CDFS_NONE;
+	node->lnkcnt = 0;
+	node->size = 0;
+	node->lba = 0;
+	node->processed = false;
+	node->opened = 0;
+	
+	link_initialize(&node->nh_link);
+	list_initialize(&node->cs_list);
+}
+
+static int create_node(fs_node_t **rfn, service_id_t service_id, int lflag,
+    fs_index_t index)
+{
+	assert((lflag & L_FILE) ^ (lflag & L_DIRECTORY));
+	
+	cdfs_node_t *node = malloc(sizeof(cdfs_node_t));
+	if (!node)
+		return ENOMEM;
+	
+	cdfs_node_initialize(node);
+	
+	node->fs_node = malloc(sizeof(fs_node_t));
+	if (!node->fs_node) {
+		free(node);
+		return ENOMEM;
+	}
+	
+	fs_node_initialize(node->fs_node);
+	node->fs_node->data = node;
+	
+	fs_node_t *rootfn;
+	int rc = cdfs_root_get(&rootfn, service_id);
+	
+	assert(rc == EOK);
+	
+	if (!rootfn)
+		node->index = CDFS_SOME_ROOT;
+	else
+		node->index = index;
+	
+	node->service_id = service_id;
+	
+	if (lflag & L_DIRECTORY)
+		node->type = CDFS_DIRECTORY;
+	else
+		node->type = CDFS_FILE;
+	
+	/* Insert the new node into the nodes hash table. */
+	unsigned long key[] = {
+		[NODES_KEY_SRVC] = node->service_id,
+		[NODES_KEY_INDEX] = node->index
+	};
+	
+	hash_table_insert(&nodes, key, &node->nh_link);
+	
+	*rfn = FS_NODE(node);
+	nodes_cached++;
+	
+	return EOK;
+}
+
+static int link_node(fs_node_t *pfn, fs_node_t *fn, const char *name)
+{
+	cdfs_node_t *parent = CDFS_NODE(pfn);
+	cdfs_node_t *node = CDFS_NODE(fn);
+	
+	assert(parent->type == CDFS_DIRECTORY);
+	
+	/* Check for duplicate entries */
+	list_foreach(parent->cs_list, link) {
+		cdfs_dentry_t *dentry =
+		    list_get_instance(link, cdfs_dentry_t, link);
+		
+		if (str_cmp(dentry->name, name) == 0)
+			return EEXIST;
+	}
+	
+	/* Allocate and initialize the dentry */
+	cdfs_dentry_t *dentry = malloc(sizeof(cdfs_dentry_t));
+	if (!dentry)
+		return ENOMEM;
+	
+	/* Populate and link the new dentry */
+	dentry->name = str_dup(name);
+	if (dentry->name == NULL) {
+		free(dentry);
+		return ENOMEM;
+	}
+	
+	link_initialize(&dentry->link);
+	dentry->index = node->index;
+	
+	node->lnkcnt++;
+	list_append(&dentry->link, &parent->cs_list);
+	
+	return EOK;
+}
+
+static bool cdfs_readdir(service_id_t service_id, fs_node_t *fs_node)
+{
+	cdfs_node_t *node = CDFS_NODE(fs_node);
+	assert(node);
+	
+	if (node->processed)
+		return true;
+	
+	uint32_t blocks = node->size / BLOCK_SIZE;
+	if ((node->size % BLOCK_SIZE) != 0)
+		blocks++;
+	
+	for (uint32_t i = 0; i < blocks; i++) {
+		block_t *block;
+		int rc = block_get(&block, service_id, node->lba + i, BLOCK_FLAGS_NONE);
+		if (rc != EOK)
+			return false;
+		
+		cdfs_dir_t *dir = (cdfs_dir_t *) block->data;
+		
+		// FIXME: skip '.' and '..'
+		
+		for (size_t offset = 0;
+		    (dir->length != 0) && (offset < BLOCK_SIZE);
+		    offset += dir->length) {
+			dir = (cdfs_dir_t *) (block->data + offset);
+			
+			cdfs_dentry_type_t dentry_type;
+			if (dir->flags & DIR_FLAG_DIRECTORY)
+				dentry_type = CDFS_DIRECTORY;
+			else
+				dentry_type = CDFS_FILE;
+			
+			// FIXME: hack - indexing by dentry byte offset on disc
+			
+			fs_node_t *fn;
+			int rc = create_node(&fn, service_id, dentry_type,
+			    (node->lba + i) * BLOCK_SIZE + offset);
+			if ((rc != EOK) || (fn == NULL))
+				return false;
+			
+			cdfs_node_t *cur = CDFS_NODE(fn);
+			cur->lba = uint32_lb(dir->lba);
+			cur->size = uint32_lb(dir->size);
+			
+			char *name = (char *) malloc(dir->name_length + 1);
+			if (name == NULL)
+				return false;
+			
+			memcpy(name, dir->name, dir->name_length);
+			name[dir->name_length] = 0;
+			
+			// FIXME: check return value
+			
+			link_node(fs_node, fn, name);
+			free(name);
+			
+			if (dentry_type == CDFS_FILE)
+				cur->processed = true;
+		}
+		
+		block_put(block);
+	}
+	
+	node->processed = true;
+	return true;
+}
+
+static fs_node_t *get_uncached_node(service_id_t service_id, fs_index_t index)
+{
+	cdfs_lba_t lba = index / BLOCK_SIZE;
+	size_t offset = index % BLOCK_SIZE;
+	
+	block_t *block;
+	int rc = block_get(&block, service_id, lba, BLOCK_FLAGS_NONE);
+	if (rc != EOK)
+		return NULL;
+	
+	cdfs_dir_t *dir = (cdfs_dir_t *) (block->data + offset);
+	
+	cdfs_dentry_type_t dentry_type;
+	if (dir->flags & DIR_FLAG_DIRECTORY)
+		dentry_type = CDFS_DIRECTORY;
+	else
+		dentry_type = CDFS_FILE;
+	
+	fs_node_t *fn;
+	rc = create_node(&fn, service_id, dentry_type, index);
+	if ((rc != EOK) || (fn == NULL))
+		return NULL;
+	
+	cdfs_node_t *node = CDFS_NODE(fn);
+	node->lba = uint32_lb(dir->lba);
+	node->size = uint32_lb(dir->size);
+	node->lnkcnt = 1;
+	
+	if (dentry_type == CDFS_FILE)
+		node->processed = true;
+	
+	block_put(block);
+	return fn;
+}
+
+static fs_node_t *get_cached_node(service_id_t service_id, fs_index_t index)
+{
+	unsigned long key[] = {
+		[NODES_KEY_SRVC] = service_id,
+		[NODES_KEY_INDEX] = index
+	};
+	
+	link_t *link = hash_table_find(&nodes, key);
+	if (link) {
+		cdfs_node_t *node =
+		    hash_table_get_instance(link, cdfs_node_t, nh_link);
+		return FS_NODE(node);
+	}
+	
+	return get_uncached_node(service_id, index);
+}
+
+static int cdfs_match(fs_node_t **fn, fs_node_t *pfn, const char *component)
+{
+	cdfs_node_t *parent = CDFS_NODE(pfn);
+	
+	if (!parent->processed) {
+		int rc = cdfs_readdir(parent->service_id, pfn);
+		if (rc != EOK)
+			return rc;
+	}
+	
+	list_foreach(parent->cs_list, link) {
+		cdfs_dentry_t *dentry =
+		    list_get_instance(link, cdfs_dentry_t, link);
+		
+		if (str_cmp(dentry->name, component) == 0) {
+			*fn = get_cached_node(parent->service_id, dentry->index);
+			return EOK;
+		}
+	}
+	
+	*fn = NULL;
+	return EOK;
+}
+
+static int cdfs_node_open(fs_node_t *fn)
+{
+	cdfs_node_t *node = CDFS_NODE(fn);
+	
+	if (!node->processed)
+		cdfs_readdir(node->service_id, fn);
+	
+	node->opened++;
+	return EOK;
+}
+
+static int cdfs_node_put(fs_node_t *fn)
+{
+	/* Nothing to do */
+	return EOK;
+}
+
+static int cdfs_create_node(fs_node_t **fn, service_id_t service_id, int lflag)
+{
+	/* Read-only */
+	return ENOTSUP;
+}
+
+static int cdfs_destroy_node(fs_node_t *fn)
+{
+	/* Read-only */
+	return ENOTSUP;
+}
+
+static int cdfs_link_node(fs_node_t *pfn, fs_node_t *cfn, const char *name)
+{
+	/* Read-only */
+	return ENOTSUP;
+}
+
+static int cdfs_unlink_node(fs_node_t *pfn, fs_node_t *cfn, const char *name)
+{
+	/* Read-only */
+	return ENOTSUP;
+}
+
+static int cdfs_has_children(bool *has_children, fs_node_t *fn)
+{
+	cdfs_node_t *node = CDFS_NODE(fn);
+	
+	if ((node->type == CDFS_DIRECTORY) && (!node->processed))
+		cdfs_readdir(node->service_id, fn);
+	
+	*has_children = !list_empty(&node->cs_list);
+	return EOK;
+}
+
+static fs_index_t cdfs_index_get(fs_node_t *fn)
+{
+	cdfs_node_t *node = CDFS_NODE(fn);
+	return node->index;
+}
+
+static aoff64_t cdfs_size_get(fs_node_t *fn)
+{
+	cdfs_node_t *node = CDFS_NODE(fn);
+	return node->size;
+}
+
+static unsigned int cdfs_lnkcnt_get(fs_node_t *fn)
+{
+	cdfs_node_t *node = CDFS_NODE(fn);
+	return node->lnkcnt;
+}
+
+static bool cdfs_is_directory(fs_node_t *fn)
+{
+	cdfs_node_t *node = CDFS_NODE(fn);
+	return (node->type == CDFS_DIRECTORY);
+}
+
+static bool cdfs_is_file(fs_node_t *fn)
+{
+	cdfs_node_t *node = CDFS_NODE(fn);
+	return (node->type == CDFS_FILE);
+}
+
+static service_id_t cdfs_device_get(fs_node_t *fn)
+{
+	return 0;
+}
+
+libfs_ops_t cdfs_libfs_ops = {
+	.root_get = cdfs_root_get,
+	.match = cdfs_match,
+	.node_get = cdfs_node_get,
+	.node_open = cdfs_node_open,
+	.node_put = cdfs_node_put,
+	.create = cdfs_create_node,
+	.destroy = cdfs_destroy_node,
+	.link = cdfs_link_node,
+	.unlink = cdfs_unlink_node,
+	.has_children = cdfs_has_children,
+	.index_get = cdfs_index_get,
+	.size_get = cdfs_size_get,
+	.lnkcnt_get = cdfs_lnkcnt_get,
+	.is_directory = cdfs_is_directory,
+	.is_file = cdfs_is_file,
+	.device_get = cdfs_device_get
+};
+
+static bool iso_readfs(service_id_t service_id, fs_node_t *rfn,
+    cdfs_lba_t altroot)
+{
+	/* First 16 blocks of isofs are empty */
+	block_t *block;
+	int rc = block_get(&block, service_id, altroot + 16, BLOCK_FLAGS_NONE);
+	if (rc != EOK)
+		return false;
+	
+	cdfs_vol_desc_t *vol_desc = (cdfs_vol_desc_t *) block->data;
+	
+	/*
+	 * Test for primary volume descriptor
+	 * and standard compliance.
+	 */
+	if ((vol_desc->type != VOL_DESC_PRIMARY) ||
+	    (bcmp(vol_desc->standard_ident, CDFS_STANDARD_IDENT, 5) != 0) ||
+	    (vol_desc->version != 1)) {
+		block_put(block);
+		return false;
+	}
+	
+	uint16_t set_size = uint16_lb(vol_desc->data.primary.set_size);
+	if (set_size > 1) {
+		/*
+		 * Technically, we don't support multi-disc sets.
+		 * But one can encounter erroneously mastered
+		 * images in the wild and it might actually work
+		 * for the first disc in the set.
+		 */
+	}
+	
+	uint16_t sequence_nr = uint16_lb(vol_desc->data.primary.sequence_nr);
+	if (sequence_nr != 1) {
+		/*
+		 * We only support the first disc
+		 * in multi-disc sets.
+		 */
+		block_put(block);
+		return false;
+	}
+	
+	uint16_t block_size = uint16_lb(vol_desc->data.primary.block_size);
+	if (block_size != BLOCK_SIZE) {
+		block_put(block);
+		return false;
+	}
+	
+	// TODO: implement path table support
+	
+	cdfs_node_t *node = CDFS_NODE(rfn);
+	node->lba = uint32_lb(vol_desc->data.primary.root_dir.lba);
+	node->size = uint32_lb(vol_desc->data.primary.root_dir.size);
+	
+	if (!cdfs_readdir(service_id, rfn)) {
+		block_put(block);
+		return false;
+	}
+	
+	block_put(block);
+	return true;
+}
+
+/* Mount a session with session start offset
+ *
+ */
+static bool cdfs_instance_init(service_id_t service_id, cdfs_lba_t altroot)
+{
+	/* Create root node */
+	fs_node_t *rfn;
+	int rc = create_node(&rfn, service_id, L_DIRECTORY, cdfs_index++);
+	
+	if ((rc != EOK) || (!rfn))
+		return false;
+	
+	/* FS root is not linked */
+	CDFS_NODE(rfn)->lnkcnt = 0;
+	CDFS_NODE(rfn)->lba = 0;
+	CDFS_NODE(rfn)->processed = false;
+	
+	/* Check if there is cdfs in given session */
+	if (!iso_readfs(service_id, rfn, altroot)) {
+		// XXX destroy node
+		return false;
+	}
+	
+	return true;
+}
+
+static int cdfs_mounted(service_id_t service_id, const char *opts,
+    fs_index_t *index, aoff64_t *size, unsigned int *lnkcnt)
+{
+	/* Initialize the block layer */
+	int rc = block_init(EXCHANGE_SERIALIZE, service_id, BLOCK_SIZE);
+	if (rc != EOK)
+		return rc;
+	
+	cdfs_lba_t altroot = 0;
+	
+	if (str_lcmp(opts, "altroot=", 8) == 0) {
+		/* User-defined alternative root on a multi-session disk */
+		if (str_uint32_t(opts + 8, NULL, 0, false, &altroot) != EOK)
+			altroot = 0;
+	} else {
+		/* Read TOC and find the last session */
+		toc_block_t *toc = block_get_toc(service_id, 1);
+		if ((toc != NULL) && (uint16_t_be2host(toc->size) == 10)) {
+			altroot = uint32_t_be2host(toc->first_lba);
+			free(toc);
+		}
+	}
+	
+	/* Initialize the block cache */
+	rc = block_cache_init(service_id, BLOCK_SIZE, 0, CACHE_MODE_WT);
+	if (rc != EOK) {
+		block_fini(service_id);
+		return rc;
+	}
+	
+	/* Check if this device is not already mounted */
+	fs_node_t *rootfn;
+	rc = cdfs_root_get(&rootfn, service_id);
+	if ((rc == EOK) && (rootfn)) {
+		cdfs_node_put(rootfn);
+		block_cache_fini(service_id);
+		block_fini(service_id);
+		
+		return EEXIST;
+	}
+	
+	/* Initialize cdfs instance */
+	if (!cdfs_instance_init(service_id, altroot)) {
+		block_cache_fini(service_id);
+		block_fini(service_id);
+		
+		return ENOMEM;
+	}
+	
+	rc = cdfs_root_get(&rootfn, service_id);
+	assert(rc == EOK);
+	
+	cdfs_node_t *root = CDFS_NODE(rootfn);
+	*index = root->index;
+	*size = root->size;
+	*lnkcnt = root->lnkcnt;
+	
+	return EOK;
+}
+
+static void cdfs_instance_done(service_id_t service_id)
+{
+	unsigned long key[] = {
+		[NODES_KEY_SRVC] = service_id
+	};
+	
+	hash_table_remove(&nodes, key, 1);
+	block_cache_fini(service_id);
+	block_fini(service_id);
+}
+
+static int cdfs_unmounted(service_id_t service_id)
+{
+	cdfs_instance_done(service_id);
+	return EOK;
+}
+
+static int cdfs_read(service_id_t service_id, fs_index_t index, aoff64_t pos,
+    size_t *rbytes)
+{
+	unsigned long key[] = {
+		[NODES_KEY_SRVC] = service_id,
+		[NODES_KEY_INDEX] = index
+	};
+	
+	link_t *link = hash_table_find(&nodes, key);
+	if (link == NULL)
+		return ENOENT;
+	
+	cdfs_node_t *node =
+	    hash_table_get_instance(link, cdfs_node_t, nh_link);
+	
+	if (!node->processed) {
+		int rc = cdfs_readdir(service_id, FS_NODE(node));
+		if (rc != EOK)
+			return rc;
+	}
+	
+	ipc_callid_t callid;
+	size_t len;
+	if (!async_data_read_receive(&callid, &len)) {
+		async_answer_0(callid, EINVAL);
+		return EINVAL;
+	}
+	
+	if (node->type == CDFS_FILE) {
+		if (pos >= node->size) {
+			*rbytes = 0;
+			async_data_read_finalize(callid, NULL, 0);
+		} else {
+			cdfs_lba_t lba = pos / BLOCK_SIZE;
+			size_t offset = pos % BLOCK_SIZE;
+			
+			*rbytes = min(len, BLOCK_SIZE - offset);
+			*rbytes = min(*rbytes, node->size - pos);
+			
+			block_t *block;
+			int rc = block_get(&block, service_id, node->lba + lba,
+			    BLOCK_FLAGS_NONE);
+			if (rc != EOK) {
+				async_answer_0(callid, rc);
+				return rc;
+			}
+			
+			async_data_read_finalize(callid, block->data + offset,
+			    *rbytes);
+			rc = block_put(block);
+			if (rc != EOK)
+				return rc;
+		}
+	} else {
+		link_t *link = list_nth(&node->cs_list, pos);
+		if (link == NULL) {
+			async_answer_0(callid, ENOENT);
+			return ENOENT;
+		}
+		
+		cdfs_dentry_t *dentry =
+		    list_get_instance(link, cdfs_dentry_t, link);
+		
+		*rbytes = 1;
+		async_data_read_finalize(callid, dentry->name,
+		    str_size(dentry->name) + 1);
+	}
+	
+	return EOK;
+}
+
+static int cdfs_write(service_id_t service_id, fs_index_t index, aoff64_t pos,
+    size_t *wbytes, aoff64_t *nsize)
+{
+	/*
+	 * As cdfs is a read-only filesystem,
+	 * the operation is not supported.
+	 */
+	
+	return ENOTSUP;
+}
+
+static int cdfs_truncate(service_id_t service_id, fs_index_t index,
+    aoff64_t size)
+{
+	/*
+	 * As cdfs is a read-only filesystem,
+	 * the operation is not supported.
+	 */
+	
+	return ENOTSUP;
+}
+
+static void cleanup_cache(service_id_t service_id)
+{
+	if (nodes_cached > NODE_CACHE_SIZE) {
+		size_t remove = nodes_cached - NODE_CACHE_SIZE;
+		
+		// FIXME: this accesses the internals of the hash table
+		//        and should be rewritten in a clean way
+		
+		for (hash_index_t chain = 0; chain < nodes.entries; chain++) {
+			for (link_t *link = nodes.entry[chain].head.next;
+			    link != &nodes.entry[chain].head;
+			    link = link->next) {
+				if (remove == 0)
+					return;
+				
+				cdfs_node_t *node =
+				    hash_table_get_instance(link, cdfs_node_t, nh_link);
+				if (node->opened == 0) {
+					link_t *tmp = link;
+					link = link->prev;
+					
+					list_remove(tmp);
+					nodes.op->remove_callback(tmp);
+					nodes_cached--;
+					remove--;
+					
+					continue;
+				}
+			}
+		}
+	}
+}
+
+static int cdfs_close(service_id_t service_id, fs_index_t index)
+{
+	/* Root node is always in memory */
+	if (index == 0)
+		return EOK;
+	
+	unsigned long key[] = {
+		[NODES_KEY_SRVC] = service_id,
+		[NODES_KEY_INDEX] = index
+	};
+	
+	link_t *link = hash_table_find(&nodes, key);
+	if (link == 0)
+		return ENOENT;
+	
+	cdfs_node_t *node =
+	    hash_table_get_instance(link, cdfs_node_t, nh_link);
+	
+	assert(node->opened > 0);
+	
+	node->opened--;
+	cleanup_cache(service_id);
+	
+	return EOK;
+}
+
+static int cdfs_destroy(service_id_t service_id, fs_index_t index)
+{
+	/*
+	 * As cdfs is a read-only filesystem,
+	 * the operation is not supported.
+	 */
+	
+	return ENOTSUP;
+}
+
+static int cdfs_sync(service_id_t service_id, fs_index_t index)
+{
+	/*
+	 * As cdfs is a read-only filesystem,
+	 * the sync operation is a no-op.
+	 */
+	
+	return EOK;
+}
+
+vfs_out_ops_t cdfs_ops = {
+	.mounted = cdfs_mounted,
+	.unmounted = cdfs_unmounted,
+	.read = cdfs_read,
+	.write = cdfs_write,
+	.truncate = cdfs_truncate,
+	.close = cdfs_close,
+	.destroy = cdfs_destroy,
+	.sync = cdfs_sync
+};
+
+/** Initialize the cdfs server
+ *
+ */
+bool cdfs_init(void)
+{
+	if (!hash_table_create(&nodes, NODES_BUCKETS, 2, &nodes_ops))
+		return false;
+	
+	return true;
+}
+
+/**
+ * @}
+ */
Index: uspace/srv/fs/cdfs/cdfs_ops.h
===================================================================
--- uspace/srv/fs/cdfs/cdfs_ops.h	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
+++ uspace/srv/fs/cdfs/cdfs_ops.h	(revision a26895ddf444f99491f445f5813ad4b7cbf05a53)
@@ -0,0 +1,44 @@
+/*
+ * Copyright (c) 2011 Martin Decky
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * - 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.
+ * - The name of the author may not be used to endorse or promote products
+ *   derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
+ */
+
+/** @addtogroup fs
+ * @{
+ */
+
+#ifndef CDFS_CDFS_OPS_H_
+#define CDFS_CDFS_OPS_H_
+
+#include <bool.h>
+
+extern bool cdfs_init(void);
+
+#endif
+
+/**
+ * @}
+ */
