source: mainline/uspace/srv/fs/fat/fat_ops.c@ e32b65a

/*
 * Copyright (c) 2008 Jakub Jermar
 * 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        fat_ops.c
 * @brief       Implementation of VFS operations for the FAT file system server.
 */

#include "fat.h"
#include "fat_dentry.h"
#include "fat_fat.h"
#include "../../vfs/vfs.h"
#include <libfs.h>
#include <libblock.h>
#include <ipc/ipc.h>
#include <ipc/services.h>
#include <ipc/devmap.h>
#include <async.h>
#include <errno.h>
#include <string.h>
#include <byteorder.h>
#include <libadt/hash_table.h>
#include <libadt/list.h>
#include <assert.h>
#include <futex.h>
#include <sys/mman.h>
#include <align.h>

/** Futex protecting the list of cached free FAT nodes. */
static futex_t ffn_futex = FUTEX_INITIALIZER;

/** List of cached free FAT nodes. */
static LIST_INITIALIZE(ffn_head);

static void fat_node_initialize(fat_node_t *node)
{
        futex_initialize(&node->lock, 1);
        node->idx = NULL;
        node->type = 0;
        link_initialize(&node->ffn_link);
        node->size = 0;
        node->lnkcnt = 0;
        node->refcnt = 0;
        node->dirty = false;
}

static void fat_node_sync(fat_node_t *node)
{
        block_t *b;
        fat_bs_t *bs;
        fat_dentry_t *d;
        uint16_t bps;
        unsigned dps;
        
        assert(node->dirty);

        bs = block_bb_get(node->idx->dev_handle);
        bps = uint16_t_le2host(bs->bps);
        dps = bps / sizeof(fat_dentry_t);
        
        /* Read the block that contains the dentry of interest. */
        b = _fat_block_get(bs, node->idx->dev_handle, node->idx->pfc,
            (node->idx->pdi * sizeof(fat_dentry_t)) / bps, BLOCK_FLAGS_NONE);

        d = ((fat_dentry_t *)b->data) + (node->idx->pdi % dps);

        d->firstc = host2uint16_t_le(node->firstc);
        if (node->type == FAT_FILE) {
                d->size = host2uint32_t_le(node->size);
        } else if (node->type == FAT_DIRECTORY) {
                d->attr = FAT_ATTR_SUBDIR;
        }
        
        /* TODO: update other fields? (e.g time fields) */
        
        b->dirty = true;                /* need to sync block */
        block_put(b);
}

static fat_node_t *fat_node_get_new(void)
{
        fat_node_t *nodep;

        futex_down(&ffn_futex);
        if (!list_empty(&ffn_head)) {
                /* Try to use a cached free node structure. */
                fat_idx_t *idxp_tmp;
                nodep = list_get_instance(ffn_head.next, fat_node_t, ffn_link);
                if (futex_trydown(&nodep->lock) == ESYNCH_WOULD_BLOCK)
                        goto skip_cache;
                idxp_tmp = nodep->idx;
                if (futex_trydown(&idxp_tmp->lock) == ESYNCH_WOULD_BLOCK) {
                        futex_up(&nodep->lock);
                        goto skip_cache;
                }
                list_remove(&nodep->ffn_link);
                futex_up(&ffn_futex);
                if (nodep->dirty)
                        fat_node_sync(nodep);
                idxp_tmp->nodep = NULL;
                futex_up(&nodep->lock);
                futex_up(&idxp_tmp->lock);
        } else {
skip_cache:
                /* Try to allocate a new node structure. */
                futex_up(&ffn_futex);
                nodep = (fat_node_t *)malloc(sizeof(fat_node_t));
                if (!nodep)
                        return NULL;
        }
        fat_node_initialize(nodep);
        
        return nodep;
}

/** Internal version of fat_node_get().
 *
 * @param idxp          Locked index structure.
 */
static void *fat_node_get_core(fat_idx_t *idxp)
{
        block_t *b;
        fat_bs_t *bs;
        fat_dentry_t *d;
        fat_node_t *nodep = NULL;
        unsigned bps;
        unsigned spc;
        unsigned dps;

        if (idxp->nodep) {
                /*
                 * We are lucky.
                 * The node is already instantiated in memory.
                 */
                futex_down(&idxp->nodep->lock);
                if (!idxp->nodep->refcnt++)
                        list_remove(&idxp->nodep->ffn_link);
                futex_up(&idxp->nodep->lock);
                return idxp->nodep;
        }

        /*
         * We must instantiate the node from the file system.
         */
        
        assert(idxp->pfc);

        nodep = fat_node_get_new();
        if (!nodep)
                return NULL;

        bs = block_bb_get(idxp->dev_handle);
        bps = uint16_t_le2host(bs->bps);
        spc = bs->spc;
        dps = bps / sizeof(fat_dentry_t);

        /* Read the block that contains the dentry of interest. */
        b = _fat_block_get(bs, idxp->dev_handle, idxp->pfc,
            (idxp->pdi * sizeof(fat_dentry_t)) / bps, BLOCK_FLAGS_NONE);
        assert(b);

        d = ((fat_dentry_t *)b->data) + (idxp->pdi % dps);
        if (d->attr & FAT_ATTR_SUBDIR) {
                /* 
                 * The only directory which does not have this bit set is the
                 * root directory itself. The root directory node is handled
                 * and initialized elsewhere.
                 */
                nodep->type = FAT_DIRECTORY;
                /*
                 * Unfortunately, the 'size' field of the FAT dentry is not
                 * defined for the directory entry type. We must determine the
                 * size of the directory by walking the FAT.
                 */
                nodep->size = bps * spc * fat_clusters_get(bs, idxp->dev_handle,
                    uint16_t_le2host(d->firstc));
        } else {
                nodep->type = FAT_FILE;
                nodep->size = uint32_t_le2host(d->size);
        }
        nodep->firstc = uint16_t_le2host(d->firstc);
        nodep->lnkcnt = 1;
        nodep->refcnt = 1;

        block_put(b);

        /* Link the idx structure with the node structure. */
        nodep->idx = idxp;
        idxp->nodep = nodep;

        return nodep;
}

/*
 * Forward declarations of FAT libfs operations.
 */
static void *fat_node_get(dev_handle_t, fs_index_t);
static void fat_node_put(void *);
static void *fat_create_node(dev_handle_t, int);
static int fat_destroy_node(void *);
static int fat_link(void *, void *, const char *);
static int fat_unlink(void *, void *);
static void *fat_match(void *, const char *);
static fs_index_t fat_index_get(void *);
static size_t fat_size_get(void *);
static unsigned fat_lnkcnt_get(void *);
static bool fat_has_children(void *);
static void *fat_root_get(dev_handle_t);
static char fat_plb_get_char(unsigned);
static bool fat_is_directory(void *);
static bool fat_is_file(void *node);

/*
 * FAT libfs operations.
 */

/** Instantiate a FAT in-core node. */
void *fat_node_get(dev_handle_t dev_handle, fs_index_t index)
{
        void *node;
        fat_idx_t *idxp;

        idxp = fat_idx_get_by_index(dev_handle, index);
        if (!idxp)
                return NULL;
        /* idxp->lock held */
        node = fat_node_get_core(idxp);
        futex_up(&idxp->lock);
        return node;
}

void fat_node_put(void *node)
{
        fat_node_t *nodep = (fat_node_t *)node;
        bool destroy = false;

        futex_down(&nodep->lock);
        if (!--nodep->refcnt) {
                if (nodep->idx) {
                        futex_down(&ffn_futex);
                        list_append(&nodep->ffn_link, &ffn_head);
                        futex_up(&ffn_futex);
                } else {
                        /*
                         * The node does not have any index structure associated
                         * with itself. This can only mean that we are releasing
                         * the node after a failed attempt to allocate the index
                         * structure for it.
                         */
                        destroy = true;
                }
        }
        futex_up(&nodep->lock);
        if (destroy)
                free(node);
}

void *fat_create_node(dev_handle_t dev_handle, int flags)
{
        fat_idx_t *idxp;
        fat_node_t *nodep;
        fat_bs_t *bs;
        fat_cluster_t mcl, lcl;
        uint16_t bps;
        int rc;

        bs = block_bb_get(dev_handle);
        bps = uint16_t_le2host(bs->bps);
        if (flags & L_DIRECTORY) {
                /* allocate a cluster */
                rc = fat_alloc_clusters(bs, dev_handle, 1, &mcl, &lcl);
                if (rc != EOK) 
                        return NULL;
        }

        nodep = fat_node_get_new();
        if (!nodep) {
                fat_free_clusters(bs, dev_handle, mcl); 
                return NULL;
        }
        idxp = fat_idx_get_new(dev_handle);
        if (!idxp) {
                fat_free_clusters(bs, dev_handle, mcl); 
                fat_node_put(nodep);
                return NULL;
        }
        /* idxp->lock held */
        if (flags & L_DIRECTORY) {
                int i;
                block_t *b;

                /*
                 * Populate the new cluster with unused dentries.
                 */
                for (i = 0; i < bs->spc; i++) {
                        b = _fat_block_get(bs, dev_handle, mcl, i,
                            BLOCK_FLAGS_NOREAD);
                        /* mark all dentries as never-used */
                        memset(b->data, 0, bps);
                        b->dirty = false;
                        block_put(b);
                }
                nodep->type = FAT_DIRECTORY;
                nodep->firstc = mcl;
                nodep->size = bps * bs->spc;
        } else {
                nodep->type = FAT_FILE;
                nodep->firstc = FAT_CLST_RES0;
                nodep->size = 0;
        }
        nodep->lnkcnt = 0;      /* not linked anywhere */
        nodep->refcnt = 1;
        nodep->dirty = true;

        nodep->idx = idxp;
        idxp->nodep = nodep;

        futex_up(&idxp->lock);
        return nodep;
}

int fat_destroy_node(void *node)
{
        fat_node_t *nodep = (fat_node_t *)node;
        fat_bs_t *bs;

        /*
         * The node is not reachable from the file system. This means that the
         * link count should be zero and that the index structure cannot be
         * found in the position hash. Obviously, we don't need to lock the node
         * nor its index structure.
         */
        assert(nodep->lnkcnt == 0);

        /*
         * The node may not have any children.
         */
        assert(fat_has_children(node) == false);

        bs = block_bb_get(nodep->idx->dev_handle);
        if (nodep->firstc != FAT_CLST_RES0) {
                assert(nodep->size);
                /* Free all clusters allocated to the node. */
                fat_free_clusters(bs, nodep->idx->dev_handle, nodep->firstc);
        }

        fat_idx_destroy(nodep->idx);
        free(nodep);
        return EOK;
}

int fat_link(void *prnt, void *chld, const char *name)
{
        fat_node_t *parentp = (fat_node_t *)prnt;
        fat_node_t *childp = (fat_node_t *)chld;
        fat_dentry_t *d;
        fat_bs_t *bs;
        block_t *b;
        int i, j;
        uint16_t bps;
        unsigned dps;
        unsigned blocks;
        fat_cluster_t mcl, lcl;
        int rc;

        futex_down(&childp->lock);
        if (childp->lnkcnt == 1) {
                /*
                 * On FAT, we don't support multiple hard links.
                 */
                futex_up(&childp->lock);
                return EMLINK;
        }
        assert(childp->lnkcnt == 0);
        futex_up(&childp->lock);

        if (!fat_dentry_name_verify(name)) {
                /*
                 * Attempt to create unsupported name.
                 */
                return ENOTSUP;
        }

        /*
         * Get us an unused parent node's dentry or grow the parent and allocate
         * a new one.
         */
        
        futex_down(&parentp->idx->lock);
        bs = block_bb_get(parentp->idx->dev_handle);
        bps = uint16_t_le2host(bs->bps);
        dps = bps / sizeof(fat_dentry_t);

        blocks = parentp->size / bps;

        for (i = 0; i < blocks; i++) {
                b = fat_block_get(bs, parentp, i, BLOCK_FLAGS_NONE);
                for (j = 0; j < dps; j++) {
                        d = ((fat_dentry_t *)b->data) + j;
                        switch (fat_classify_dentry(d)) {
                        case FAT_DENTRY_SKIP:
                        case FAT_DENTRY_VALID:
                                /* skipping used and meta entries */
                                continue;
                        case FAT_DENTRY_FREE:
                        case FAT_DENTRY_LAST:
                                /* found an empty slot */
                                goto hit;
                        }
                }
                block_put(b);
        }
        
        /*
         * We need to grow the parent in order to create a new unused dentry.
         */
        if (parentp->idx->pfc == FAT_CLST_ROOT) {
                /* Can't grow the root directory. */
                futex_up(&parentp->idx->lock);
                return ENOSPC;
        }
        rc = fat_alloc_clusters(bs, parentp->idx->dev_handle, 1, &mcl, &lcl);
        if (rc != EOK) {
                futex_up(&parentp->idx->lock);
                return rc;
        }
        fat_append_clusters(bs, parentp, mcl);
        b = fat_block_get(bs, parentp, i, BLOCK_FLAGS_NOREAD);
        d = (fat_dentry_t *)b->data;
        /*
         * Clear all dentries in the block except for the first one (the first
         * dentry will be cleared in the next step).
         */
        memset(d + 1, 0, bps - sizeof(fat_dentry_t));

hit:
        /*
         * At this point we only establish the link between the parent and the
         * child.  The dentry, except of the name and the extension, will remain
         * uninitialized until the corresponding node is synced. Thus the valid
         * dentry data is kept in the child node structure.
         */
        memset(d, 0, sizeof(fat_dentry_t));
        fat_dentry_name_set(d, name);
        b->dirty = true;                /* need to sync block */
        block_put(b);
        futex_up(&parentp->idx->lock);

        futex_down(&childp->idx->lock);
        
        /*
         * If possible, create the Sub-directory Identifier Entry and the
         * Sub-directory Parent Pointer Entry (i.e. "." and ".."). These entries
         * are not mandatory according to Standard ECMA-107 and HelenOS VFS does
         * not use them anyway, so this is rather a sign of our good will.
         */
        b = fat_block_get(bs, childp, 0, BLOCK_FLAGS_NONE);
        d = (fat_dentry_t *)b->data;
        if (fat_classify_dentry(d) == FAT_DENTRY_LAST ||
            strcmp(d->name, FAT_NAME_DOT) == 0) {
                memset(d, 0, sizeof(fat_dentry_t));
                strcpy(d->name, FAT_NAME_DOT);
                strcpy(d->ext, FAT_EXT_PAD);
                d->attr = FAT_ATTR_SUBDIR;
                d->firstc = host2uint16_t_le(childp->firstc);
                /* TODO: initialize also the date/time members. */
        }
        d++;
        if (fat_classify_dentry(d) == FAT_DENTRY_LAST ||
            strcmp(d->name, FAT_NAME_DOT_DOT) == 0) {
                memset(d, 0, sizeof(fat_dentry_t));
                strcpy(d->name, FAT_NAME_DOT_DOT);
                strcpy(d->ext, FAT_EXT_PAD);
                d->attr = FAT_ATTR_SUBDIR;
                d->firstc = (parentp->firstc == FAT_CLST_ROOT) ?
                    host2uint16_t_le(FAT_CLST_RES0) :
                    host2uint16_t_le(parentp->firstc);
                /* TODO: initialize also the date/time members. */
        }
        b->dirty = true;                /* need to sync block */
        block_put(b);

        childp->idx->pfc = parentp->firstc;
        childp->idx->pdi = i * dps + j;
        futex_up(&childp->idx->lock);

        futex_down(&childp->lock);
        childp->lnkcnt = 1;
        childp->dirty = true;           /* need to sync node */
        futex_up(&childp->lock);

        /*
         * Hash in the index structure into the position hash.
         */
        fat_idx_hashin(childp->idx);

        return EOK;
}

int fat_unlink(void *prnt, void *chld)
{
        fat_node_t *parentp = (fat_node_t *)prnt;
        fat_node_t *childp = (fat_node_t *)chld;
        fat_bs_t *bs;
        fat_dentry_t *d;
        uint16_t bps;
        block_t *b;

        futex_down(&parentp->lock);
        futex_down(&childp->lock);
        assert(childp->lnkcnt == 1);
        futex_down(&childp->idx->lock);
        bs = block_bb_get(childp->idx->dev_handle);
        bps = uint16_t_le2host(bs->bps);

        b = _fat_block_get(bs, childp->idx->dev_handle, childp->idx->pfc,
            (childp->idx->pdi * sizeof(fat_dentry_t)) / bps,
            BLOCK_FLAGS_NONE);
        d = (fat_dentry_t *)b->data +
            (childp->idx->pdi % (bps / sizeof(fat_dentry_t)));
        /* mark the dentry as not-currently-used */
        d->name[0] = FAT_DENTRY_ERASED;
        b->dirty = true;                /* need to sync block */
        block_put(b);

        /* remove the index structure from the position hash */
        fat_idx_hashout(childp->idx);
        /* clear position information */
        childp->idx->pfc = FAT_CLST_RES0;
        childp->idx->pdi = 0;
        futex_up(&childp->idx->lock);
        childp->lnkcnt = 0;
        childp->dirty = true;
        futex_up(&childp->lock);
        futex_up(&parentp->lock);

        return EOK;
}

void *fat_match(void *prnt, const char *component)
{
        fat_bs_t *bs;
        fat_node_t *parentp = (fat_node_t *)prnt;
        char name[FAT_NAME_LEN + 1 + FAT_EXT_LEN + 1];
        unsigned i, j;
        unsigned bps;           /* bytes per sector */
        unsigned dps;           /* dentries per sector */
        unsigned blocks;
        fat_dentry_t *d;
        block_t *b;

        futex_down(&parentp->idx->lock);
        bs = block_bb_get(parentp->idx->dev_handle);
        bps = uint16_t_le2host(bs->bps);
        dps = bps / sizeof(fat_dentry_t);
        blocks = parentp->size / bps;
        for (i = 0; i < blocks; i++) {
                b = fat_block_get(bs, parentp, i, BLOCK_FLAGS_NONE);
                for (j = 0; j < dps; j++) { 
                        d = ((fat_dentry_t *)b->data) + j;
                        switch (fat_classify_dentry(d)) {
                        case FAT_DENTRY_SKIP:
                        case FAT_DENTRY_FREE:
                                continue;
                        case FAT_DENTRY_LAST:
                                block_put(b);
                                futex_up(&parentp->idx->lock);
                                return NULL;
                        default:
                        case FAT_DENTRY_VALID:
                                fat_dentry_name_get(d, name);
                                break;
                        }
                        if (fat_dentry_namecmp(name, component) == 0) {
                                /* hit */
                                void *node;
                                /*
                                 * Assume tree hierarchy for locking.  We
                                 * already have the parent and now we are going
                                 * to lock the child.  Never lock in the oposite
                                 * order.
                                 */
                                fat_idx_t *idx = fat_idx_get_by_pos(
                                    parentp->idx->dev_handle, parentp->firstc,
                                    i * dps + j);
                                futex_up(&parentp->idx->lock);
                                if (!idx) {
                                        /*
                                         * Can happen if memory is low or if we
                                         * run out of 32-bit indices.
                                         */
                                        block_put(b);
                                        return NULL;
                                }
                                node = fat_node_get_core(idx);
                                futex_up(&idx->lock);
                                block_put(b);
                                return node;
                        }
                }
                block_put(b);
        }

        futex_up(&parentp->idx->lock);
        return NULL;
}

fs_index_t fat_index_get(void *node)
{
        fat_node_t *fnodep = (fat_node_t *)node;
        if (!fnodep)
                return 0;
        return fnodep->idx->index;
}

size_t fat_size_get(void *node)
{
        return ((fat_node_t *)node)->size;
}

unsigned fat_lnkcnt_get(void *node)
{
        return ((fat_node_t *)node)->lnkcnt;
}

bool fat_has_children(void *node)
{
        fat_bs_t *bs;
        fat_node_t *nodep = (fat_node_t *)node;
        unsigned bps;
        unsigned dps;
        unsigned blocks;
        block_t *b;
        unsigned i, j;

        if (nodep->type != FAT_DIRECTORY)
                return false;
        
        futex_down(&nodep->idx->lock);
        bs = block_bb_get(nodep->idx->dev_handle);
        bps = uint16_t_le2host(bs->bps);
        dps = bps / sizeof(fat_dentry_t);

        blocks = nodep->size / bps;

        for (i = 0; i < blocks; i++) {
                fat_dentry_t *d;
        
                b = fat_block_get(bs, nodep, i, BLOCK_FLAGS_NONE);
                for (j = 0; j < dps; j++) {
                        d = ((fat_dentry_t *)b->data) + j;
                        switch (fat_classify_dentry(d)) {
                        case FAT_DENTRY_SKIP:
                        case FAT_DENTRY_FREE:
                                continue;
                        case FAT_DENTRY_LAST:
                                block_put(b);
                                futex_up(&nodep->idx->lock);
                                return false;
                        default:
                        case FAT_DENTRY_VALID:
                                block_put(b);
                                futex_up(&nodep->idx->lock);
                                return true;
                        }
                        block_put(b);
                        futex_up(&nodep->idx->lock);
                        return true;
                }
                block_put(b);
        }

        futex_up(&nodep->idx->lock);
        return false;
}

void *fat_root_get(dev_handle_t dev_handle)
{
        return fat_node_get(dev_handle, 0);
}

char fat_plb_get_char(unsigned pos)
{
        return fat_reg.plb_ro[pos % PLB_SIZE];
}

bool fat_is_directory(void *node)
{
        return ((fat_node_t *)node)->type == FAT_DIRECTORY;
}

bool fat_is_file(void *node)
{
        return ((fat_node_t *)node)->type == FAT_FILE;
}

/** libfs operations */
libfs_ops_t fat_libfs_ops = {
        .match = fat_match,
        .node_get = fat_node_get,
        .node_put = fat_node_put,
        .create = fat_create_node,
        .destroy = fat_destroy_node,
        .link = fat_link,
        .unlink = fat_unlink,
        .index_get = fat_index_get,
        .size_get = fat_size_get,
        .lnkcnt_get = fat_lnkcnt_get,
        .has_children = fat_has_children,
        .root_get = fat_root_get,
        .plb_get_char = fat_plb_get_char,
        .is_directory = fat_is_directory,
        .is_file = fat_is_file
};

/*
 * VFS operations.
 */

void fat_mounted(ipc_callid_t rid, ipc_call_t *request)
{
        dev_handle_t dev_handle = (dev_handle_t) IPC_GET_ARG1(*request);
        fat_bs_t *bs;
        uint16_t bps;
        uint16_t rde;
        int rc;

        /* initialize libblock */
        rc = block_init(dev_handle, BS_SIZE);
        if (rc != EOK) {
                ipc_answer_0(rid, rc);
                return;
        }

        /* prepare the boot block */
        rc = block_bb_read(dev_handle, BS_BLOCK * BS_SIZE, BS_SIZE);
        if (rc != EOK) {
                block_fini(dev_handle);
                ipc_answer_0(rid, rc);
                return;
        }

        /* get the buffer with the boot sector */
        bs = block_bb_get(dev_handle);
        
        /* Read the number of root directory entries. */
        bps = uint16_t_le2host(bs->bps);
        rde = uint16_t_le2host(bs->root_ent_max);

        if (bps != BS_SIZE) {
                block_fini(dev_handle);
                ipc_answer_0(rid, ENOTSUP);
                return;
        }

        /* Initialize the block cache */
        rc = block_cache_init(dev_handle, bps, 0 /* XXX */);
        if (rc != EOK) {
                block_fini(dev_handle);
                ipc_answer_0(rid, rc);
                return;
        }

        rc = fat_idx_init_by_dev_handle(dev_handle);
        if (rc != EOK) {
                block_fini(dev_handle);
                ipc_answer_0(rid, rc);
                return;
        }

        /* Initialize the root node. */
        fat_node_t *rootp = (fat_node_t *)malloc(sizeof(fat_node_t));
        if (!rootp) {
                block_fini(dev_handle);
                fat_idx_fini_by_dev_handle(dev_handle);
                ipc_answer_0(rid, ENOMEM);
                return;
        }
        fat_node_initialize(rootp);

        fat_idx_t *ridxp = fat_idx_get_by_pos(dev_handle, FAT_CLST_ROOTPAR, 0);
        if (!ridxp) {
                block_fini(dev_handle);
                free(rootp);
                fat_idx_fini_by_dev_handle(dev_handle);
                ipc_answer_0(rid, ENOMEM);
                return;
        }
        assert(ridxp->index == 0);
        /* ridxp->lock held */

        rootp->type = FAT_DIRECTORY;
        rootp->firstc = FAT_CLST_ROOT;
        rootp->refcnt = 1;
        rootp->lnkcnt = 0;      /* FS root is not linked */
        rootp->size = rde * sizeof(fat_dentry_t);
        rootp->idx = ridxp;
        ridxp->nodep = rootp;
        
        futex_up(&ridxp->lock);

        ipc_answer_3(rid, EOK, ridxp->index, rootp->size, rootp->lnkcnt);
}

void fat_mount(ipc_callid_t rid, ipc_call_t *request)
{
        ipc_answer_0(rid, ENOTSUP);
}

void fat_lookup(ipc_callid_t rid, ipc_call_t *request)
{
        libfs_lookup(&fat_libfs_ops, fat_reg.fs_handle, rid, request);
}

void fat_read(ipc_callid_t rid, ipc_call_t *request)
{
        dev_handle_t dev_handle = (dev_handle_t)IPC_GET_ARG1(*request);
        fs_index_t index = (fs_index_t)IPC_GET_ARG2(*request);
        off_t pos = (off_t)IPC_GET_ARG3(*request);
        fat_node_t *nodep = (fat_node_t *)fat_node_get(dev_handle, index);
        fat_bs_t *bs;
        uint16_t bps;
        size_t bytes;
        block_t *b;

        if (!nodep) {
                ipc_answer_0(rid, ENOENT);
                return;
        }

        ipc_callid_t callid;
        size_t len;
        if (!ipc_data_read_receive(&callid, &len)) {
                fat_node_put(nodep);
                ipc_answer_0(callid, EINVAL);
                ipc_answer_0(rid, EINVAL);
                return;
        }

        bs = block_bb_get(dev_handle);
        bps = uint16_t_le2host(bs->bps);

        if (nodep->type == FAT_FILE) {
                /*
                 * Our strategy for regular file reads is to read one block at
                 * most and make use of the possibility to return less data than
                 * requested. This keeps the code very simple.
                 */
                if (pos >= nodep->size) {
                        /* reading beyond the EOF */
                        bytes = 0;
                        (void) ipc_data_read_finalize(callid, NULL, 0);
                } else {
                        bytes = min(len, bps - pos % bps);
                        bytes = min(bytes, nodep->size - pos);
                        b = fat_block_get(bs, nodep, pos / bps,
                            BLOCK_FLAGS_NONE);
                        (void) ipc_data_read_finalize(callid, b->data + pos % bps,
                            bytes);
                        block_put(b);
                }
        } else {
                unsigned bnum;
                off_t spos = pos;
                char name[FAT_NAME_LEN + 1 + FAT_EXT_LEN + 1];
                fat_dentry_t *d;

                assert(nodep->type == FAT_DIRECTORY);
                assert(nodep->size % bps == 0);
                assert(bps % sizeof(fat_dentry_t) == 0);

                /*
                 * Our strategy for readdir() is to use the position pointer as
                 * an index into the array of all dentries. On entry, it points
                 * to the first unread dentry. If we skip any dentries, we bump
                 * the position pointer accordingly.
                 */
                bnum = (pos * sizeof(fat_dentry_t)) / bps;
                while (bnum < nodep->size / bps) {
                        off_t o;

                        b = fat_block_get(bs, nodep, bnum, BLOCK_FLAGS_NONE);
                        for (o = pos % (bps / sizeof(fat_dentry_t));
                            o < bps / sizeof(fat_dentry_t);
                            o++, pos++) {
                                d = ((fat_dentry_t *)b->data) + o;
                                switch (fat_classify_dentry(d)) {
                                case FAT_DENTRY_SKIP:
                                case FAT_DENTRY_FREE:
                                        continue;
                                case FAT_DENTRY_LAST:
                                        block_put(b);
                                        goto miss;
                                default:
                                case FAT_DENTRY_VALID:
                                        fat_dentry_name_get(d, name);
                                        block_put(b);
                                        goto hit;
                                }
                        }
                        block_put(b);
                        bnum++;
                }
miss:
                fat_node_put(nodep);
                ipc_answer_0(callid, ENOENT);
                ipc_answer_1(rid, ENOENT, 0);
                return;
hit:
                (void) ipc_data_read_finalize(callid, name, strlen(name) + 1);
                bytes = (pos - spos) + 1;
        }

        fat_node_put(nodep);
        ipc_answer_1(rid, EOK, (ipcarg_t)bytes);
}

void fat_write(ipc_callid_t rid, ipc_call_t *request)
{
        dev_handle_t dev_handle = (dev_handle_t)IPC_GET_ARG1(*request);
        fs_index_t index = (fs_index_t)IPC_GET_ARG2(*request);
        off_t pos = (off_t)IPC_GET_ARG3(*request);
        fat_node_t *nodep = (fat_node_t *)fat_node_get(dev_handle, index);
        fat_bs_t *bs;
        size_t bytes;
        block_t *b;
        uint16_t bps;
        unsigned spc;
        unsigned bpc;           /* bytes per cluster */
        off_t boundary;
        int flags = BLOCK_FLAGS_NONE;
        
        if (!nodep) {
                ipc_answer_0(rid, ENOENT);
                return;
        }
        
        ipc_callid_t callid;
        size_t len;
        if (!ipc_data_write_receive(&callid, &len)) {
                fat_node_put(nodep);
                ipc_answer_0(callid, EINVAL);
                ipc_answer_0(rid, EINVAL);
                return;
        }

        bs = block_bb_get(dev_handle);
        bps = uint16_t_le2host(bs->bps);
        spc = bs->spc;
        bpc = bps * spc;

        /*
         * In all scenarios, we will attempt to write out only one block worth
         * of data at maximum. There might be some more efficient approaches,
         * but this one greatly simplifies fat_write(). Note that we can afford
         * to do this because the client must be ready to handle the return
         * value signalizing a smaller number of bytes written. 
         */ 
        bytes = min(len, bps - pos % bps);
        if (bytes == bps)
                flags |= BLOCK_FLAGS_NOREAD;
        
        boundary = ROUND_UP(nodep->size, bpc);
        if (pos < boundary) {
                /*
                 * This is the easier case - we are either overwriting already
                 * existing contents or writing behind the EOF, but still within
                 * the limits of the last cluster. The node size may grow to the
                 * next block size boundary.
                 */
                fat_fill_gap(bs, nodep, FAT_CLST_RES0, pos);
                b = fat_block_get(bs, nodep, pos / bps, flags);
                (void) ipc_data_write_finalize(callid, b->data + pos % bps,
                    bytes);
                b->dirty = true;                /* need to sync block */
                block_put(b);
                if (pos + bytes > nodep->size) {
                        nodep->size = pos + bytes;
                        nodep->dirty = true;    /* need to sync node */
                }
                ipc_answer_2(rid, EOK, bytes, nodep->size);     
                fat_node_put(nodep);
                return;
        } else {
                /*
                 * This is the more difficult case. We must allocate new
                 * clusters for the node and zero them out.
                 */
                int status;
                unsigned nclsts;
                fat_cluster_t mcl, lcl; 
 
                nclsts = (ROUND_UP(pos + bytes, bpc) - boundary) / bpc;
                /* create an independent chain of nclsts clusters in all FATs */
                status = fat_alloc_clusters(bs, dev_handle, nclsts, &mcl, &lcl);
                if (status != EOK) {
                        /* could not allocate a chain of nclsts clusters */
                        fat_node_put(nodep);
                        ipc_answer_0(callid, status);
                        ipc_answer_0(rid, status);
                        return;
                }
                /* zero fill any gaps */
                fat_fill_gap(bs, nodep, mcl, pos);
                b = _fat_block_get(bs, dev_handle, lcl, (pos / bps) % spc,
                    flags);
                (void) ipc_data_write_finalize(callid, b->data + pos % bps,
                    bytes);
                b->dirty = true;                /* need to sync block */
                block_put(b);
                /*
                 * Append the cluster chain starting in mcl to the end of the
                 * node's cluster chain.
                 */
                fat_append_clusters(bs, nodep, mcl);
                nodep->size = pos + bytes;
                nodep->dirty = true;            /* need to sync node */
                ipc_answer_2(rid, EOK, bytes, nodep->size);
                fat_node_put(nodep);
                return;
        }
}

void fat_truncate(ipc_callid_t rid, ipc_call_t *request)
{
        dev_handle_t dev_handle = (dev_handle_t)IPC_GET_ARG1(*request);
        fs_index_t index = (fs_index_t)IPC_GET_ARG2(*request);
        size_t size = (off_t)IPC_GET_ARG3(*request);
        fat_node_t *nodep = (fat_node_t *)fat_node_get(dev_handle, index);
        fat_bs_t *bs;
        uint16_t bps;
        uint8_t spc;
        unsigned bpc;   /* bytes per cluster */
        int rc;

        if (!nodep) {
                ipc_answer_0(rid, ENOENT);
                return;
        }

        bs = block_bb_get(dev_handle);
        bps = uint16_t_le2host(bs->bps);
        spc = bs->spc;
        bpc = bps * spc;

        if (nodep->size == size) {
                rc = EOK;
        } else if (nodep->size < size) {
                /*
                 * The standard says we have the freedom to grow the node.
                 * For now, we simply return an error.
                 */
                rc = EINVAL;
        } else if (ROUND_UP(nodep->size, bpc) == ROUND_UP(size, bpc)) {
                /*
                 * The node will be shrunk, but no clusters will be deallocated.
                 */
                nodep->size = size;
                nodep->dirty = true;            /* need to sync node */
                rc = EOK;       
        } else {
                /*
                 * The node will be shrunk, clusters will be deallocated.
                 */
                if (size == 0) {
                        fat_chop_clusters(bs, nodep, FAT_CLST_RES0);
                } else {
                        fat_cluster_t lastc;
                        (void) fat_cluster_walk(bs, dev_handle, nodep->firstc,
                            &lastc, (size - 1) / bpc);
                        fat_chop_clusters(bs, nodep, lastc);
                }
                nodep->size = size;
                nodep->dirty = true;            /* need to sync node */
                rc = EOK;       
        }
        fat_node_put(nodep);
        ipc_answer_0(rid, rc);
        return;
}

void fat_destroy(ipc_callid_t rid, ipc_call_t *request)
{
        dev_handle_t dev_handle = (dev_handle_t)IPC_GET_ARG1(*request);
        fs_index_t index = (fs_index_t)IPC_GET_ARG2(*request);
        int rc;

        fat_node_t *nodep = fat_node_get(dev_handle, index);
        if (!nodep) {
                ipc_answer_0(rid, ENOENT);
                return;
        }

        rc = fat_destroy_node(nodep);
        ipc_answer_0(rid, rc);
}

/**
 * @}
 */