HelenOS sources

/* [<][>][^][v][top][bottom][index][help] */
This source file includes following definitions.
main
insert_dentries
init_inode_table
make_root_ino
make_root_ino2
init_superblock
write_superblock
write_superblock3
init_bitmaps
mark_bmap
write_block
help_cmd_mkmfs
is_power_of_two
/*
 * Copyright (c) 2024 Jiri Svoboda
 * Copyright (c) 2011 Maurizio Lombardi
 * 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 mkmfs
 * @{
 */

/**
 * @file        mkmfs.c
 * @brief       Tool for creating new Minix file systems.
 *
 */

#include <stdio.h>
#include <stdlib.h>
#include <block.h>
#include <errno.h>
#include <inttypes.h>
#include <getopt.h>
#include <mem.h>
#include <str.h>
#include <time.h>
#include <minix.h>

#define NAME    "mkmfs"

#define FREE    0
#define USED    1

#define UPPER(n, size)          (((n) / (size)) + (((n) % (size)) != 0))
#define NEXT_DENTRY(p, dirsize) (p += (dirsize))

typedef enum {
        HELP_SHORT,
        HELP_LONG
} help_level_t;

/* Generic MFS superblock */
struct mfs_sb_info {
        uint64_t n_inodes;
        uint64_t n_zones;
        aoff64_t dev_nblocks;
        unsigned long ibmap_blocks;
        unsigned long zbmap_blocks;
        unsigned long first_data_zone;
        unsigned long itable_size;
        int log2_zone_size;
        int ino_per_block;
        int dirsize;
        uint32_t max_file_size;
        uint16_t magic;
        uint32_t block_size;
        int fs_version;
        bool longnames;
};

static void     help_cmd_mkmfs(help_level_t level);
static bool     is_power_of_two(uint32_t n);
static errno_t  init_superblock(struct mfs_sb_info *sb);
static errno_t  write_superblock(const struct mfs_sb_info *sbi);
static errno_t  write_superblock3(const struct mfs_sb_info *sbi);
static errno_t  init_bitmaps(const struct mfs_sb_info *sb);
static errno_t  init_inode_table(const struct mfs_sb_info *sb);
static errno_t  make_root_ino(const struct mfs_sb_info *sb);
static errno_t  make_root_ino2(const struct mfs_sb_info *sb);
static void     mark_bmap(uint32_t *bmap, int idx, int v);
static errno_t  insert_dentries(const struct mfs_sb_info *sb);

static inline errno_t write_block(aoff64_t off, size_t size, const void *data);

static service_id_t service_id;
static int shift;

static struct option const long_options[] = {
        { "help", no_argument, 0, 'h' },
        { "long-names", no_argument, 0, 'l' },
        { "block-size", required_argument, 0, 'b' },
        { "inodes", required_argument, 0, 'i' },
        { NULL, no_argument, 0, '1' },
        { NULL, no_argument, 0, '2' },
        { 0, 0, 0, 0 }
};

int main (int argc, char **argv)
{
        errno_t rc;
        int c, opt_ind;
        char *device_name;
        size_t devblock_size;

        struct mfs_sb_info sb;

        /* Default is MinixFS V3 */
        sb.magic = MFS_MAGIC_V3;
        sb.fs_version = 3;

        /* Default block size is 4Kb */
        sb.block_size = MFS_MAX_BLOCKSIZE;
        sb.dirsize = MFS3_DIRSIZE;
        sb.n_inodes = 0;
        sb.longnames = false;
        sb.ino_per_block = V3_INODES_PER_BLOCK(MFS_MAX_BLOCKSIZE);

        if (argc == 1) {
                help_cmd_mkmfs(HELP_SHORT);
                printf("Incorrect number of arguments, try `mkmfs --help'\n");
                exit(0);
        }

        c = 0;
        optind = 0;
        opt_ind = 0;
        while (c != -1) {
                c = getopt_long(argc, argv, "lh12b:i:",
                    long_options, &opt_ind);
                switch (c) {
                case 'h':
                        help_cmd_mkmfs(HELP_LONG);
                        exit(0);
                case '1':
                        sb.magic = MFS_MAGIC_V1;
                        sb.block_size = MFS_BLOCKSIZE;
                        sb.fs_version = 1;
                        sb.ino_per_block = V1_INODES_PER_BLOCK;
                        sb.dirsize = MFS_DIRSIZE;
                        break;
                case '2':
                        sb.magic = MFS_MAGIC_V2;
                        sb.block_size = MFS_BLOCKSIZE;
                        sb.fs_version = 2;
                        sb.ino_per_block = V2_INODES_PER_BLOCK;
                        sb.dirsize = MFS_DIRSIZE;
                        break;
                case 'b':
                        sb.block_size = (uint32_t) strtol(optarg, NULL, 10);
                        break;
                case 'i':
                        sb.n_inodes = (uint64_t) strtol(optarg, NULL, 10);
                        break;
                case 'l':
                        sb.longnames = true;
                        sb.dirsize = MFSL_DIRSIZE;
                        break;
                }
        }

        if (sb.block_size < MFS_MIN_BLOCKSIZE ||
            sb.block_size > MFS_MAX_BLOCKSIZE) {
                printf(NAME ":Error! Invalid block size.\n");
                exit(0);
        } else if (!is_power_of_two(sb.block_size)) {
                /* Block size must be a power of 2. */
                printf(NAME ":Error! Invalid block size.\n");
                exit(0);
        } else if (sb.block_size > MFS_BLOCKSIZE &&
            sb.fs_version != 3) {
                printf(NAME ":Error! Block size > 1024 is "
                    "supported by V3 filesystem only.\n");
                exit(0);
        } else if (sb.fs_version == 3 && sb.longnames) {
                printf(NAME ":Error! Long filenames are supported "
                    "by V1/V2 filesystem only.\n");
                exit(0);
        }

        if (sb.block_size == MFS_MIN_BLOCKSIZE)
                shift = 1;
        else if (sb.block_size == MFS_MAX_BLOCKSIZE)
                shift = 3;
        else
                shift = 2;

        argv += optind;

        device_name = argv[0];

        if (!device_name) {
                help_cmd_mkmfs(HELP_LONG);
                exit(0);
        }

        rc = loc_service_get_id(device_name, &service_id, 0);
        if (rc != EOK) {
                printf(NAME ": Error resolving device `%s'.\n", device_name);
                return 2;
        }

        rc = block_init(service_id);
        if (rc != EOK)  {
                printf(NAME ": Error initializing libblock.\n");
                return 2;
        }

        rc = block_get_bsize(service_id, &devblock_size);
        if (rc != EOK) {
                printf(NAME ": Error determining device block size.\n");
                return 2;
        }

        rc = block_get_nblocks(service_id, &sb.dev_nblocks);
        if (rc != EOK) {
                printf(NAME ": Warning, failed to obtain "
                    "block device size.\n");
        } else {
                printf(NAME ": Block device has %" PRIuOFF64 " blocks.\n",
                    sb.dev_nblocks);
        }

        if (devblock_size != 512) {
                printf(NAME ": Error. Device block size is not 512 bytes.\n");
                return 2;
        }

        /* Minimum block size is 1 Kb */
        sb.dev_nblocks /= 2;

        printf(NAME ": Creating Minix file system on device\n");
        printf(NAME ": Writing superblock\n");

        /* Initialize superblock */
        if (init_superblock(&sb) != EOK) {
                printf(NAME ": Error. Superblock initialization failed\n");
                return 2;
        }

        printf(NAME ": Initializing bitmaps\n");

        /* Initialize bitmaps */
        if (init_bitmaps(&sb) != EOK) {
                printf(NAME ": Error. Bitmaps initialization failed\n");
                return 2;
        }

        printf(NAME ": Initializing the inode table\n");

        /* Init inode table */
        if (init_inode_table(&sb) != EOK) {
                printf(NAME ": Error. Inode table initialization failed\n");
                return 2;
        }

        printf(NAME ": Creating the root directory inode\n");

        /* Make the root inode */
        if (sb.fs_version == 1)
                rc = make_root_ino(&sb);
        else
                rc = make_root_ino2(&sb);

        if (rc != EOK) {
                printf(NAME ": Error. Root inode initialization failed\n");
                return 2;
        }

        /* Insert directory entries . and .. */
        if (insert_dentries(&sb) != EOK) {
                printf(NAME ": Error. Root directory initialization failed\n");
                return 2;
        }

        block_fini(service_id);

        return 0;
}

/**Inserts the '.' and '..' directory entries in the root directory.
 *
 * @param sb            Pointer to the superblock structure.
 *
 * @return              EOK on success or an error code.
 */
static errno_t insert_dentries(const struct mfs_sb_info *sb)
{
        void *root_block;
        uint8_t *dentry_ptr;
        errno_t rc;
        const long root_dblock = sb->first_data_zone;

        root_block = malloc(sb->block_size);
        memset(root_block, 0x00, sb->block_size);

        if (!root_block)
                return ENOMEM;

        dentry_ptr = root_block;

        if (sb->fs_version != 3) {
                /* Directory entries for V1/V2 filesystem */
                struct mfs_dentry *dentry = root_block;

                dentry->d_inum = MFS_ROOT_INO;
                memcpy(dentry->d_name, ".\0", 2);

                dentry = (struct mfs_dentry *) NEXT_DENTRY(dentry_ptr,
                    sb->dirsize);

                dentry->d_inum = MFS_ROOT_INO;
                memcpy(dentry->d_name, "..\0", 3);
        } else {
                /* Directory entries for V3 filesystem */
                struct mfs3_dentry *dentry = root_block;

                dentry->d_inum = MFS_ROOT_INO;
                memcpy(dentry->d_name, ".\0", 2);

                dentry = (struct mfs3_dentry *) NEXT_DENTRY(dentry_ptr,
                    sb->dirsize);

                dentry->d_inum = MFS_ROOT_INO;
                memcpy(dentry->d_name, "..\0", 3);
        }

        rc = write_block(root_dblock, 1, root_block);

        free(root_block);
        return rc;
}

/**Initialize the inode table.
 *
 * @param sb            Pointer to the superblock structure.
 *
 * @return              EOK on success or an error code.
 */
static errno_t init_inode_table(const struct mfs_sb_info *sb)
{
        unsigned int i;
        uint8_t *itable_buf;
        errno_t rc = EOK;

        long itable_off = sb->zbmap_blocks + sb->ibmap_blocks + 2;
        unsigned long itable_size = sb->itable_size;

        itable_buf = malloc(sb->block_size);

        if (!itable_buf)
                return ENOMEM;

        memset(itable_buf, 0x00, sb->block_size);

        for (i = 0; i < itable_size; ++i, ++itable_off) {
                rc = write_block(itable_off, 1, itable_buf);

                if (rc != EOK)
                        break;
        }

        free(itable_buf);
        return rc;
}

/**Initialize a V1 root inode.
 *
 * @param sb            Ponter to the superblock structure.
 *
 * @return              EOK on success or an error code.
 */
static errno_t make_root_ino(const struct mfs_sb_info *sb)
{
        struct mfs_inode *ino_buf;
        errno_t rc;

        const long itable_off = sb->zbmap_blocks + sb->ibmap_blocks + 2;

        const time_t sec = time(NULL);

        ino_buf = malloc(MFS_BLOCKSIZE);

        if (!ino_buf)
                return ENOMEM;

        memset(ino_buf, 0x00, MFS_BLOCKSIZE);

        ino_buf[MFS_ROOT_INO - 1].i_mode = S_IFDIR;
        ino_buf[MFS_ROOT_INO - 1].i_uid = 0;
        ino_buf[MFS_ROOT_INO - 1].i_gid = 0;
        ino_buf[MFS_ROOT_INO - 1].i_size = (sb->longnames ? MFSL_DIRSIZE :
            MFS_DIRSIZE) * 2;
        ino_buf[MFS_ROOT_INO - 1].i_mtime = sec;
        ino_buf[MFS_ROOT_INO - 1].i_nlinks = 2;
        ino_buf[MFS_ROOT_INO - 1].i_dzone[0] = sb->first_data_zone;

        rc = write_block(itable_off, 1, ino_buf);

        free(ino_buf);
        return rc;
}

/**Initialize a Minix V2 root inode on disk, also valid for V3 filesystem.
 *
 * @param sb            Pointer to the superblock structure.
 *
 * @return              EOK on success or an error code.
 */
static errno_t make_root_ino2(const struct mfs_sb_info *sb)
{
        struct mfs2_inode *ino_buf;
        errno_t rc;

        /* Compute offset of the first inode table block */
        const long itable_off = sb->zbmap_blocks + sb->ibmap_blocks + 2;

        const time_t sec = time(NULL);

        ino_buf = malloc(sb->block_size);

        if (!ino_buf)
                return ENOMEM;

        memset(ino_buf, 0x00, sb->block_size);

        ino_buf[MFS_ROOT_INO - 1].i_mode = S_IFDIR;
        ino_buf[MFS_ROOT_INO - 1].i_uid = 0;
        ino_buf[MFS_ROOT_INO - 1].i_gid = 0;
        ino_buf[MFS_ROOT_INO - 1].i_size = MFS3_DIRSIZE * 2;
        ino_buf[MFS_ROOT_INO - 1].i_mtime = sec;
        ino_buf[MFS_ROOT_INO - 1].i_atime = sec;
        ino_buf[MFS_ROOT_INO - 1].i_ctime = sec;
        ino_buf[MFS_ROOT_INO - 1].i_nlinks = 2;
        ino_buf[MFS_ROOT_INO - 1].i_dzone[0] = sb->first_data_zone;

        rc = write_block(itable_off, 1, ino_buf);

        free(ino_buf);
        return rc;
}

/**Initialize the superblock structure on disk.
 *
 * @param sb            Pointer to the superblock structure.
 *
 * @return              EOK on success or an error code.
 */
static errno_t init_superblock(struct mfs_sb_info *sb)
{
        aoff64_t inodes;
        unsigned long ind;
        unsigned long ind2;
        unsigned long zones;
        errno_t rc;

        if (sb->longnames)
                sb->magic = sb->fs_version == 1 ? MFS_MAGIC_V1L :
                    MFS_MAGIC_V2L;

        /* Compute the number of zones on disk */

        if (sb->fs_version == 1) {
                /* Valid only for MFS V1 */
                sb->n_zones = sb->dev_nblocks > UINT16_MAX ?
                    UINT16_MAX : sb->dev_nblocks;
                ind = MFS_BLOCKSIZE / sizeof(uint16_t);
                ind2 = ind * ind;
                sb->max_file_size = (V1_NR_DIRECT_ZONES + ind + ind2) *
                    MFS_BLOCKSIZE;
        } else {
                /* Valid for MFS V2/V3 */
                size_t ptrsize;
                if (sb->fs_version == 2)
                        ptrsize = sizeof(uint16_t);
                else
                        ptrsize = sizeof(uint32_t);

                ind = sb->block_size / ptrsize;
                ind2 = ind * ind;
                zones = V2_NR_DIRECT_ZONES + ind + ind2;
                sb->max_file_size = zones * sb->block_size;
                sb->n_zones = sb->dev_nblocks > UINT32_MAX ?
                    UINT32_MAX : sb->dev_nblocks;

                if (sb->fs_version == 3) {
                        if (INT32_MAX / sb->block_size < zones)
                                sb->max_file_size = INT32_MAX;
                        sb->ino_per_block = V3_INODES_PER_BLOCK(sb->block_size);
                        sb->n_zones /= (sb->block_size / MFS_MIN_BLOCKSIZE);
                }
        }

        /* Round up the number of inodes to fill block size */
        if (sb->n_inodes == 0)
                inodes = sb->dev_nblocks / 3;
        else
                inodes = sb->n_inodes;

        if (inodes % sb->ino_per_block)
                inodes = ((inodes / sb->ino_per_block) + 1) *
                    sb->ino_per_block;

        if (sb->fs_version < 3)
                sb->n_inodes = inodes > UINT16_MAX ? UINT16_MAX : inodes;
        else
                sb->n_inodes = inodes > UINT32_MAX ? UINT32_MAX : inodes;

        /* Compute inode bitmap size in blocks */
        sb->ibmap_blocks = UPPER(sb->n_inodes, sb->block_size * 8);

        /* Compute inode table size */
        sb->itable_size = sb->n_inodes / sb->ino_per_block;

        /* Compute zone bitmap size in blocks */
        sb->zbmap_blocks = UPPER(sb->n_zones, sb->block_size * 8);

        /* Compute first data zone position */
        sb->first_data_zone = 2 + sb->itable_size +
            sb->zbmap_blocks + sb->ibmap_blocks;

        /* Set log2 of zone to block ratio to zero */
        sb->log2_zone_size = 0;

        /* Check for errors */
        if (sb->first_data_zone >= sb->n_zones) {
                printf(NAME ": Error! Insufficient disk space");
                return ENOMEM;
        }

        /* Superblock is now ready to be written on disk */
        printf(NAME ": %d block size\n", sb->block_size);
        printf(NAME ": %d inodes\n", (uint32_t) sb->n_inodes);
        printf(NAME ": %d zones\n", (uint32_t) sb->n_zones);
        printf(NAME ": inode table blocks = %ld\n", sb->itable_size);
        printf(NAME ": inode bitmap blocks = %ld\n", sb->ibmap_blocks);
        printf(NAME ": zone bitmap blocks = %ld\n", sb->zbmap_blocks);
        printf(NAME ": first data zone = %d\n", (uint32_t)sb->first_data_zone);
        printf(NAME ": max file size = %u\n", sb->max_file_size);
        printf(NAME ": long fnames = %s\n", sb->longnames ? "Yes" : "No");

        if (sb->fs_version == 3)
                rc = write_superblock3(sb);
        else
                rc = write_superblock(sb);

        return rc;
}

/**Write the V1/V2 superblock on disk.
 *
 * @param sbi           Pointer to the superblock structure to write on disk.
 *
 * @return              EOK on success or an error code.
 */
static errno_t write_superblock(const struct mfs_sb_info *sbi)
{
        struct mfs_superblock *sb;
        errno_t rc;

        sb = malloc(MFS_SUPERBLOCK_SIZE);

        if (!sb)
                return ENOMEM;

        sb->s_ninodes = (uint16_t) sbi->n_inodes;
        sb->s_nzones = (uint16_t) sbi->n_zones;
        sb->s_nzones2 = (uint32_t) sbi->n_zones;
        sb->s_ibmap_blocks = (uint16_t) sbi->ibmap_blocks;
        sb->s_zbmap_blocks = (uint16_t) sbi->zbmap_blocks;
        sb->s_first_data_zone = (uint16_t) sbi->first_data_zone;
        sb->s_log2_zone_size = sbi->log2_zone_size;
        sb->s_max_file_size = sbi->max_file_size;
        sb->s_magic = sbi->magic;
        sb->s_state = MFS_VALID_FS;

        rc = write_block(MFS_SUPERBLOCK, 1, sb);
        free(sb);

        return rc;
}

/**Write the V3s superblock on disk.
 *
 * @param sbi           Pointer to the superblock structure to write on disk.
 *
 * @return              EOK on success or an error code.
 */
static errno_t write_superblock3(const struct mfs_sb_info *sbi)
{
        struct mfs3_superblock *sb;
        errno_t rc;

        sb = malloc(MFS_SUPERBLOCK_SIZE);

        if (!sb)
                return ENOMEM;

        sb->s_ninodes = (uint32_t) sbi->n_inodes;
        sb->s_nzones = (uint32_t) sbi->n_zones;
        sb->s_ibmap_blocks = (uint16_t) sbi->ibmap_blocks;
        sb->s_zbmap_blocks = (uint16_t) sbi->zbmap_blocks;
        sb->s_first_data_zone = (uint16_t) sbi->first_data_zone;
        sb->s_log2_zone_size = sbi->log2_zone_size;
        sb->s_max_file_size = sbi->max_file_size;
        sb->s_magic = sbi->magic;
        sb->s_block_size = sbi->block_size;
        sb->s_disk_version = 3;

        rc = block_write_direct(service_id, MFS_SUPERBLOCK << 1, 1 << 1, sb);
        free(sb);

        return rc;
}

/**Initialize the inode and block bitmaps on disk.
 *
 * @param sb            Pointer to the superblock structure.
 *
 * @return              EOK on success or an error code.
 */
static errno_t init_bitmaps(const struct mfs_sb_info *sb)
{
        uint32_t *ibmap_buf, *zbmap_buf;
        uint8_t *ibmap_buf8, *zbmap_buf8;
        const unsigned int ibmap_nblocks = sb->ibmap_blocks;
        const unsigned int zbmap_nblocks = sb->zbmap_blocks;
        unsigned int i;
        errno_t rc = EOK;

        ibmap_buf = malloc(ibmap_nblocks * sb->block_size);
        zbmap_buf = malloc(zbmap_nblocks * sb->block_size);

        if (!ibmap_buf || !zbmap_buf) {
                rc = ENOMEM;
                goto exit;
        }

        memset(ibmap_buf, 0xFF, ibmap_nblocks * sb->block_size);
        memset(zbmap_buf, 0xFF, zbmap_nblocks * sb->block_size);

        for (i = 2; i < sb->n_inodes + 1; ++i)
                mark_bmap(ibmap_buf, i, FREE);

        for (i = 2; i < sb->n_zones - sb->first_data_zone; ++i)
                mark_bmap(zbmap_buf, i, FREE);

        ibmap_buf8 = (uint8_t *) ibmap_buf;
        zbmap_buf8 = (uint8_t *) zbmap_buf;

        int start_block = 2;

        for (i = 0; i < ibmap_nblocks; ++i) {
                if ((rc = write_block(start_block + i,
                    1, (ibmap_buf8 + i * sb->block_size))) != EOK)
                        goto exit;
        }

        start_block = 2 + ibmap_nblocks;

        for (i = 0; i < zbmap_nblocks; ++i) {
                if ((rc = write_block(start_block + i,
                    1, (zbmap_buf8 + i * sb->block_size))) != EOK)
                        goto exit;
        }

exit:
        free(ibmap_buf);
        free(zbmap_buf);

        return rc;
}

/**Mark a bitmap entry as used or free.
 *
 * @param bmap          32-bit pointer to the bitmap in memory.
 * @param idx           The index in the bitmap of the bit to set at 1 or 0.
 * @param v             FREE to clear the bit, USED to set the bit.
 */
static void mark_bmap(uint32_t *bmap, int idx, int v)
{
        if (v == FREE)
                bmap[idx / 32] &= ~(1 << (idx % 32));
        else
                bmap[idx / 32] |= 1 << (idx % 32);
}

/**Write a block on disk.
 *
 * @param off           64-bit block offset on disk.
 * @param size          size of the block.
 * @param data          Pointer to the block content.
 *
 * @return              EOK on success or a error number.
 */
static inline errno_t write_block(aoff64_t off, size_t size, const void *data)
{
        if (shift == 3) {
                errno_t rc;
                aoff64_t tmp_off = off << 1;
                uint8_t *data_ptr = (uint8_t *) data;

                rc = block_write_direct(service_id, tmp_off << 2,
                    size << 2, data_ptr);

                if (rc != EOK)
                        return rc;

                data_ptr += 2048;
                tmp_off++;

                return block_write_direct(service_id, tmp_off << 2,
                    size << 2, data_ptr);
        }
        return block_write_direct(service_id, off << shift,
            size << shift, data);
}

static void help_cmd_mkmfs(help_level_t level)
{
        if (level == HELP_SHORT) {
                printf(NAME ": tool to create new Minix file systems\n");
        } else {
                printf("Usage: [options] device\n"
                    "-1         Make a Minix version 1 filesystem\n"
                    "-2         Make a Minix version 2 filesystem\n"
                    "-b ##      Specify the block size in bytes (V3 only),\n"
                    "           valid block size values are 1024, 2048 and"
                    /* ...   */ " 4096 bytes per block\n"
                    "-i ##      Specify the number of inodes"
                    /* ...   */ " for the filesystem\n"
                    "-l         Use 30-char long filenames (V1/V2 only)\n");
        }
}

/** Check if a given number is a power of two.
 *
 * @param n     The number to check.
 *
 * @return      true if it is a power of two, false otherwise.
 */
static bool is_power_of_two(uint32_t n)
{
        if (n == 0)
                return false;

        return (n & (n - 1)) == 0;
}

/**
 * @}
 */
/* [<][>][^][v][top][bottom][index][help] */
HelenOS homepage, sources at GitHub
HelenOS sources

root/uspace/app/mkmfs/mkmfs.c

DEFINITIONS
