aboutsummaryrefslogtreecommitdiff
path: root/toys/other
diff options
context:
space:
mode:
authorRob Landley <rob@landley.net>2013-02-22 11:22:01 -0600
committerRob Landley <rob@landley.net>2013-02-22 11:22:01 -0600
commitdd009d622ad7b6094efe073f0974a5e3a523cc32 (patch)
tree52856e68356c848845a6c54c60322d146a4c69f4 /toys/other
parent865afc55d04ede07db830c767bec75fb67c3ff6f (diff)
downloadtoybox-dd009d622ad7b6094efe073f0974a5e3a523cc32.tar.gz
Move some unfinished commands to the "pending" directory.
Diffstat (limited to 'toys/other')
-rw-r--r--toys/other/mke2fs.c656
1 files changed, 0 insertions, 656 deletions
diff --git a/toys/other/mke2fs.c b/toys/other/mke2fs.c
deleted file mode 100644
index 1a83e4f6..00000000
--- a/toys/other/mke2fs.c
+++ /dev/null
@@ -1,656 +0,0 @@
-/* mke2fs.c - Create an ext2 filesystem image.
- *
- * Copyright 2006, 2007 Rob Landley <rob@landley.net>
-
-// Still to go: "E:jJ:L:m:O:"
-USE_MKE2FS(NEWTOY(mke2fs, "<1>2g:Fnqm#N#i#b#", TOYFLAG_SBIN))
-
-config MKE2FS
- bool "mke2fs (unfinished and broken by dirtree changes)"
- default n
- help
- usage: mke2fs [-Fnq] [-b ###] [-N|i ###] [-m ###] device
-
- Create an ext2 filesystem on a block device or filesystem image.
-
- -F Force to run on a mounted device
- -n Don't write to device
- -q Quiet (no output)
- -b size Block size (1024, 2048, or 4096)
- -N inodes Allocate this many inodes
- -i bytes Allocate one inode for every XXX bytes of device
- -m percent Reserve this percent of filesystem space for root user
-
-config MKE2FS_JOURNAL
- bool "Journaling support (ext3)"
- default n
- depends on MKE2FS
- help
- usage: [-j] [-J size=###,device=XXX]
-
- -j Create journal (ext3)
- -J Journal options
- size: Number of blocks (1024-102400)
- device: Specify an external journal
-
-config MKE2FS_GEN
- bool "Generate (gene2fs)"
- default n
- depends on MKE2FS
- help
- usage: gene2fs [options] device filename
-
- The [options] are the same as mke2fs.
-
-config MKE2FS_LABEL
- bool "Label support"
- default n
- depends on MKE2FS
- help
- usage: mke2fs [-L label] [-M path] [-o string]
-
- -L Volume label
- -M Path to mount point
- -o Created by
-
-config MKE2FS_EXTENDED
- bool "Extended options"
- default n
- depends on MKE2FS
- help
- usage: mke2fs [-E stride=###] [-O option[,option]]
-
- -E stride= Set RAID stripe size (in blocks)
- -O [opts] Specify fewer ext2 option flags (for old kernels)
- All of these are on by default (as appropriate)
- none Clear default options (all but journaling)
- dir_index Use htree indexes for large directories
- filetype Store file type info in directory entry
- has_journal Set by -j
- journal_dev Set by -J device=XXX
- sparse_super Don't allocate huge numbers of redundant superblocks
-*/
-
-#define FOR_mke2fs
-#include "toys.h"
-
-GLOBALS(
- // Command line arguments.
- long blocksize;
- long bytes_per_inode;
- long inodes; // Total inodes in filesystem.
- long reserved_percent; // Integer precent of space to reserve for root.
- char *gendir; // Where to read dirtree from.
-
- // Internal data.
- struct dirtree *dt; // Tree of files to copy into the new filesystem.
- unsigned treeblocks; // Blocks used by dt
- unsigned treeinodes; // Inodes used by dt
-
- unsigned blocks; // Total blocks in the filesystem.
- unsigned freeblocks; // Free blocks in the filesystem.
- unsigned inodespg; // Inodes per group
- unsigned groups; // Total number of block groups.
- unsigned blockbits; // Bits per block. (Also blocks per group.)
-
- // For gene2fs
- unsigned nextblock; // Next data block to allocate
- unsigned nextgroup; // Next group we'll be allocating from
- int fsfd; // File descriptor of filesystem (to output to).
-
- struct ext2_superblock sb;
-)
-
-#define INODES_RESERVED 10
-
-static uint32_t div_round_up(uint32_t a, uint32_t b)
-{
- uint32_t c = a/b;
-
- if (a%b) c++;
- return c;
-}
-
-// Calculate data blocks plus index blocks needed to hold a file.
-
-static uint32_t file_blocks_used(uint64_t size, uint32_t *blocklist)
-{
- uint32_t dblocks = (uint32_t)((size+(TT.blocksize-1))/TT.blocksize);
- uint32_t idx=TT.blocksize/4, iblocks=0, diblocks=0, tiblocks=0;
-
- // Fill out index blocks in inode.
-
- if (blocklist) {
- int i;
-
- // Direct index blocks
- for (i=0; i<13 && i<dblocks; i++) blocklist[i] = i;
- // Singly indirect index blocks
- if (dblocks > 13+idx) blocklist[13] = 13+idx;
- // Doubly indirect index blocks
- idx = 13 + idx + (idx*idx);
- if (dblocks > idx) blocklist[14] = idx;
-
- return 0;
- }
-
- // Account for direct, singly, doubly, and triply indirect index blocks
-
- if (dblocks > 12) {
- iblocks = ((dblocks-13)/idx)+1;
- if (iblocks > 1) {
- diblocks = ((iblocks-2)/idx)+1;
- if (diblocks > 1)
- tiblocks = ((diblocks-2)/idx)+1;
- }
- }
-
- return dblocks + iblocks + diblocks + tiblocks;
-}
-
-// Use the parent pointer to iterate through the tree non-recursively.
-static struct dirtree *treenext(struct dirtree *this)
-{
- while (this && !this->next) this = this->parent;
- if (this) this = this->next;
-
- return this;
-}
-
-// Recursively calculate the number of blocks used by each inode in the tree.
-// Returns blocks used by this directory, assigns bytes used to *size.
-// Writes total block count to TT.treeblocks and inode count to TT.treeinodes.
-
-static long check_treesize(struct dirtree *that, off_t *size)
-{
- long blocks;
-
- while (that) {
- *size += sizeof(struct ext2_dentry) + strlen(that->name);
-
- if (that->child)
- that->st.st_blocks = check_treesize(that->child, &that->st.st_size);
- else if (S_ISREG(that->st.st_mode)) {
- that->st.st_blocks = file_blocks_used(that->st.st_size, 0);
- TT.treeblocks += that->st.st_blocks;
- }
- that = that->next;
- }
- TT.treeblocks += blocks = file_blocks_used(*size, 0);
- TT.treeinodes++;
-
- return blocks;
-}
-
-// Calculate inode numbers and link counts.
-//
-// To do this right I need to copy the tree and sort it, but here's a really
-// ugly n^2 way of dealing with the problem that doesn't scale well to large
-// numbers of files (> 100,000) but can be done in very little code.
-// This rewrites inode numbers to their final values, allocating depth first.
-
-static void check_treelinks(struct dirtree *tree)
-{
- struct dirtree *current=tree, *that;
- long inode = INODES_RESERVED;
-
- while (current) {
- ++inode;
- // Since we can't hardlink to directories, we know their link count.
- if (S_ISDIR(current->st.st_mode)) current->st.st_nlink = 2;
- else {
- dev_t new = current->st.st_dev;
-
- if (!new) continue;
-
- // Look for other copies of current node
- current->st.st_nlink = 0;
- for (that = tree; that; that = treenext(that)) {
- if (current->st.st_ino == that->st.st_ino &&
- current->st.st_dev == that->st.st_dev)
- {
- current->st.st_nlink++;
- current->st.st_ino = inode;
- }
- }
- }
- current->st.st_ino = inode;
- current = treenext(current);
- }
-}
-
-// According to http://www.opengroup.org/onlinepubs/9629399/apdxa.htm
-// we should generate a uuid structure by reading a clock with 100 nanosecond
-// precision, normalizing it to the start of the gregorian calendar in 1582,
-// and looking up our eth0 mac address.
-//
-// On the other hand, we have 128 bits to come up with a unique identifier, of
-// which 6 have a defined value. /dev/urandom it is.
-
-static void create_uuid(char *uuid)
-{
- // Read 128 random bits
- int fd = xopen("/dev/urandom", O_RDONLY);
- xreadall(fd, uuid, 16);
- close(fd);
-
- // Claim to be a DCE format UUID.
- uuid[6] = (uuid[6] & 0x0F) | 0x40;
- uuid[8] = (uuid[8] & 0x3F) | 0x80;
-
- // rfc2518 section 6.4.1 suggests if we're not using a macaddr, we should
- // set bit 1 of the node ID, which is the mac multicast bit. This means we
- // should never collide with anybody actually using a macaddr.
- uuid[11] = uuid[11] | 128;
-}
-
-// Calculate inodes per group from total inodes.
-static uint32_t get_inodespg(uint32_t inodes)
-{
- uint32_t temp;
-
- // Round up to fill complete inode blocks.
- temp = (inodes + TT.groups - 1) / TT.groups;
- inodes = TT.blocksize/sizeof(struct ext2_inode);
- return ((temp + inodes - 1)/inodes)*inodes;
-}
-
-// Fill out superblock and TT structures.
-
-static void init_superblock(struct ext2_superblock *sb)
-{
- uint32_t temp;
-
- // Set log_block_size and log_frag_size.
-
- for (temp = 0; temp < 4; temp++) if (TT.blocksize == 1024<<temp) break;
- if (temp==4) error_exit("bad blocksize");
- sb->log_block_size = sb->log_frag_size = SWAP_LE32(temp);
-
- // Fill out blocks_count, r_blocks_count, first_data_block
-
- sb->blocks_count = SWAP_LE32(TT.blocks);
- sb->free_blocks_count = SWAP_LE32(TT.freeblocks);
- temp = (TT.blocks * (uint64_t)TT.reserved_percent) / 100;
- sb->r_blocks_count = SWAP_LE32(temp);
-
- sb->first_data_block = SWAP_LE32(TT.blocksize == 1024 ? 1 : 0);
-
- // Set blocks_per_group and frags_per_group, which is the size of an
- // allocation bitmap that fits in one block (I.E. how many bits per block)?
-
- sb->blocks_per_group = sb->frags_per_group = SWAP_LE32(TT.blockbits);
-
- // Set inodes_per_group and total inodes_count
- sb->inodes_per_group = SWAP_LE32(TT.inodespg);
- sb->inodes_count = SWAP_LE32(TT.inodespg * TT.groups);
-
- // Determine free inodes.
- temp = TT.inodespg*TT.groups - INODES_RESERVED;
- if (temp < TT.treeinodes) error_exit("Not enough inodes.\n");
- sb->free_inodes_count = SWAP_LE32(temp - TT.treeinodes);
-
- // Fill out the rest of the superblock.
- sb->max_mnt_count=0xFFFF;
- sb->wtime = sb->lastcheck = sb->mkfs_time = SWAP_LE32(time(NULL));
- sb->magic = SWAP_LE32(0xEF53);
- sb->state = sb->errors = SWAP_LE16(1);
-
- sb->rev_level = SWAP_LE32(1);
- sb->first_ino = SWAP_LE32(INODES_RESERVED+1);
- sb->inode_size = SWAP_LE16(sizeof(struct ext2_inode));
- sb->feature_incompat = SWAP_LE32(EXT2_FEATURE_INCOMPAT_FILETYPE);
- sb->feature_ro_compat = SWAP_LE32(EXT2_FEATURE_RO_COMPAT_SPARSE_SUPER);
-
- create_uuid(sb->uuid);
-
- // TODO If we're called as mke3fs or mkfs.ext3, do a journal.
-
- //if (strchr(toys.which->name,'3'))
- // sb->feature_compat |= SWAP_LE32(EXT3_FEATURE_COMPAT_HAS_JOURNAL);
-}
-
-// Does this group contain a superblock backup (and group descriptor table)?
-static int is_sb_group(uint32_t group)
-{
- int i;
-
- // Superblock backups are on groups 0, 1, and powers of 3, 5, and 7.
- if(!group || group==1) return 1;
- for (i=3; i<9; i+=2) {
- int j = i;
- while (j<group) j*=i;
- if (j==group) return 1;
- }
- return 0;
-}
-
-
-// Number of blocks used in group by optional superblock/group list backup.
-static int group_superblock_overhead(uint32_t group)
-{
- int used;
-
- if (!is_sb_group(group)) return 0;
-
- // How many blocks does the group descriptor table take up?
- used = TT.groups * sizeof(struct ext2_group);
- used += TT.blocksize - 1;
- used /= TT.blocksize;
- // Plus the superblock itself.
- used++;
- // And a corner case.
- if (!group && TT.blocksize == 1024) used++;
-
- return used;
-}
-
-// Number of blocks used in group to store superblock/group/inode list
-static int group_overhead(uint32_t group)
-{
- // Return superblock backup overhead (if any), plus block/inode
- // allocation bitmaps, plus inode tables.
- return group_superblock_overhead(group) + 2 + get_inodespg(TT.inodespg)
- / (TT.blocksize/sizeof(struct ext2_inode));
-}
-
-// In bitmap "array" set "len" bits starting at position "start" (from 0).
-static void bits_set(char *array, int start, int len)
-{
- while(len) {
- if ((start&7) || len<8) {
- array[start/8]|=(1<<(start&7));
- start++;
- len--;
- } else {
- array[start/8]=255;
- start+=8;
- len-=8;
- }
- }
-}
-
-// Seek past len bytes (to maintain sparse file), or write zeroes if output
-// not seekable
-static void put_zeroes(int len)
-{
- if(-1 == lseek(TT.fsfd, len, SEEK_SET)) {
- memset(toybuf, 0, sizeof(toybuf));
- while (len) {
- int out = len > sizeof(toybuf) ? sizeof(toybuf) : len;
- xwrite(TT.fsfd, toybuf, out);
- len -= out;
- }
- }
-}
-
-// Fill out an inode structure from struct stat info in dirtree.
-static void fill_inode(struct ext2_inode *in, struct dirtree *that)
-{
- uint32_t fbu[15];
- int temp;
-
- file_blocks_used(that->st.st_size, fbu);
-
- // If that inode needs data blocks allocated to it.
- if (that->st.st_size) {
- int i, group = TT.nextblock/TT.blockbits;
-
- // TODO: teach this about indirect blocks.
- for (i=0; i<15; i++) {
- // If we just jumped into a new group, skip group overhead blocks.
- while (group >= TT.nextgroup)
- TT.nextblock += group_overhead(TT.nextgroup++);
- }
- }
- // TODO : S_ISREG/DIR/CHR/BLK/FIFO/LNK/SOCK(m)
- in->mode = SWAP_LE32(that->st.st_mode);
-
- in->uid = SWAP_LE16(that->st.st_uid & 0xFFFF);
- in->uid_high = SWAP_LE16(that->st.st_uid >> 16);
- in->gid = SWAP_LE16(that->st.st_gid & 0xFFFF);
- in->gid_high = SWAP_LE16(that->st.st_gid >> 16);
- in->size = SWAP_LE32(that->st.st_size & 0xFFFFFFFF);
-
- // Contortions to make the compiler not generate a warning for x>>32
- // when x is 32 bits. The optimizer should clean this up.
- if (sizeof(that->st.st_size) > 4) temp = 32;
- else temp = 0;
- if (temp) in->dir_acl = SWAP_LE32(that->st.st_size >> temp);
-
- in->atime = SWAP_LE32(that->st.st_atime);
- in->ctime = SWAP_LE32(that->st.st_ctime);
- in->mtime = SWAP_LE32(that->st.st_mtime);
-
- in->links_count = SWAP_LE16(that->st.st_nlink);
- in->blocks = SWAP_LE32(that->st.st_blocks);
- // in->faddr
-}
-
-// Works like an archiver.
-// The first argument is the name of the file to create. If it already
-// exists, that size will be used.
-
-void mke2fs_main(void)
-{
- int i, temp;
- off_t length;
- uint32_t usedblocks, usedinodes, dtiblk, dtbblk;
- struct dirtree *dti, *dtb;
-
- // Handle command line arguments.
-
- if (toys.optargs[1]) {
- sscanf(toys.optargs[1], "%u", &TT.blocks);
- temp = O_RDWR|O_CREAT;
- } else temp = O_RDWR;
- if (!TT.reserved_percent) TT.reserved_percent = 5;
-
- // TODO: Check if filesystem is mounted here
-
- // For mke?fs, open file. For gene?fs, create file.
- TT.fsfd = xcreate(*toys.optargs, temp, 0777);
-
- // Determine appropriate block size and block count from file length.
- // (If no length, default to 4k. They can override it on the cmdline.)
-
- length = fdlength(TT.fsfd);
- if (!TT.blocksize) TT.blocksize = (length && length < 1<<29) ? 1024 : 4096;
- TT.blockbits = 8*TT.blocksize;
- if (!TT.blocks) TT.blocks = length/TT.blocksize;
-
- // Collect gene2fs list or lost+found, calculate requirements.
-
- if (TT.gendir) {
- strncpy(toybuf, TT.gendir, sizeof(toybuf));
- dti = dirtree_read(toybuf, NULL, NULL);
- } else {
- dti = xzalloc(sizeof(struct dirtree)+11);
- strcpy(dti->name, "lost+found");
- dti->st.st_mode = S_IFDIR|0755;
- dti->st.st_ctime = dti->st.st_mtime = time(NULL);
- }
-
- // Add root directory inode. This is iterated through for when finding
- // blocks, but not when finding inodes. The tree's parent pointers don't
- // point back into this.
-
- dtb = xzalloc(sizeof(struct dirtree)+1);
- dtb->st.st_mode = S_IFDIR|0755;
- dtb->st.st_ctime = dtb->st.st_mtime = time(NULL);
- dtb->child = dti;
-
- // Figure out how much space is used by preset files
- length = check_treesize(dtb, &(dtb->st.st_size));
- check_treelinks(dtb);
-
- // Figure out how many total inodes we need.
-
- if (!TT.inodes) {
- if (!TT.bytes_per_inode) TT.bytes_per_inode = 8192;
- TT.inodes = (TT.blocks * (uint64_t)TT.blocksize) / TT.bytes_per_inode;
- }
-
- // If we're generating a filesystem and have no idea how many blocks it
- // needs, start with a minimal guess, find the overhead of that many
- // groups, and loop until this is enough groups to store this many blocks.
- if (!TT.blocks) TT.groups = (TT.treeblocks/TT.blockbits)+1;
- else TT.groups = div_round_up(TT.blocks, TT.blockbits);
-
- for (;;) {
- temp = TT.treeblocks;
-
- for (i = 0; i<TT.groups; i++) temp += group_overhead(i);
-
- if (TT.blocks) {
- if (TT.blocks < temp) error_exit("Not enough space.\n");
- break;
- }
- if (temp <= TT.groups * TT.blockbits) {
- TT.blocks = temp;
- break;
- }
- TT.groups++;
- }
- TT.freeblocks = TT.blocks - temp;
-
- // Now we know all the TT data, initialize superblock structure.
-
- init_superblock(&TT.sb);
-
- // Start writing. Skip the first 1k to avoid the boot sector (if any).
- put_zeroes(1024);
-
- // Loop through block groups, write out each one.
- dtiblk = dtbblk = usedblocks = usedinodes = 0;
- for (i=0; i<TT.groups; i++) {
- struct ext2_inode *in = (struct ext2_inode *)toybuf;
- uint32_t start, itable, used, end;
- int j, slot;
-
- // Where does this group end?
- end = TT.blockbits;
- if ((i+1)*TT.blockbits > TT.blocks) end = TT.blocks & (TT.blockbits-1);
-
- // Blocks used by inode table
- itable = (TT.inodespg*sizeof(struct ext2_inode))/TT.blocksize;
-
- // If a superblock goes here, write it out.
- start = group_superblock_overhead(i);
- if (start) {
- struct ext2_group *bg = (struct ext2_group *)toybuf;
- int treeblocks = TT.treeblocks, treeinodes = TT.treeinodes;
-
- TT.sb.block_group_nr = SWAP_LE16(i);
-
- // Write superblock and pad it up to block size
- xwrite(TT.fsfd, &TT.sb, sizeof(struct ext2_superblock));
- temp = TT.blocksize - sizeof(struct ext2_superblock);
- if (!i && TT.blocksize > 1024) temp -= 1024;
- memset(toybuf, 0, TT.blocksize);
- xwrite(TT.fsfd, toybuf, temp);
-
- // Loop through groups to write group descriptor table.
- for(j=0; j<TT.groups; j++) {
-
- // Figure out what sector this group starts in.
- used = group_superblock_overhead(j);
-
- // Find next array slot in this block (flush block if full).
- slot = j % (TT.blocksize/sizeof(struct ext2_group));
- if (!slot) {
- if (j) xwrite(TT.fsfd, bg, TT.blocksize);
- memset(bg, 0, TT.blocksize);
- }
-
- // How many free inodes in this group?
- temp = TT.inodespg;
- if (!i) temp -= INODES_RESERVED;
- if (temp > treeinodes) {
- treeinodes -= temp;
- temp = 0;
- } else {
- temp -= treeinodes;
- treeinodes = 0;
- }
- bg[slot].free_inodes_count = SWAP_LE16(temp);
-
- // How many free blocks in this group?
- temp = TT.inodespg/(TT.blocksize/sizeof(struct ext2_inode)) + 2;
- temp = end-used-temp;
- if (temp > treeblocks) {
- treeblocks -= temp;
- temp = 0;
- } else {
- temp -= treeblocks;
- treeblocks = 0;
- }
- bg[slot].free_blocks_count = SWAP_LE32(temp);
-
- // Fill out rest of group structure
- used += j*TT.blockbits;
- bg[slot].block_bitmap = SWAP_LE32(used++);
- bg[slot].inode_bitmap = SWAP_LE32(used++);
- bg[slot].inode_table = SWAP_LE32(used);
- bg[slot].used_dirs_count = 0; // (TODO)
- }
- xwrite(TT.fsfd, bg, TT.blocksize);
- }
-
- // Now write out stuff that every block group has.
-
- // Write block usage bitmap
-
- start += 2 + itable;
- memset(toybuf, 0, TT.blocksize);
- bits_set(toybuf, 0, start);
- bits_set(toybuf, end, TT.blockbits-end);
- temp = TT.treeblocks - usedblocks;
- if (temp) {
- if (end-start > temp) temp = end-start;
- bits_set(toybuf, start, temp);
- }
- xwrite(TT.fsfd, toybuf, TT.blocksize);
-
- // Write inode bitmap
- memset(toybuf, 0, TT.blocksize);
- j = 0;
- if (!i) bits_set(toybuf, 0, j = INODES_RESERVED);
- bits_set(toybuf, TT.inodespg, slot = TT.blockbits-TT.inodespg);
- temp = TT.treeinodes - usedinodes;
- if (temp) {
- if (slot-j > temp) temp = slot-j;
- bits_set(toybuf, j, temp);
- }
- xwrite(TT.fsfd, toybuf, TT.blocksize);
-
- // Write inode table for this group (TODO)
- for (j = 0; j<TT.inodespg; j++) {
- slot = j % (TT.blocksize/sizeof(struct ext2_inode));
- if (!slot) {
- if (j) xwrite(TT.fsfd, in, TT.blocksize);
- memset(in, 0, TT.blocksize);
- }
- if (!i && j<INODES_RESERVED) {
- // Write root inode
- if (j == 2) fill_inode(in+slot, dtb);
- } else if (dti) {
- fill_inode(in+slot, dti);
- dti = treenext(dti);
- }
- }
- xwrite(TT.fsfd, in, TT.blocksize);
-
- while (dtb) {
- // TODO write index data block
- // TODO write root directory data block
- // TODO write directory data block
- // TODO write file data block
- put_zeroes(TT.blocksize);
- start++;
- if (start == end) break;
- }
- // Write data blocks (TODO)
- put_zeroes((end-start) * TT.blocksize);
- }
-}