/*
* Free an inode from the filesystem
*/
void testfs_free_inode(struct inode *inode)
{
struct buffer_head *bitmap_bh = NULL;
struct super_block *sb = inode->i_sb;
struct testfs_super_block *tsb = TESTFS_SB(sb)->s_ts;
unsigned int ino = inode->i_ino;
BUG_ON(!tsb);
testfs_debug("Freeing inode %u\n",ino);
if (ino <= tsb->s_first_nonmeta_inode || ino >= tsb->s_max_inodes) {
testfs_error("Invalid inode number to be freed %u\n",ino);
goto error_return;
}
clear_inode(inode);
bitmap_bh = read_inode_bitmap(sb);
if (inode_already_freed(bitmap_bh->b_data, ino)) {
testfs_error("Inode already free %u\n",ino);
goto error_return;
}
testfs_clear_inode_bit(bitmap_bh->b_data, ino);
testfs_release_inode(sb);
mark_buffer_dirty(bitmap_bh);
error_return:
return;
}
/*
* load_inode_bitmap loads the inode bitmap for a blocks group
*
* It maintains a cache for the last bitmaps loaded. This cache is managed
* with a LRU algorithm.
*
* Notes:
* 1/ There is one cache per mounted file system.
* 2/ If the file system contains less than EXT2_MAX_GROUP_LOADED groups,
* this function reads the bitmap without maintaining a LRU cache.
*
* Return the buffer_head of the bitmap or the ERR_PTR(error)
*/
static struct buffer_head *load_inode_bitmap (struct super_block * sb,
unsigned int block_group)
{
int i, slot = 0;
struct ext2_sb_info *sbi = &sb->u.ext2_sb;
struct buffer_head *bh = sbi->s_inode_bitmap[0];
if (block_group >= sbi->s_groups_count)
ext2_panic (sb, "load_inode_bitmap",
"block_group >= groups_count - "
"block_group = %d, groups_count = %lu",
block_group, sbi->s_groups_count);
if (sbi->s_loaded_inode_bitmaps > 0 &&
sbi->s_inode_bitmap_number[0] == block_group && bh)
goto found;
if (sbi->s_groups_count <= EXT2_MAX_GROUP_LOADED) {
slot = block_group;
bh = sbi->s_inode_bitmap[slot];
if (!bh)
goto read_it;
if (sbi->s_inode_bitmap_number[slot] == slot)
goto found;
ext2_panic (sb, "load_inode_bitmap",
"block_group != inode_bitmap_number");
}
bh = NULL;
for (i = 0; i < sbi->s_loaded_inode_bitmaps &&
sbi->s_inode_bitmap_number[i] != block_group;
i++)
;
if (i < sbi->s_loaded_inode_bitmaps)
bh = sbi->s_inode_bitmap[i];
else if (sbi->s_loaded_inode_bitmaps < EXT2_MAX_GROUP_LOADED)
sbi->s_loaded_inode_bitmaps++;
else
brelse (sbi->s_inode_bitmap[--i]);
while (i--) {
sbi->s_inode_bitmap_number[i+1] = sbi->s_inode_bitmap_number[i];
sbi->s_inode_bitmap[i+1] = sbi->s_inode_bitmap[i];
}
read_it:
if (!bh)
bh = read_inode_bitmap (sb, block_group);
sbi->s_inode_bitmap_number[slot] = block_group;
sbi->s_inode_bitmap[slot] = bh;
if (!bh)
return ERR_PTR(-EIO);
found:
return bh;
}
/* Verify that we are loading a valid orphan from disk */
struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino)
{
unsigned long max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count);
unsigned long block_group;
int bit;
struct buffer_head *bitmap_bh = NULL;
struct inode *inode = NULL;
/* Error cases - e2fsck has already cleaned up for us */
if (ino > max_ino) {
ext3_warning(sb, __FUNCTION__,
"bad orphan ino %lu! e2fsck was run?\n", ino);
goto out;
}
block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb);
bitmap_bh = read_inode_bitmap(sb, block_group);
if (!bitmap_bh) {
ext3_warning(sb, __FUNCTION__,
"inode bitmap error for orphan %lu\n", ino);
goto out;
}
/* Having the inode bit set should be a 100% indicator that this
* is a valid orphan (no e2fsck run on fs). Orphans also include
* inodes that were being truncated, so we can't check i_nlink==0.
*/
if (!ext3_test_bit(bit, bitmap_bh->b_data) ||
!(inode = iget(sb, ino)) || is_bad_inode(inode) ||
NEXT_ORPHAN(inode) > max_ino) {
ext3_warning(sb, __FUNCTION__,
"bad orphan inode %lu! e2fsck was run?\n", ino);
printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n",
bit, (unsigned long long)bitmap_bh->b_blocknr,
ext3_test_bit(bit, bitmap_bh->b_data));
printk(KERN_NOTICE "inode=%p\n", inode);
if (inode) {
printk(KERN_NOTICE "is_bad_inode(inode)=%d\n",
is_bad_inode(inode));
printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n",
NEXT_ORPHAN(inode));
printk(KERN_NOTICE "max_ino=%lu\n", max_ino);
}
/* Avoid freeing blocks if we got a bad deleted inode */
if (inode && inode->i_nlink == 0)
inode->i_blocks = 0;
iput(inode);
inode = NULL;
}
out:
brelse(bitmap_bh);
return inode;
}
unsigned long ext3_count_free_inodes (struct super_block * sb)
{
unsigned long desc_count;
struct ext3_group_desc *gdp;
int i;
#ifdef EXT3FS_DEBUG
struct ext3_super_block *es;
unsigned long bitmap_count, x;
struct buffer_head *bitmap_bh = NULL;
lock_super (sb);
es = EXT3_SB(sb)->s_es;
desc_count = 0;
bitmap_count = 0;
gdp = NULL;
for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) {
gdp = ext3_get_group_desc (sb, i, NULL);
if (!gdp)
continue;
desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
brelse(bitmap_bh);
bitmap_bh = read_inode_bitmap(sb, i);
if (!bitmap_bh)
continue;
x = ext3_count_free(bitmap_bh, EXT3_INODES_PER_GROUP(sb) / 8);
printk("group %d: stored = %d, counted = %lu\n",
i, le16_to_cpu(gdp->bg_free_inodes_count), x);
bitmap_count += x;
}
brelse(bitmap_bh);
printk("ext3_count_free_inodes: stored = %u, computed = %lu, %lu\n",
le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
unlock_super(sb);
return desc_count;
#else
desc_count = 0;
for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) {
gdp = ext3_get_group_desc (sb, i, NULL);
if (!gdp)
continue;
desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
cond_resched();
}
return desc_count;
#endif
}
struct inode *testfs_new_inode(struct inode *dir, int mode)
{
struct super_block *sb = dir->i_sb;
struct testfs_sb_info *tsbi = TESTFS_SB(sb);
struct testfs_inode_info *tsi;
struct buffer_head *bitmap_bh = NULL;
struct inode *inode;
unsigned int ino = 0;
inode = new_inode(sb);
if(!inode)
{
testfs_debug("Could not allocate inode from inode cache\n");
return ERR_PTR(-ENOMEM);
}
tsi = TESTFS_I(inode);
bitmap_bh = read_inode_bitmap(sb);
ino = testfs_find_free_inode(bitmap_bh->b_data, sb);
if(!ino)
{
testfs_debug("Could not find any free inode. File system full\n");
return ERR_PTR(-ENOSPC);
}
testfs_debug("Allocated new inode (%u)\n",ino);
testfs_set_inode_bit(bitmap_bh->b_data, ino);
inode->i_ino = ino;
inode->i_mode = mode;
inode->i_gid = current->fsgid;
inode->i_uid = current->fsuid;
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
testfs_debug("Successfully allocated inodes....\n");
memset(tsi->i_data, 0 ,sizeof(tsi->i_data));
tsi->i_data[0] = ino;
tsi->state = TESTFS_INODE_ALLOCATED;
tsbi->s_free_inodes--;
sb->s_dirt = 1;
insert_inode_hash(inode);
mark_inode_dirty(inode);
mark_buffer_dirty(bitmap_bh);
sync_dirty_buffer(bitmap_bh);
testfs_debug("returning now\n");
return inode;
}
/* Called at mount-time, super-block is locked */
void ext3_check_inodes_bitmap (struct super_block * sb)
{
struct ext3_super_block * es;
unsigned long desc_count, bitmap_count, x;
struct buffer_head *bitmap_bh = NULL;
struct ext3_group_desc * gdp;
int i;
es = EXT3_SB(sb)->s_es;
desc_count = 0;
bitmap_count = 0;
gdp = NULL;
for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) {
gdp = ext3_get_group_desc (sb, i, NULL);
if (!gdp)
continue;
desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
brelse(bitmap_bh);
bitmap_bh = read_inode_bitmap(sb, i);
if (!bitmap_bh)
continue;
x = ext3_count_free(bitmap_bh, EXT3_INODES_PER_GROUP(sb) / 8);
if (le16_to_cpu(gdp->bg_free_inodes_count) != x)
ext3_error (sb, "ext3_check_inodes_bitmap",
"Wrong free inodes count in group %d, "
"stored = %d, counted = %lu", i,
le16_to_cpu(gdp->bg_free_inodes_count), x);
bitmap_count += x;
}
brelse(bitmap_bh);
if (le32_to_cpu(es->s_free_inodes_count) != bitmap_count)
ext3_error (sb, "ext3_check_inodes_bitmap",
"Wrong free inodes count in super block, "
"stored = %lu, counted = %lu",
(unsigned long)le32_to_cpu(es->s_free_inodes_count),
bitmap_count);
}
/*
* NOTE! When we get the inode, we're the only people
* that have access to it, and as such there are no
* race conditions we have to worry about. The inode
* is not on the hash-lists, and it cannot be reached
* through the filesystem because the directory entry
* has been deleted earlier.
*
* HOWEVER: we must make sure that we get no aliases,
* which means that we have to call "clear_inode()"
* _before_ we mark the inode not in use in the inode
* bitmaps. Otherwise a newly created file might use
* the same inode number (not actually the same pointer
* though), and then we'd have two inodes sharing the
* same inode number and space on the harddisk.
*/
void ext3_free_inode (handle_t *handle, struct inode * inode)
{
struct super_block * sb = inode->i_sb;
int is_directory;
unsigned long ino;
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *bh2;
unsigned long block_group;
unsigned long bit;
struct ext3_group_desc * gdp;
struct ext3_super_block * es;
struct ext3_sb_info *sbi;
int fatal = 0, err;
if (atomic_read(&inode->i_count) > 1) {
printk ("ext3_free_inode: inode has count=%d\n",
atomic_read(&inode->i_count));
return;
}
if (inode->i_nlink) {
printk ("ext3_free_inode: inode has nlink=%d\n",
inode->i_nlink);
return;
}
if (!sb) {
printk("ext3_free_inode: inode on nonexistent device\n");
return;
}
sbi = EXT3_SB(sb);
ino = inode->i_ino;
ext3_debug ("freeing inode %lu\n", ino);
/*
* Note: we must free any quota before locking the superblock,
* as writing the quota to disk may need the lock as well.
*/
DQUOT_INIT(inode);
ext3_xattr_delete_inode(handle, inode);
DQUOT_FREE_INODE(inode);
DQUOT_DROP(inode);
is_directory = S_ISDIR(inode->i_mode);
/* Do this BEFORE marking the inode not in use or returning an error */
clear_inode (inode);
es = EXT3_SB(sb)->s_es;
if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
ext3_error (sb, "ext3_free_inode",
"reserved or nonexistent inode %lu", ino);
goto error_return;
}
block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb);
bitmap_bh = read_inode_bitmap(sb, block_group);
if (!bitmap_bh)
goto error_return;
BUFFER_TRACE(bitmap_bh, "get_write_access");
fatal = ext3_journal_get_write_access(handle, bitmap_bh);
if (fatal)
goto error_return;
/* Ok, now we can actually update the inode bitmaps.. */
if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
bit, bitmap_bh->b_data))
ext3_error (sb, "ext3_free_inode",
"bit already cleared for inode %lu", ino);
else {
gdp = ext3_get_group_desc (sb, block_group, &bh2);
BUFFER_TRACE(bh2, "get_write_access");
fatal = ext3_journal_get_write_access(handle, bh2);
if (fatal) goto error_return;
if (gdp) {
spin_lock(sb_bgl_lock(sbi, block_group));
gdp->bg_free_inodes_count = cpu_to_le16(
le16_to_cpu(gdp->bg_free_inodes_count) + 1);
if (is_directory)
gdp->bg_used_dirs_count = cpu_to_le16(
le16_to_cpu(gdp->bg_used_dirs_count) - 1);
spin_unlock(sb_bgl_lock(sbi, block_group));
percpu_counter_inc(&sbi->s_freeinodes_counter);
if (is_directory)
percpu_counter_dec(&sbi->s_dirs_counter);
}
BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle, bh2);
if (!fatal) fatal = err;
}
BUFFER_TRACE(bitmap_bh, "call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle, bitmap_bh);
if (!fatal)
fatal = err;
sb->s_dirt = 1;
error_return:
brelse(bitmap_bh);
ext3_std_error(sb, fatal);
}
/*
* There are two policies for allocating an inode. If the new inode is
* a directory, then a forward search is made for a block group with both
* free space and a low directory-to-inode ratio; if that fails, then of
* the groups with above-average free space, that group with the fewest
* directories already is chosen.
*
* For other inodes, search forward from the parent directory's block
* group to find a free inode.
*/
struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode)
{
struct super_block *sb;
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *bh2;
int group;
unsigned long ino = 0;
struct inode * inode;
struct ext3_group_desc * gdp = NULL;
struct ext3_super_block * es;
struct ext3_inode_info *ei;
struct ext3_sb_info *sbi;
int err = 0;
struct inode *ret;
int i;
/* Cannot create files in a deleted directory */
if (!dir || !dir->i_nlink)
return ERR_PTR(-EPERM);
sb = dir->i_sb;
inode = new_inode(sb);
if (!inode)
return ERR_PTR(-ENOMEM);
ei = EXT3_I(inode);
sbi = EXT3_SB(sb);
es = sbi->s_es;
if (S_ISDIR(mode)) {
if (test_opt (sb, OLDALLOC))
group = find_group_dir(sb, dir);
else
group = find_group_orlov(sb, dir);
} else
group = find_group_other(sb, dir);
err = -ENOSPC;
if (group == -1)
goto out;
for (i = 0; i < sbi->s_groups_count; i++) {
err = -EIO;
gdp = ext3_get_group_desc(sb, group, &bh2);
if (!gdp)
goto fail;
brelse(bitmap_bh);
bitmap_bh = read_inode_bitmap(sb, group);
if (!bitmap_bh)
goto fail;
ino = 0;
repeat_in_this_group:
ino = ext3_find_next_zero_bit((unsigned long *)
bitmap_bh->b_data, EXT3_INODES_PER_GROUP(sb), ino);
if (ino < EXT3_INODES_PER_GROUP(sb)) {
BUFFER_TRACE(bitmap_bh, "get_write_access");
err = ext3_journal_get_write_access(handle, bitmap_bh);
if (err)
goto fail;
if (!ext3_set_bit_atomic(sb_bgl_lock(sbi, group),
ino, bitmap_bh->b_data)) {
/* we won it */
BUFFER_TRACE(bitmap_bh,
"call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle,
bitmap_bh);
if (err)
goto fail;
goto got;
}
/* we lost it */
journal_release_buffer(handle, bitmap_bh);
if (++ino < EXT3_INODES_PER_GROUP(sb))
goto repeat_in_this_group;
}
/*
* This case is possible in concurrent environment. It is very
* rare. We cannot repeat the find_group_xxx() call because
* that will simply return the same blockgroup, because the
* group descriptor metadata has not yet been updated.
* So we just go onto the next blockgroup.
*/
if (++group == sbi->s_groups_count)
group = 0;
}
err = -ENOSPC;
goto out;
got:
ino += group * EXT3_INODES_PER_GROUP(sb) + 1;
if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
ext3_error (sb, "ext3_new_inode",
"reserved inode or inode > inodes count - "
"block_group = %d, inode=%lu", group, ino);
err = -EIO;
goto fail;
}
BUFFER_TRACE(bh2, "get_write_access");
err = ext3_journal_get_write_access(handle, bh2);
if (err) goto fail;
spin_lock(sb_bgl_lock(sbi, group));
gdp->bg_free_inodes_count =
cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1);
if (S_ISDIR(mode)) {
gdp->bg_used_dirs_count =
cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1);
}
spin_unlock(sb_bgl_lock(sbi, group));
BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle, bh2);
if (err) goto fail;
percpu_counter_dec(&sbi->s_freeinodes_counter);
if (S_ISDIR(mode))
percpu_counter_inc(&sbi->s_dirs_counter);
sb->s_dirt = 1;
inode->i_uid = current->fsuid;
if (test_opt (sb, GRPID))
inode->i_gid = dir->i_gid;
else if (dir->i_mode & S_ISGID) {
inode->i_gid = dir->i_gid;
if (S_ISDIR(mode))
mode |= S_ISGID;
} else
inode->i_gid = current->fsgid;
inode->i_mode = mode;
inode->i_ino = ino;
/* This is the optimal IO size (for stat), not the fs block size */
inode->i_blksize = PAGE_SIZE;
inode->i_blocks = 0;
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
memset(ei->i_data, 0, sizeof(ei->i_data));
ei->i_dir_start_lookup = 0;
ei->i_disksize = 0;
ei->i_flags = EXT3_I(dir)->i_flags & ~EXT3_INDEX_FL;
if (S_ISLNK(mode))
ei->i_flags &= ~(EXT3_IMMUTABLE_FL|EXT3_APPEND_FL);
/* dirsync only applies to directories */
if (!S_ISDIR(mode))
ei->i_flags &= ~EXT3_DIRSYNC_FL;
#ifdef EXT3_FRAGMENTS
ei->i_faddr = 0;
ei->i_frag_no = 0;
ei->i_frag_size = 0;
#endif
ei->i_file_acl = 0;
ei->i_dir_acl = 0;
ei->i_dtime = 0;
ei->i_block_alloc_info = NULL;
ei->i_block_group = group;
ext3_set_inode_flags(inode);
if (IS_DIRSYNC(inode))
handle->h_sync = 1;
insert_inode_hash(inode);
spin_lock(&sbi->s_next_gen_lock);
inode->i_generation = sbi->s_next_generation++;
spin_unlock(&sbi->s_next_gen_lock);
ei->i_state = EXT3_STATE_NEW;
ei->i_extra_isize =
(EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) ?
sizeof(struct ext3_inode) - EXT3_GOOD_OLD_INODE_SIZE : 0;
ret = inode;
if(DQUOT_ALLOC_INODE(inode)) {
err = -EDQUOT;
goto fail_drop;
}
err = ext3_init_acl(handle, inode, dir);
if (err)
goto fail_free_drop;
err = ext3_init_security(handle,inode, dir);
if (err)
goto fail_free_drop;
err = ext3_mark_inode_dirty(handle, inode);
if (err) {
ext3_std_error(sb, err);
goto fail_free_drop;
}
ext3_debug("allocating inode %lu\n", inode->i_ino);
goto really_out;
fail:
ext3_std_error(sb, err);
out:
iput(inode);
ret = ERR_PTR(err);
really_out:
brelse(bitmap_bh);
return ret;
fail_free_drop:
DQUOT_FREE_INODE(inode);
fail_drop:
DQUOT_DROP(inode);
inode->i_flags |= S_NOQUOTA;
inode->i_nlink = 0;
iput(inode);
brelse(bitmap_bh);
return ERR_PTR(err);
}
/*
* NOTE! When we get the inode, we're the only people
* that have access to it, and as such there are no
* race conditions we have to worry about. The inode
* is not on the hash-lists, and it cannot be reached
* through the filesystem because the directory entry
* has been deleted earlier.
*
* HOWEVER: we must make sure that we get no aliases,
* which means that we have to call "clear_inode()"
* _before_ we mark the inode not in use in the inode
* bitmaps. Otherwise a newly created file might use
* the same inode number (not actually the same pointer
* though), and then we'd have two inodes sharing the
* same inode number and space on the harddisk.
*/
void ext3_free_inode (handle_t *handle, struct inode * inode)
{
struct super_block * sb = inode->i_sb;
int is_directory;
unsigned long ino;
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *bh2;
unsigned long block_group;
unsigned long bit;
struct ext3_group_desc * gdp;
struct ext3_super_block * es;
struct ext3_sb_info *sbi;
int fatal = 0, err;
if (atomic_read(&inode->i_count) > 1) {
printk ("ext3_free_inode: inode has count=%d\n",
atomic_read(&inode->i_count));
return;
}
if (inode->i_nlink) {
printk ("ext3_free_inode: inode has nlink=%d\n",
inode->i_nlink);
return;
}
if (!sb) {
printk("ext3_free_inode: inode on nonexistent device\n");
return;
}
sbi = EXT3_SB(sb);
ino = inode->i_ino;
ext3_debug ("freeing inode %lu\n", ino);
trace_ext3_free_inode(inode);
is_directory = S_ISDIR(inode->i_mode);
es = EXT3_SB(sb)->s_es;
if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
ext3_error (sb, "ext3_free_inode",
"reserved or nonexistent inode %lu", ino);
goto error_return;
}
block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb);
bitmap_bh = read_inode_bitmap(sb, block_group);
if (!bitmap_bh)
goto error_return;
BUFFER_TRACE(bitmap_bh, "get_write_access");
fatal = ext3_journal_get_write_access(handle, bitmap_bh);
if (fatal)
goto error_return;
/* Ok, now we can actually update the inode bitmaps.. */
if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
bit, bitmap_bh->b_data))
ext3_error (sb, "ext3_free_inode",
"bit already cleared for inode %lu", ino);
else {
gdp = ext3_get_group_desc (sb, block_group, &bh2);
BUFFER_TRACE(bh2, "get_write_access");
fatal = ext3_journal_get_write_access(handle, bh2);
if (fatal) goto error_return;
if (gdp) {
spin_lock(sb_bgl_lock(sbi, block_group));
le16_add_cpu(&gdp->bg_free_inodes_count, 1);
if (is_directory)
le16_add_cpu(&gdp->bg_used_dirs_count, -1);
spin_unlock(sb_bgl_lock(sbi, block_group));
percpu_counter_inc(&sbi->s_freeinodes_counter);
if (is_directory)
percpu_counter_dec(&sbi->s_dirs_counter);
}
BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle, bh2);
if (!fatal) fatal = err;
}
BUFFER_TRACE(bitmap_bh, "call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle, bitmap_bh);
if (!fatal)
fatal = err;
error_return:
brelse(bitmap_bh);
ext3_std_error(sb, fatal);
}
/* Verify that we are loading a valid orphan from disk */
struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino)
{
unsigned long max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count);
unsigned long block_group;
int bit;
struct buffer_head *bitmap_bh;
struct inode *inode = NULL;
long err = -EIO;
/* Error cases - e2fsck has already cleaned up for us */
if (ino > max_ino) {
ext3_warning(sb, __func__,
"bad orphan ino %lu! e2fsck was run?", ino);
goto error;
}
block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb);
bitmap_bh = read_inode_bitmap(sb, block_group);
if (!bitmap_bh) {
ext3_warning(sb, __func__,
"inode bitmap error for orphan %lu", ino);
goto error;
}
/* Having the inode bit set should be a 100% indicator that this
* is a valid orphan (no e2fsck run on fs). Orphans also include
* inodes that were being truncated, so we can't check i_nlink==0.
*/
if (!ext3_test_bit(bit, bitmap_bh->b_data))
goto bad_orphan;
inode = ext3_iget(sb, ino);
if (IS_ERR(inode))
goto iget_failed;
/*
* If the orphans has i_nlinks > 0 then it should be able to be
* truncated, otherwise it won't be removed from the orphan list
* during processing and an infinite loop will result.
*/
if (inode->i_nlink && !ext3_can_truncate(inode))
goto bad_orphan;
if (NEXT_ORPHAN(inode) > max_ino)
goto bad_orphan;
brelse(bitmap_bh);
return inode;
iget_failed:
err = PTR_ERR(inode);
inode = NULL;
bad_orphan:
ext3_warning(sb, __func__,
"bad orphan inode %lu! e2fsck was run?", ino);
printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n",
bit, (unsigned long long)bitmap_bh->b_blocknr,
ext3_test_bit(bit, bitmap_bh->b_data));
printk(KERN_NOTICE "inode=%p\n", inode);
if (inode) {
printk(KERN_NOTICE "is_bad_inode(inode)=%d\n",
is_bad_inode(inode));
printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n",
NEXT_ORPHAN(inode));
printk(KERN_NOTICE "max_ino=%lu\n", max_ino);
printk(KERN_NOTICE "i_nlink=%u\n", inode->i_nlink);
/* Avoid freeing blocks if we got a bad deleted inode */
if (inode->i_nlink == 0)
inode->i_blocks = 0;
iput(inode);
}
brelse(bitmap_bh);
error:
return ERR_PTR(err);
}
/*
* load_inode_bitmap loads the inode bitmap for a blocks group
*
* It maintains a cache for the last bitmaps loaded. This cache is managed
* with a LRU algorithm.
*
* Notes:
* 1/ There is one cache per mounted file system.
* 2/ If the file system contains less than EXT3_MAX_GROUP_LOADED groups,
* this function reads the bitmap without maintaining a LRU cache.
*
* Return the slot used to store the bitmap, or a -ve error code.
*/
static int load_inode_bitmap (struct super_block * sb,
unsigned int block_group)
{
struct ext3_sb_info *sbi = EXT3_SB(sb);
unsigned long inode_bitmap_number;
struct buffer_head * inode_bitmap;
int i, j, retval = 0;
if (block_group >= sbi->s_groups_count)
ext3_panic (sb, "load_inode_bitmap",
"block_group >= groups_count - "
"block_group = %d, groups_count = %lu",
block_group, sbi->s_groups_count);
if (sbi->s_loaded_inode_bitmaps > 0 &&
sbi->s_inode_bitmap_number[0] == block_group &&
sbi->s_inode_bitmap[0] != NULL)
return 0;
if (sbi->s_groups_count <= EXT3_MAX_GROUP_LOADED) {
if (sbi->s_inode_bitmap[block_group]) {
if (sbi->s_inode_bitmap_number[block_group] !=
block_group)
ext3_panic(sb, "load_inode_bitmap",
"block_group != inode_bitmap_number");
return block_group;
}
retval = read_inode_bitmap(sb, block_group, block_group);
if (retval < 0)
return retval;
return block_group;
}
for (i = 0; i < sbi->s_loaded_inode_bitmaps &&
sbi->s_inode_bitmap_number[i] != block_group; i++)
/* do nothing */;
if (i < sbi->s_loaded_inode_bitmaps &&
sbi->s_inode_bitmap_number[i] == block_group) {
inode_bitmap_number = sbi->s_inode_bitmap_number[i];
inode_bitmap = sbi->s_inode_bitmap[i];
for (j = i; j > 0; j--) {
sbi->s_inode_bitmap_number[j] =
sbi->s_inode_bitmap_number[j - 1];
sbi->s_inode_bitmap[j] = sbi->s_inode_bitmap[j - 1];
}
sbi->s_inode_bitmap_number[0] = inode_bitmap_number;
sbi->s_inode_bitmap[0] = inode_bitmap;
/*
* There's still one special case here --- if inode_bitmap == 0
* then our last attempt to read the bitmap failed and we have
* just ended up caching that failure. Try again to read it.
*/
if (!inode_bitmap)
retval = read_inode_bitmap (sb, block_group, 0);
} else {
if (sbi->s_loaded_inode_bitmaps < EXT3_MAX_GROUP_LOADED)
sbi->s_loaded_inode_bitmaps++;
else
brelse(sbi->s_inode_bitmap[EXT3_MAX_GROUP_LOADED - 1]);
for (j = sbi->s_loaded_inode_bitmaps - 1; j > 0; j--) {
sbi->s_inode_bitmap_number[j] =
sbi->s_inode_bitmap_number[j - 1];
sbi->s_inode_bitmap[j] = sbi->s_inode_bitmap[j - 1];
}
retval = read_inode_bitmap (sb, block_group, 0);
}
return retval;
}
/*
* load_inode_bitmap loads the inode bitmap for a blocks group
*
* It maintains a cache for the last bitmaps loaded. This cache is managed
* with a LRU algorithm.
*
* Notes:
* 1/ There is one cache per mounted file system.
* 2/ If the file system contains less than EXT2_MAX_GROUP_LOADED groups,
* this function reads the bitmap without maintaining a LRU cache.
*/
static int
load_inode_bitmap(struct mount *mp, unsigned int block_group)
{
struct ext2_sb_info *sb = VFSTOEXT2(mp)->um_e2fs;
int i, j;
unsigned long inode_bitmap_number;
struct buf *inode_bitmap;
if (block_group >= sb->s_groups_count)
panic ("load_inode_bitmap:"
"block_group >= groups_count - "
"block_group = %d, groups_count = %lu",
block_group, sb->s_groups_count);
if (sb->s_loaded_inode_bitmaps > 0 &&
sb->s_inode_bitmap_number[0] == block_group)
return 0;
if (sb->s_groups_count <= EXT2_MAX_GROUP_LOADED) {
if (sb->s_inode_bitmap[block_group]) {
if (sb->s_inode_bitmap_number[block_group] !=
block_group)
panic ( "load_inode_bitmap:"
"block_group != inode_bitmap_number");
else
return block_group;
} else {
read_inode_bitmap (mp, block_group, block_group);
return block_group;
}
}
for (i = 0; i < sb->s_loaded_inode_bitmaps &&
sb->s_inode_bitmap_number[i] != block_group;
i++)
;
if (i < sb->s_loaded_inode_bitmaps &&
sb->s_inode_bitmap_number[i] == block_group) {
inode_bitmap_number = sb->s_inode_bitmap_number[i];
inode_bitmap = sb->s_inode_bitmap[i];
for (j = i; j > 0; j--) {
sb->s_inode_bitmap_number[j] =
sb->s_inode_bitmap_number[j - 1];
sb->s_inode_bitmap[j] =
sb->s_inode_bitmap[j - 1];
}
sb->s_inode_bitmap_number[0] = inode_bitmap_number;
sb->s_inode_bitmap[0] = inode_bitmap;
} else {
if (sb->s_loaded_inode_bitmaps < EXT2_MAX_GROUP_LOADED)
sb->s_loaded_inode_bitmaps++;
else
ULCK_BUF(sb->s_inode_bitmap[EXT2_MAX_GROUP_LOADED - 1])
for (j = sb->s_loaded_inode_bitmaps - 1; j > 0; j--) {
sb->s_inode_bitmap_number[j] =
sb->s_inode_bitmap_number[j - 1];
sb->s_inode_bitmap[j] =
sb->s_inode_bitmap[j - 1];
}
read_inode_bitmap (mp, block_group, 0);
}
return 0;
}
/*
* There are two policies for allocating an inode. If the new inode is
* a directory, then a forward search is made for a block group with both
* free space and a low directory-to-inode ratio; if that fails, then of
* the groups with above-average free space, that group with the fewest
* directories already is chosen.
*
* For other inodes, search forward from the parent directory's block
* group to find a free inode.
*/
struct inode *ext3_new_inode(handle_t *handle, struct inode * dir,
const struct qstr *qstr, umode_t mode)
{
struct super_block *sb;
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *bh2;
int group;
unsigned long ino = 0;
struct inode * inode;
struct ext3_group_desc * gdp = NULL;
struct ext3_super_block * es;
struct ext3_inode_info *ei;
struct ext3_sb_info *sbi;
int err = 0;
struct inode *ret;
int i;
/* Cannot create files in a deleted directory */
if (!dir || !dir->i_nlink)
return ERR_PTR(-EPERM);
sb = dir->i_sb;
trace_ext3_request_inode(dir, mode);
inode = new_inode(sb);
if (!inode)
return ERR_PTR(-ENOMEM);
ei = EXT3_I(inode);
sbi = EXT3_SB(sb);
es = sbi->s_es;
if (S_ISDIR(mode))
group = find_group_orlov(sb, dir);
else
group = find_group_other(sb, dir);
err = -ENOSPC;
if (group == -1)
goto out;
for (i = 0; i < sbi->s_groups_count; i++) {
err = -EIO;
gdp = ext3_get_group_desc(sb, group, &bh2);
if (!gdp)
goto fail;
brelse(bitmap_bh);
bitmap_bh = read_inode_bitmap(sb, group);
if (!bitmap_bh)
goto fail;
ino = 0;
repeat_in_this_group:
ino = ext3_find_next_zero_bit((unsigned long *)
bitmap_bh->b_data, EXT3_INODES_PER_GROUP(sb), ino);
if (ino < EXT3_INODES_PER_GROUP(sb)) {
BUFFER_TRACE(bitmap_bh, "get_write_access");
err = ext3_journal_get_write_access(handle, bitmap_bh);
if (err)
goto fail;
if (!ext3_set_bit_atomic(sb_bgl_lock(sbi, group),
ino, bitmap_bh->b_data)) {
/* we won it */
BUFFER_TRACE(bitmap_bh,
"call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle,
bitmap_bh);
if (err)
goto fail;
goto got;
}
/* we lost it */
journal_release_buffer(handle, bitmap_bh);
if (++ino < EXT3_INODES_PER_GROUP(sb))
goto repeat_in_this_group;
}
/*
* This case is possible in concurrent environment. It is very
* rare. We cannot repeat the find_group_xxx() call because
* that will simply return the same blockgroup, because the
* group descriptor metadata has not yet been updated.
* So we just go onto the next blockgroup.
*/
if (++group == sbi->s_groups_count)
group = 0;
}
err = -ENOSPC;
goto out;
got:
ino += group * EXT3_INODES_PER_GROUP(sb) + 1;
if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
ext3_error (sb, "ext3_new_inode",
"reserved inode or inode > inodes count - "
"block_group = %d, inode=%lu", group, ino);
err = -EIO;
goto fail;
}
BUFFER_TRACE(bh2, "get_write_access");
err = ext3_journal_get_write_access(handle, bh2);
if (err) goto fail;
spin_lock(sb_bgl_lock(sbi, group));
le16_add_cpu(&gdp->bg_free_inodes_count, -1);
if (S_ISDIR(mode)) {
le16_add_cpu(&gdp->bg_used_dirs_count, 1);
}
spin_unlock(sb_bgl_lock(sbi, group));
BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle, bh2);
if (err) goto fail;
percpu_counter_dec(&sbi->s_freeinodes_counter);
if (S_ISDIR(mode))
percpu_counter_inc(&sbi->s_dirs_counter);
if (test_opt(sb, GRPID)) {
inode->i_mode = mode;
inode->i_uid = current_fsuid();
inode->i_gid = dir->i_gid;
} else
inode_init_owner(inode, dir, mode);
inode->i_ino = ino;
/* This is the optimal IO size (for stat), not the fs block size */
inode->i_blocks = 0;
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
memset(ei->i_data, 0, sizeof(ei->i_data));
ei->i_dir_start_lookup = 0;
ei->i_disksize = 0;
ei->i_flags =
ext3_mask_flags(mode, EXT3_I(dir)->i_flags & EXT3_FL_INHERITED);
#ifdef EXT3_FRAGMENTS
ei->i_faddr = 0;
ei->i_frag_no = 0;
ei->i_frag_size = 0;
#endif
ei->i_file_acl = 0;
ei->i_dir_acl = 0;
ei->i_dtime = 0;
ei->i_block_alloc_info = NULL;
ei->i_block_group = group;
ext3_set_inode_flags(inode);
if (IS_DIRSYNC(inode))
handle->h_sync = 1;
if (insert_inode_locked(inode) < 0) {
/*
* Likely a bitmap corruption causing inode to be allocated
* twice.
*/
err = -EIO;
goto fail;
}
spin_lock(&sbi->s_next_gen_lock);
inode->i_generation = sbi->s_next_generation++;
spin_unlock(&sbi->s_next_gen_lock);
ei->i_state_flags = 0;
ext3_set_inode_state(inode, EXT3_STATE_NEW);
/* See comment in ext3_iget for explanation */
if (ino >= EXT3_FIRST_INO(sb) + 1 &&
EXT3_INODE_SIZE(sb) > EXT3_GOOD_OLD_INODE_SIZE) {
ei->i_extra_isize =
sizeof(struct ext3_inode) - EXT3_GOOD_OLD_INODE_SIZE;
} else {
ei->i_extra_isize = 0;
}
ret = inode;
dquot_initialize(inode);
err = dquot_alloc_inode(inode);
if (err)
goto fail_drop;
err = ext3_init_acl(handle, inode, dir);
if (err)
goto fail_free_drop;
err = ext3_init_security(handle, inode, dir, qstr);
if (err)
goto fail_free_drop;
err = ext3_mark_inode_dirty(handle, inode);
if (err) {
ext3_std_error(sb, err);
goto fail_free_drop;
}
ext3_debug("allocating inode %lu\n", inode->i_ino);
trace_ext3_allocate_inode(inode, dir, mode);
goto really_out;
fail:
ext3_std_error(sb, err);
out:
iput(inode);
ret = ERR_PTR(err);
really_out:
brelse(bitmap_bh);
return ret;
fail_free_drop:
dquot_free_inode(inode);
fail_drop:
dquot_drop(inode);
inode->i_flags |= S_NOQUOTA;
clear_nlink(inode);
unlock_new_inode(inode);
iput(inode);
brelse(bitmap_bh);
return ERR_PTR(err);
}
