/*
* 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, 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;
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)) {
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));
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;
inode->i_nlink = 0;
unlock_new_inode(inode);
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);
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);
}
static int find_group_orlov(struct super_block *sb, struct inode *parent)
{
int parent_group = EXT3_I(parent)->i_block_group;
struct ext3_sb_info *sbi = EXT3_SB(sb);
struct ext3_super_block *es = sbi->s_es;
int ngroups = sbi->s_groups_count;
int inodes_per_group = EXT3_INODES_PER_GROUP(sb);
unsigned int freei, avefreei;
ext3_fsblk_t freeb, avefreeb;
ext3_fsblk_t blocks_per_dir;
unsigned int ndirs;
int max_debt, max_dirs, min_inodes;
ext3_grpblk_t min_blocks;
int group = -1, i;
struct ext3_group_desc *desc;
freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter);
avefreei = freei / ngroups;
freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
avefreeb = freeb / ngroups;
ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter);
if ((parent == sb->s_root->d_inode) ||
(EXT3_I(parent)->i_flags & EXT3_TOPDIR_FL)) {
int best_ndir = inodes_per_group;
int best_group = -1;
get_random_bytes(&group, sizeof(group));
parent_group = (unsigned)group % ngroups;
for (i = 0; i < ngroups; i++) {
group = (parent_group + i) % ngroups;
desc = ext3_get_group_desc (sb, group, NULL);
if (!desc || !desc->bg_free_inodes_count)
continue;
if (le16_to_cpu(desc->bg_used_dirs_count) >= best_ndir)
continue;
if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei)
continue;
if (le16_to_cpu(desc->bg_free_blocks_count) < avefreeb)
continue;
best_group = group;
best_ndir = le16_to_cpu(desc->bg_used_dirs_count);
}
if (best_group >= 0)
return best_group;
goto fallback;
}
blocks_per_dir = (le32_to_cpu(es->s_blocks_count) - freeb) / ndirs;
max_dirs = ndirs / ngroups + inodes_per_group / 16;
min_inodes = avefreei - inodes_per_group / 4;
min_blocks = avefreeb - EXT3_BLOCKS_PER_GROUP(sb) / 4;
max_debt = EXT3_BLOCKS_PER_GROUP(sb) / max(blocks_per_dir, (ext3_fsblk_t)BLOCK_COST);
if (max_debt * INODE_COST > inodes_per_group)
max_debt = inodes_per_group / INODE_COST;
if (max_debt > 255)
max_debt = 255;
if (max_debt == 0)
max_debt = 1;
for (i = 0; i < ngroups; i++) {
group = (parent_group + i) % ngroups;
desc = ext3_get_group_desc (sb, group, NULL);
if (!desc || !desc->bg_free_inodes_count)
continue;
if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs)
continue;
if (le16_to_cpu(desc->bg_free_inodes_count) < min_inodes)
continue;
if (le16_to_cpu(desc->bg_free_blocks_count) < min_blocks)
continue;
return group;
}
fallback:
for (i = 0; i < ngroups; i++) {
group = (parent_group + i) % ngroups;
desc = ext3_get_group_desc (sb, group, NULL);
if (!desc || !desc->bg_free_inodes_count)
continue;
if (le16_to_cpu(desc->bg_free_inodes_count) >= avefreei)
return group;
}
if (avefreei) {
/*
* The free-inodes counter is approximate, and for really small
* filesystems the above test can fail to find any blockgroups
*/
avefreei = 0;
goto fallback;
}
return -1;
}
/* 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);
}
int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
unsigned long arg)
{
struct ext3_inode_info *ei = EXT3_I(inode);
unsigned int flags;
unsigned short rsv_window_size;
ext3_debug ("cmd = %u, arg = %lu\n", cmd, arg);
switch (cmd) {
case EXT3_IOC_GETFLAGS:
ext3_get_inode_flags(ei);
flags = ei->i_flags & EXT3_FL_USER_VISIBLE;
return put_user(flags, (int __user *) arg);
case EXT3_IOC_SETFLAGS: {
handle_t *handle = NULL;
int err;
struct ext3_iloc iloc;
unsigned int oldflags;
unsigned int jflag;
if (IS_RDONLY(inode))
return -EROFS;
if (!is_owner_or_cap(inode))
return -EACCES;
if (get_user(flags, (int __user *) arg))
return -EFAULT;
if (!S_ISDIR(inode->i_mode))
flags &= ~EXT3_DIRSYNC_FL;
mutex_lock(&inode->i_mutex);
/* Is it quota file? Do not allow user to mess with it */
if (IS_NOQUOTA(inode)) {
mutex_unlock(&inode->i_mutex);
return -EPERM;
}
oldflags = ei->i_flags;
/* The JOURNAL_DATA flag is modifiable only by root */
jflag = flags & EXT3_JOURNAL_DATA_FL;
/*
* The IMMUTABLE and APPEND_ONLY flags can only be changed by
* the relevant capability.
*
* This test looks nicer. Thanks to Pauline Middelink
*/
if ((flags ^ oldflags) & (EXT3_APPEND_FL | EXT3_IMMUTABLE_FL)) {
if (!capable(CAP_LINUX_IMMUTABLE)) {
mutex_unlock(&inode->i_mutex);
return -EPERM;
}
}
/*
* The JOURNAL_DATA flag can only be changed by
* the relevant capability.
*/
if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL)) {
if (!capable(CAP_SYS_RESOURCE)) {
mutex_unlock(&inode->i_mutex);
return -EPERM;
}
}
handle = ext3_journal_start(inode, 1);
if (IS_ERR(handle)) {
mutex_unlock(&inode->i_mutex);
return PTR_ERR(handle);
}
if (IS_SYNC(inode))
handle->h_sync = 1;
err = ext3_reserve_inode_write(handle, inode, &iloc);
if (err)
goto flags_err;
flags = flags & EXT3_FL_USER_MODIFIABLE;
flags |= oldflags & ~EXT3_FL_USER_MODIFIABLE;
ei->i_flags = flags;
ext3_set_inode_flags(inode);
inode->i_ctime = CURRENT_TIME_SEC;
err = ext3_mark_iloc_dirty(handle, inode, &iloc);
flags_err:
ext3_journal_stop(handle);
if (err) {
mutex_unlock(&inode->i_mutex);
return err;
}
if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL))
err = ext3_change_inode_journal_flag(inode, jflag);
mutex_unlock(&inode->i_mutex);
return err;
}
case EXT3_IOC_GETVERSION:
case EXT3_IOC_GETVERSION_OLD:
return put_user(inode->i_generation, (int __user *) arg);
case EXT3_IOC_SETVERSION:
case EXT3_IOC_SETVERSION_OLD: {
handle_t *handle;
struct ext3_iloc iloc;
__u32 generation;
int err;
if (!is_owner_or_cap(inode))
return -EPERM;
if (IS_RDONLY(inode))
return -EROFS;
if (get_user(generation, (int __user *) arg))
return -EFAULT;
handle = ext3_journal_start(inode, 1);
if (IS_ERR(handle))
return PTR_ERR(handle);
err = ext3_reserve_inode_write(handle, inode, &iloc);
if (err == 0) {
inode->i_ctime = CURRENT_TIME_SEC;
inode->i_generation = generation;
err = ext3_mark_iloc_dirty(handle, inode, &iloc);
}
ext3_journal_stop(handle);
return err;
}
#ifdef CONFIG_JBD_DEBUG
case EXT3_IOC_WAIT_FOR_READONLY:
/*
* This is racy - by the time we're woken up and running,
* the superblock could be released. And the module could
* have been unloaded. So sue me.
*
* Returns 1 if it slept, else zero.
*/
{
struct super_block *sb = inode->i_sb;
DECLARE_WAITQUEUE(wait, current);
int ret = 0;
set_current_state(TASK_INTERRUPTIBLE);
add_wait_queue(&EXT3_SB(sb)->ro_wait_queue, &wait);
if (timer_pending(&EXT3_SB(sb)->turn_ro_timer)) {
schedule();
ret = 1;
}
remove_wait_queue(&EXT3_SB(sb)->ro_wait_queue, &wait);
return ret;
}
#endif
case EXT3_IOC_GETRSVSZ:
if (test_opt(inode->i_sb, RESERVATION)
&& S_ISREG(inode->i_mode)
&& ei->i_block_alloc_info) {
rsv_window_size = ei->i_block_alloc_info->rsv_window_node.rsv_goal_size;
return put_user(rsv_window_size, (int __user *)arg);
}
return -ENOTTY;
case EXT3_IOC_SETRSVSZ: {
if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode))
return -ENOTTY;
if (IS_RDONLY(inode))
return -EROFS;
if (!is_owner_or_cap(inode))
return -EACCES;
if (get_user(rsv_window_size, (int __user *)arg))
return -EFAULT;
if (rsv_window_size > EXT3_MAX_RESERVE_BLOCKS)
rsv_window_size = EXT3_MAX_RESERVE_BLOCKS;
/*
* need to allocate reservation structure for this inode
* before set the window size
*/
mutex_lock(&ei->truncate_mutex);
if (!ei->i_block_alloc_info)
ext3_init_block_alloc_info(inode);
if (ei->i_block_alloc_info){
struct ext3_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node;
rsv->rsv_goal_size = rsv_window_size;
}
mutex_unlock(&ei->truncate_mutex);
return 0;
}
case EXT3_IOC_GROUP_EXTEND: {
ext3_fsblk_t n_blocks_count;
struct super_block *sb = inode->i_sb;
int err;
if (!capable(CAP_SYS_RESOURCE))
return -EPERM;
if (IS_RDONLY(inode))
return -EROFS;
if (get_user(n_blocks_count, (__u32 __user *)arg))
return -EFAULT;
err = ext3_group_extend(sb, EXT3_SB(sb)->s_es, n_blocks_count);
journal_lock_updates(EXT3_SB(sb)->s_journal);
journal_flush(EXT3_SB(sb)->s_journal);
journal_unlock_updates(EXT3_SB(sb)->s_journal);
return err;
}
case EXT3_IOC_GROUP_ADD: {
struct ext3_new_group_data input;
struct super_block *sb = inode->i_sb;
int err;
if (!capable(CAP_SYS_RESOURCE))
return -EPERM;
if (IS_RDONLY(inode))
return -EROFS;
if (copy_from_user(&input, (struct ext3_new_group_input __user *)arg,
sizeof(input)))
return -EFAULT;
err = ext3_group_add(sb, &input);
journal_lock_updates(EXT3_SB(sb)->s_journal);
journal_flush(EXT3_SB(sb)->s_journal);
journal_unlock_updates(EXT3_SB(sb)->s_journal);
return err;
}
default:
return -ENOTTY;
}
}
/*
* 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++) {
gdp = ext3_get_group_desc(sb, group, &bh2);
err = -EIO;
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)) {
int credits = 0;
BUFFER_TRACE(bitmap_bh, "get_write_access");
err = ext3_journal_get_write_access_credits(handle,
bitmap_bh, &credits);
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, credits);
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;
memset(ei->i_data, 0, sizeof(ei->i_data));
ei->i_next_alloc_block = 0;
ei->i_next_alloc_goal = 0;
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_rsv_window.rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
ei->i_rsv_window.rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
atomic_set(&ei->i_rsv_window.rsv_goal_size, EXT3_DEFAULT_RESERVE_BLOCKS);
atomic_set(&ei->i_rsv_window.rsv_alloc_hit, 0);
seqlock_init(&ei->i_rsv_window.rsv_seqlock);
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;
ret = inode;
if(DQUOT_ALLOC_INODE(inode)) {
DQUOT_DROP(inode);
err = -EDQUOT;
goto fail2;
}
err = ext3_init_acl(handle, inode, dir);
if (err) {
DQUOT_FREE_INODE(inode);
goto fail2;
}
err = ext3_mark_inode_dirty(handle, inode);
if (err) {
ext3_std_error(sb, err);
DQUOT_FREE_INODE(inode);
goto fail2;
}
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;
fail2:
inode->i_flags |= S_NOQUOTA;
inode->i_nlink = 0;
iput(inode);
brelse(bitmap_bh);
return ERR_PTR(err);
}
/*
* Called at inode eviction from icache
*/
void ext3_evict_inode (struct inode *inode)
{
struct ext3_inode_info *ei = EXT3_I(inode);
struct ext3_block_alloc_info *rsv;
handle_t *handle;
int want_delete = 0;
trace_ext3_evict_inode(inode);
if (!inode->i_nlink && !is_bad_inode(inode)) {
dquot_initialize(inode);
want_delete = 1;
}
/*
* When journalling data dirty buffers are tracked only in the journal.
* So although mm thinks everything is clean and ready for reaping the
* inode might still have some pages to write in the running
* transaction or waiting to be checkpointed. Thus calling
* journal_invalidatepage() (via truncate_inode_pages()) to discard
* these buffers can cause data loss. Also even if we did not discard
* these buffers, we would have no way to find them after the inode
* is reaped and thus user could see stale data if he tries to read
* them before the transaction is checkpointed. So be careful and
* force everything to disk here... We use ei->i_datasync_tid to
* store the newest transaction containing inode's data.
*
* Note that directories do not have this problem because they don't
* use page cache.
*
* The s_journal check handles the case when ext3_get_journal() fails
* and puts the journal inode.
*/
if (inode->i_nlink && ext3_should_journal_data(inode) &&
EXT3_SB(inode->i_sb)->s_journal &&
(S_ISLNK(inode->i_mode) || S_ISREG(inode->i_mode)) &&
inode->i_ino != EXT3_JOURNAL_INO) {
tid_t commit_tid = atomic_read(&ei->i_datasync_tid);
journal_t *journal = EXT3_SB(inode->i_sb)->s_journal;
log_start_commit(journal, commit_tid);
log_wait_commit(journal, commit_tid);
filemap_write_and_wait(&inode->i_data);
}
truncate_inode_pages(&inode->i_data, 0);
ext3_discard_reservation(inode);
rsv = ei->i_block_alloc_info;
ei->i_block_alloc_info = NULL;
if (unlikely(rsv))
kfree(rsv);
if (!want_delete)
goto no_delete;
handle = start_transaction(inode);
if (IS_ERR(handle)) {
/*
* If we're going to skip the normal cleanup, we still need to
* make sure that the in-core orphan linked list is properly
* cleaned up.
*/
ext3_orphan_del(NULL, inode);
goto no_delete;
}
if (IS_SYNC(inode))
handle->h_sync = 1;
inode->i_size = 0;
if (inode->i_blocks)
ext3_truncate(inode);
/*
* Kill off the orphan record created when the inode lost the last
* link. Note that ext3_orphan_del() has to be able to cope with the
* deletion of a non-existent orphan - ext3_truncate() could
* have removed the record.
*/
ext3_orphan_del(handle, inode);
ei->i_dtime = get_seconds();
/*
* One subtle ordering requirement: if anything has gone wrong
* (transaction abort, IO errors, whatever), then we can still
* do these next steps (the fs will already have been marked as
* having errors), but we can't free the inode if the mark_dirty
* fails.
*/
if (ext3_mark_inode_dirty(handle, inode)) {
/* If that failed, just dquot_drop() and be done with that */
dquot_drop(inode);
clear_inode(inode);
} else {
ext3_xattr_delete_inode(handle, inode);
dquot_free_inode(inode);
dquot_drop(inode);
clear_inode(inode);
ext3_free_inode(handle, inode);
}
ext3_journal_stop(handle);
return;
no_delete:
clear_inode(inode);
dquot_drop(inode);
}
int ext3_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
{
struct timespec before,after;
printk(KERN_DEBUG"fsync start\n");
getnstimeofday(&before);
struct inode *inode = file->f_mapping->host;
struct ext3_inode_info *ei = EXT3_I(inode);
journal_t *journal = EXT3_SB(inode->i_sb)->s_journal;
int ret, needs_barrier = 0;
tid_t commit_tid;
trace_ext3_sync_file_enter(file, datasync);
if (inode->i_sb->s_flags & MS_RDONLY) {
/* Make sure that we read updated state */
smp_rmb();
if (EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)
{
getnstimeofday(&after);
printk(KERN_DEBUG"fsync time:%.01f us",after.tv_nsec-before.tv_nsec);
return -EROFS;
}
getnstimeofday(&after);
printk(KERN_DEBUG"fsync time:%.01f us",after.tv_nsec-before.tv_nsec);
return 0;
}
ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
if (ret)
goto out;
J_ASSERT(ext3_journal_current_handle() == NULL);
/*
* data=writeback,ordered:
* The caller's filemap_fdatawrite()/wait will sync the data.
* Metadata is in the journal, we wait for a proper transaction
* to commit here.
*
* data=journal:
* filemap_fdatawrite won't do anything (the buffers are clean).
* ext3_force_commit will write the file data into the journal and
* will wait on that.
* filemap_fdatawait() will encounter a ton of newly-dirtied pages
* (they were dirtied by commit). But that's OK - the blocks are
* safe in-journal, which is all fsync() needs to ensure.
*/
if (ext3_should_journal_data(inode)) {
ret = ext3_force_commit(inode->i_sb);
goto out;
}
if (datasync)
commit_tid = atomic_read(&ei->i_datasync_tid);
else
commit_tid = atomic_read(&ei->i_sync_tid);
if (test_opt(inode->i_sb, BARRIER) &&
!journal_trans_will_send_data_barrier(journal, commit_tid))
needs_barrier = 1;
log_start_commit(journal, commit_tid);
ret = log_wait_commit(journal, commit_tid);
/*
* In case we didn't commit a transaction, we have to flush
* disk caches manually so that data really is on persistent
* storage
*/
if (needs_barrier) {
int err;
err = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
if (!ret)
ret = err;
}
out:
getnstimeofday(&after);
printk(KERN_DEBUG"fsync time:%.01f us",after.tv_nsec-before.tv_nsec);
trace_ext3_sync_file_exit(inode, ret);
return ret;
}
