static int ocfs2_truncate_for_delete(struct ocfs2_super *osb,
struct inode *inode,
struct buffer_head *fe_bh)
{
int status = 0;
struct ocfs2_dinode *fe;
handle_t *handle = NULL;
mlog_entry_void();
fe = (struct ocfs2_dinode *) fe_bh->b_data;
/*
* This check will also skip truncate of inodes with inline
* data and fast symlinks.
*/
if (fe->i_clusters) {
if (ocfs2_should_order_data(inode))
ocfs2_begin_ordered_truncate(inode, 0);
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
handle = NULL;
mlog_errno(status);
goto out;
}
status = ocfs2_journal_access_di(handle, INODE_CACHE(inode),
fe_bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto out;
}
i_size_write(inode, 0);
status = ocfs2_mark_inode_dirty(handle, inode, fe_bh);
if (status < 0) {
mlog_errno(status);
goto out;
}
ocfs2_commit_trans(osb, handle);
handle = NULL;
status = ocfs2_commit_truncate(osb, inode, fe_bh);
if (status < 0) {
mlog_errno(status);
goto out;
}
}
out:
if (handle)
ocfs2_commit_trans(osb, handle);
mlog_exit(status);
return status;
}
handle_t *ocfs2_start_walk_page_trans(struct inode *inode,
struct page *page,
unsigned from,
unsigned to)
{
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
handle_t *handle = NULL;
int ret = 0;
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (!handle) {
ret = -ENOMEM;
mlog_errno(ret);
goto out;
}
if (ocfs2_should_order_data(inode)) {
ret = walk_page_buffers(handle,
page_buffers(page),
from, to, NULL,
ocfs2_journal_dirty_data);
if (ret < 0)
mlog_errno(ret);
}
out:
if (ret) {
if (handle)
ocfs2_commit_trans(osb, handle);
handle = ERR_PTR(ret);
}
return handle;
}
handle_t *ocfs2_start_walk_page_trans(struct inode *inode,
struct page *page,
unsigned from,
unsigned to)
{
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
handle_t *handle;
int ret = 0;
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (IS_ERR(handle)) {
ret = -ENOMEM;
mlog_errno(ret);
goto out;
}
if (ocfs2_should_order_data(inode)) {
ret = ocfs2_jbd2_file_inode(handle, inode);
if (ret < 0)
mlog_errno(ret);
}
out:
if (ret) {
if (!IS_ERR(handle))
ocfs2_commit_trans(osb, handle);
handle = ERR_PTR(ret);
}
return handle;
}
int ocfs2_update_inode_atime(struct inode *inode,
struct buffer_head *bh)
{
int ret;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
handle_t *handle;
mlog_entry_void();
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (handle == NULL) {
ret = -ENOMEM;
mlog_errno(ret);
goto out;
}
inode->i_atime = CURRENT_TIME;
ret = ocfs2_mark_inode_dirty(handle, inode, bh);
if (ret < 0)
mlog_errno(ret);
ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
out:
mlog_exit(ret);
return ret;
}
/*
* Step 2: By now, we've completed the journal recovery, we've stamped
* a clean local alloc on disk and dropped the node out of the
* recovery map. Dlm locks will no longer stall, so lets clear out the
* main bitmap.
*/
int ocfs2_complete_local_alloc_recovery(struct ocfs2_super *osb,
struct ocfs2_dinode *alloc)
{
int status;
handle_t *handle;
struct buffer_head *main_bm_bh = NULL;
struct inode *main_bm_inode;
main_bm_inode = ocfs2_get_system_file_inode(osb,
GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT);
if (!main_bm_inode) {
status = -EINVAL;
mlog_errno(status);
goto out;
}
inode_lock(main_bm_inode);
status = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1);
if (status < 0) {
mlog_errno(status);
goto out_mutex;
}
handle = ocfs2_start_trans(osb, OCFS2_WINDOW_MOVE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
handle = NULL;
mlog_errno(status);
goto out_unlock;
}
/* we want the bitmap change to be recorded on disk asap */
handle->h_sync = 1;
status = ocfs2_sync_local_to_main(osb, handle, alloc,
main_bm_inode, main_bm_bh);
if (status < 0)
mlog_errno(status);
ocfs2_commit_trans(osb, handle);
out_unlock:
ocfs2_inode_unlock(main_bm_inode, 1);
out_mutex:
inode_unlock(main_bm_inode);
brelse(main_bm_bh);
iput(main_bm_inode);
out:
if (!status)
ocfs2_init_steal_slots(osb);
if (status)
mlog_errno(status);
return status;
}
/* Some parts of this taken from generic_cont_expand, which turned out
* to be too fragile to do exactly what we need without us having to
* worry about recursive locking in ->write_begin() and ->write_end(). */
static int ocfs2_write_zero_page(struct inode *inode,
u64 size)
{
struct address_space *mapping = inode->i_mapping;
struct page *page;
unsigned long index;
unsigned int offset;
handle_t *handle = NULL;
int ret;
offset = (size & (PAGE_CACHE_SIZE-1)); /* Within page */
/* ugh. in prepare/commit_write, if from==to==start of block, we
** skip the prepare. make sure we never send an offset for the start
** of a block
*/
if ((offset & (inode->i_sb->s_blocksize - 1)) == 0) {
offset++;
}
index = size >> PAGE_CACHE_SHIFT;
page = grab_cache_page(mapping, index);
if (!page) {
ret = -ENOMEM;
mlog_errno(ret);
goto out;
}
ret = ocfs2_prepare_write_nolock(inode, page, offset, offset);
if (ret < 0) {
mlog_errno(ret);
goto out_unlock;
}
if (ocfs2_should_order_data(inode)) {
handle = ocfs2_start_walk_page_trans(inode, page, offset,
offset);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
handle = NULL;
goto out_unlock;
}
}
/* must not update i_size! */
ret = block_commit_write(page, offset, offset);
if (ret < 0)
mlog_errno(ret);
else
ret = 0;
if (handle)
ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
out_unlock:
unlock_page(page);
page_cache_release(page);
out:
return ret;
}
static int ocfs2_orphan_for_truncate(struct ocfs2_super *osb,
struct inode *inode,
struct buffer_head *fe_bh,
u64 new_i_size)
{
int status;
handle_t *handle;
struct ocfs2_dinode *di;
mlog_entry_void();
/* TODO: This needs to actually orphan the inode in this
* transaction. */
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
goto out;
}
status = ocfs2_journal_access(handle, inode, fe_bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto out_commit;
}
/*
* Do this before setting i_size.
*/
status = ocfs2_zero_tail_for_truncate(inode, handle, new_i_size);
if (status) {
mlog_errno(status);
goto out_commit;
}
i_size_write(inode, new_i_size);
inode->i_blocks = ocfs2_align_bytes_to_sectors(new_i_size);
inode->i_ctime = inode->i_mtime = CURRENT_TIME;
di = (struct ocfs2_dinode *) fe_bh->b_data;
di->i_size = cpu_to_le64(new_i_size);
di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec);
di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
status = ocfs2_journal_dirty(handle, fe_bh);
if (status < 0)
mlog_errno(status);
out_commit:
ocfs2_commit_trans(osb, handle);
out:
mlog_exit(status);
return status;
}
/*
* Helper function to set i_mode in memory and disk. Some call paths
* will not have di_bh or a journal handle to pass, in which case it
* will create it's own.
*/
static int ocfs2_acl_set_mode(struct inode *inode, struct buffer_head *di_bh,
handle_t *handle, umode_t new_mode)
{
int ret, commit_handle = 0;
struct ocfs2_dinode *di;
if (di_bh == NULL) {
ret = ocfs2_read_inode_block(inode, &di_bh);
if (ret) {
mlog_errno(ret);
goto out;
}
} else
get_bh(di_bh);
if (handle == NULL) {
handle = ocfs2_start_trans(OCFS2_SB(inode->i_sb),
OCFS2_INODE_UPDATE_CREDITS);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
mlog_errno(ret);
goto out_brelse;
}
commit_handle = 1;
}
di = (struct ocfs2_dinode *)di_bh->b_data;
ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), di_bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (ret) {
mlog_errno(ret);
goto out_commit;
}
inode->i_mode = new_mode;
inode->i_ctime = current_time(inode);
di->i_mode = cpu_to_le16(inode->i_mode);
di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 0);
ocfs2_journal_dirty(handle, di_bh);
out_commit:
if (commit_handle)
ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
out_brelse:
brelse(di_bh);
out:
return ret;
}
static int ocfs2_write_remove_suid(struct inode *inode)
{
int ret;
struct buffer_head *bh = NULL;
struct ocfs2_inode_info *oi = OCFS2_I(inode);
handle_t *handle;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct ocfs2_dinode *di;
mlog_entry("(Inode %llu, mode 0%o)\n",
(unsigned long long)oi->ip_blkno, inode->i_mode);
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (handle == NULL) {
ret = -ENOMEM;
mlog_errno(ret);
goto out;
}
ret = ocfs2_read_block(osb, oi->ip_blkno, &bh, OCFS2_BH_CACHED, inode);
if (ret < 0) {
mlog_errno(ret);
goto out_trans;
}
ret = ocfs2_journal_access(handle, inode, bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (ret < 0) {
mlog_errno(ret);
goto out_bh;
}
inode->i_mode &= ~S_ISUID;
if ((inode->i_mode & S_ISGID) && (inode->i_mode & S_IXGRP))
inode->i_mode &= ~S_ISGID;
di = (struct ocfs2_dinode *) bh->b_data;
di->i_mode = cpu_to_le16(inode->i_mode);
ret = ocfs2_journal_dirty(handle, bh);
if (ret < 0)
mlog_errno(ret);
out_bh:
brelse(bh);
out_trans:
ocfs2_commit_trans(osb, handle);
out:
mlog_exit(ret);
return ret;
}
static int ocfs2_simple_size_update(struct inode *inode,
struct buffer_head *di_bh,
u64 new_i_size)
{
int ret;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
handle_t *handle = NULL;
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (handle == NULL) {
ret = -ENOMEM;
mlog_errno(ret);
goto out;
}
ret = ocfs2_set_inode_size(handle, inode, di_bh,
new_i_size);
if (ret < 0)
mlog_errno(ret);
ocfs2_commit_trans(osb, handle);
out:
return ret;
}
static int ocfs2_extend_allocation(struct inode *inode,
u32 clusters_to_add)
{
int status = 0;
int restart_func = 0;
int drop_alloc_sem = 0;
int credits;
u32 prev_clusters, logical_start;
struct buffer_head *bh = NULL;
struct ocfs2_dinode *fe = NULL;
handle_t *handle = NULL;
struct ocfs2_alloc_context *data_ac = NULL;
struct ocfs2_alloc_context *meta_ac = NULL;
enum ocfs2_alloc_restarted why;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
mlog_entry("(clusters_to_add = %u)\n", clusters_to_add);
/*
* This function only exists for file systems which don't
* support holes.
*/
BUG_ON(ocfs2_sparse_alloc(osb));
status = ocfs2_read_block(osb, OCFS2_I(inode)->ip_blkno, &bh,
OCFS2_BH_CACHED, inode);
if (status < 0) {
mlog_errno(status);
goto leave;
}
fe = (struct ocfs2_dinode *) bh->b_data;
if (!OCFS2_IS_VALID_DINODE(fe)) {
OCFS2_RO_ON_INVALID_DINODE(inode->i_sb, fe);
status = -EIO;
goto leave;
}
logical_start = OCFS2_I(inode)->ip_clusters;
restart_all:
BUG_ON(le32_to_cpu(fe->i_clusters) != OCFS2_I(inode)->ip_clusters);
/* blocks peope in read/write from reading our allocation
* until we're done changing it. We depend on i_mutex to block
* other extend/truncate calls while we're here. Ordering wrt
* start_trans is important here -- always do it before! */
down_write(&OCFS2_I(inode)->ip_alloc_sem);
drop_alloc_sem = 1;
status = ocfs2_lock_allocators(inode, fe, clusters_to_add, &data_ac,
&meta_ac);
if (status) {
mlog_errno(status);
goto leave;
}
credits = ocfs2_calc_extend_credits(osb->sb, fe, clusters_to_add);
handle = ocfs2_start_trans(osb, credits);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
handle = NULL;
mlog_errno(status);
goto leave;
}
restarted_transaction:
/* reserve a write to the file entry early on - that we if we
* run out of credits in the allocation path, we can still
* update i_size. */
status = ocfs2_journal_access(handle, inode, bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto leave;
}
prev_clusters = OCFS2_I(inode)->ip_clusters;
status = ocfs2_do_extend_allocation(osb,
inode,
&logical_start,
clusters_to_add,
bh,
handle,
data_ac,
meta_ac,
&why);
if ((status < 0) && (status != -EAGAIN)) {
if (status != -ENOSPC)
mlog_errno(status);
goto leave;
}
status = ocfs2_journal_dirty(handle, bh);
if (status < 0) {
mlog_errno(status);
goto leave;
}
spin_lock(&OCFS2_I(inode)->ip_lock);
clusters_to_add -= (OCFS2_I(inode)->ip_clusters - prev_clusters);
spin_unlock(&OCFS2_I(inode)->ip_lock);
if (why != RESTART_NONE && clusters_to_add) {
if (why == RESTART_META) {
mlog(0, "restarting function.\n");
restart_func = 1;
} else {
BUG_ON(why != RESTART_TRANS);
mlog(0, "restarting transaction.\n");
/* TODO: This can be more intelligent. */
credits = ocfs2_calc_extend_credits(osb->sb,
fe,
clusters_to_add);
status = ocfs2_extend_trans(handle, credits);
if (status < 0) {
/* handle still has to be committed at
* this point. */
status = -ENOMEM;
mlog_errno(status);
goto leave;
}
goto restarted_transaction;
}
}
mlog(0, "fe: i_clusters = %u, i_size=%llu\n",
le32_to_cpu(fe->i_clusters),
(unsigned long long)le64_to_cpu(fe->i_size));
mlog(0, "inode: ip_clusters=%u, i_size=%lld\n",
OCFS2_I(inode)->ip_clusters, i_size_read(inode));
leave:
if (drop_alloc_sem) {
up_write(&OCFS2_I(inode)->ip_alloc_sem);
drop_alloc_sem = 0;
}
if (handle) {
ocfs2_commit_trans(osb, handle);
handle = NULL;
}
if (data_ac) {
ocfs2_free_alloc_context(data_ac);
data_ac = NULL;
}
if (meta_ac) {
ocfs2_free_alloc_context(meta_ac);
meta_ac = NULL;
}
if ((!status) && restart_func) {
restart_func = 0;
goto restart_all;
}
if (bh) {
brelse(bh);
bh = NULL;
}
mlog_exit(status);
return status;
}
/*
* return any unused bits to the bitmap and write out a clean
* local_alloc.
*
* local_alloc_bh is optional. If not passed, we will simply use the
* one off osb. If you do pass it however, be warned that it *will* be
* returned brelse'd and NULL'd out.*/
void ocfs2_shutdown_local_alloc(struct ocfs2_super *osb)
{
int status;
handle_t *handle;
struct inode *local_alloc_inode = NULL;
struct buffer_head *bh = NULL;
struct buffer_head *main_bm_bh = NULL;
struct inode *main_bm_inode = NULL;
struct ocfs2_dinode *alloc_copy = NULL;
struct ocfs2_dinode *alloc = NULL;
cancel_delayed_work(&osb->la_enable_wq);
flush_workqueue(osb->ocfs2_wq);
if (osb->local_alloc_state == OCFS2_LA_UNUSED)
goto out;
local_alloc_inode =
ocfs2_get_system_file_inode(osb,
LOCAL_ALLOC_SYSTEM_INODE,
osb->slot_num);
if (!local_alloc_inode) {
status = -ENOENT;
mlog_errno(status);
goto out;
}
osb->local_alloc_state = OCFS2_LA_DISABLED;
ocfs2_resmap_uninit(&osb->osb_la_resmap);
main_bm_inode = ocfs2_get_system_file_inode(osb,
GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT);
if (!main_bm_inode) {
status = -EINVAL;
mlog_errno(status);
goto out;
}
inode_lock(main_bm_inode);
status = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1);
if (status < 0) {
mlog_errno(status);
goto out_mutex;
}
/* WINDOW_MOVE_CREDITS is a bit heavy... */
handle = ocfs2_start_trans(osb, OCFS2_WINDOW_MOVE_CREDITS);
if (IS_ERR(handle)) {
mlog_errno(PTR_ERR(handle));
handle = NULL;
goto out_unlock;
}
bh = osb->local_alloc_bh;
alloc = (struct ocfs2_dinode *) bh->b_data;
alloc_copy = kmalloc(bh->b_size, GFP_NOFS);
if (!alloc_copy) {
status = -ENOMEM;
goto out_commit;
}
memcpy(alloc_copy, alloc, bh->b_size);
status = ocfs2_journal_access_di(handle, INODE_CACHE(local_alloc_inode),
bh, OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto out_commit;
}
ocfs2_clear_local_alloc(alloc);
ocfs2_journal_dirty(handle, bh);
brelse(bh);
osb->local_alloc_bh = NULL;
osb->local_alloc_state = OCFS2_LA_UNUSED;
status = ocfs2_sync_local_to_main(osb, handle, alloc_copy,
main_bm_inode, main_bm_bh);
if (status < 0)
mlog_errno(status);
out_commit:
ocfs2_commit_trans(osb, handle);
out_unlock:
brelse(main_bm_bh);
ocfs2_inode_unlock(main_bm_inode, 1);
out_mutex:
inode_unlock(main_bm_inode);
iput(main_bm_inode);
out:
iput(local_alloc_inode);
kfree(alloc_copy);
}
static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags,
unsigned mask)
{
struct ocfs2_inode_info *ocfs2_inode = OCFS2_I(inode);
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
handle_t *handle = NULL;
struct buffer_head *bh = NULL;
unsigned oldflags;
int status;
mutex_lock(&inode->i_mutex);
status = ocfs2_inode_lock(inode, &bh, 1);
if (status < 0) {
mlog_errno(status);
goto bail;
}
status = -EACCES;
if (!inode_owner_or_capable(inode))
goto bail_unlock;
if (!S_ISDIR(inode->i_mode))
flags &= ~OCFS2_DIRSYNC_FL;
oldflags = ocfs2_inode->ip_attr;
flags = flags & mask;
flags |= oldflags & ~mask;
/*
* The IMMUTABLE and APPEND_ONLY flags can only be changed by
* the relevant capability.
*/
status = -EPERM;
if ((oldflags & OCFS2_IMMUTABLE_FL) || ((flags ^ oldflags) &
(OCFS2_APPEND_FL | OCFS2_IMMUTABLE_FL))) {
if (!capable(CAP_LINUX_IMMUTABLE))
goto bail_unlock;
}
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
goto bail_unlock;
}
ocfs2_inode->ip_attr = flags;
ocfs2_set_inode_flags(inode);
status = ocfs2_mark_inode_dirty(handle, inode, bh);
if (status < 0)
mlog_errno(status);
ocfs2_commit_trans(osb, handle);
bail_unlock:
ocfs2_inode_unlock(inode, 1);
bail:
mutex_unlock(&inode->i_mutex);
brelse(bh);
return status;
}
static int ocfs2_remove_inode(struct inode *inode,
struct buffer_head *di_bh,
struct inode *orphan_dir_inode,
struct buffer_head *orphan_dir_bh)
{
int status;
struct inode *inode_alloc_inode = NULL;
struct buffer_head *inode_alloc_bh = NULL;
handle_t *handle;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct ocfs2_dinode *di = (struct ocfs2_dinode *) di_bh->b_data;
inode_alloc_inode =
ocfs2_get_system_file_inode(osb, INODE_ALLOC_SYSTEM_INODE,
le16_to_cpu(di->i_suballoc_slot));
if (!inode_alloc_inode) {
status = -EEXIST;
mlog_errno(status);
goto bail;
}
mutex_lock(&inode_alloc_inode->i_mutex);
status = ocfs2_inode_lock(inode_alloc_inode, &inode_alloc_bh, 1);
if (status < 0) {
mutex_unlock(&inode_alloc_inode->i_mutex);
mlog_errno(status);
goto bail;
}
handle = ocfs2_start_trans(osb, OCFS2_DELETE_INODE_CREDITS +
ocfs2_quota_trans_credits(inode->i_sb));
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
goto bail_unlock;
}
if (!(OCFS2_I(inode)->ip_flags & OCFS2_INODE_SKIP_ORPHAN_DIR)) {
status = ocfs2_orphan_del(osb, handle, orphan_dir_inode, inode,
orphan_dir_bh);
if (status < 0) {
mlog_errno(status);
goto bail_commit;
}
}
/* set the inodes dtime */
status = ocfs2_journal_access_di(handle, INODE_CACHE(inode), di_bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto bail_commit;
}
di->i_dtime = cpu_to_le64(CURRENT_TIME.tv_sec);
di->i_flags &= cpu_to_le32(~(OCFS2_VALID_FL | OCFS2_ORPHANED_FL));
ocfs2_journal_dirty(handle, di_bh);
ocfs2_remove_from_cache(INODE_CACHE(inode), di_bh);
dquot_free_inode(inode);
status = ocfs2_free_dinode(handle, inode_alloc_inode,
inode_alloc_bh, di);
if (status < 0)
mlog_errno(status);
bail_commit:
ocfs2_commit_trans(osb, handle);
bail_unlock:
ocfs2_inode_unlock(inode_alloc_inode, 1);
mutex_unlock(&inode_alloc_inode->i_mutex);
brelse(inode_alloc_bh);
bail:
iput(inode_alloc_inode);
return status;
}
/*
* return any unused bits to the bitmap and write out a clean
* local_alloc.
*
* local_alloc_bh is optional. If not passed, we will simply use the
* one off osb. If you do pass it however, be warned that it *will* be
* returned brelse'd and NULL'd out.*/
void ocfs2_shutdown_local_alloc(struct ocfs2_super *osb)
{
int status;
struct ocfs2_journal_handle *handle = NULL;
struct inode *local_alloc_inode = NULL;
struct buffer_head *bh = NULL;
struct buffer_head *main_bm_bh = NULL;
struct inode *main_bm_inode = NULL;
struct ocfs2_dinode *alloc_copy = NULL;
struct ocfs2_dinode *alloc = NULL;
mlog_entry_void();
if (osb->local_alloc_state == OCFS2_LA_UNUSED)
goto bail;
local_alloc_inode =
ocfs2_get_system_file_inode(osb,
LOCAL_ALLOC_SYSTEM_INODE,
osb->slot_num);
if (!local_alloc_inode) {
status = -ENOENT;
mlog_errno(status);
goto bail;
}
osb->local_alloc_state = OCFS2_LA_DISABLED;
handle = ocfs2_alloc_handle(osb);
if (!handle) {
status = -ENOMEM;
mlog_errno(status);
goto bail;
}
main_bm_inode = ocfs2_get_system_file_inode(osb,
GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT);
if (!main_bm_inode) {
status = -EINVAL;
mlog_errno(status);
goto bail;
}
ocfs2_handle_add_inode(handle, main_bm_inode);
status = ocfs2_meta_lock(main_bm_inode, handle, &main_bm_bh, 1);
if (status < 0) {
mlog_errno(status);
goto bail;
}
/* WINDOW_MOVE_CREDITS is a bit heavy... */
handle = ocfs2_start_trans(osb, handle, OCFS2_WINDOW_MOVE_CREDITS);
if (IS_ERR(handle)) {
mlog_errno(PTR_ERR(handle));
handle = NULL;
goto bail;
}
bh = osb->local_alloc_bh;
alloc = (struct ocfs2_dinode *) bh->b_data;
alloc_copy = kmalloc(bh->b_size, GFP_KERNEL);
if (!alloc_copy) {
status = -ENOMEM;
goto bail;
}
memcpy(alloc_copy, alloc, bh->b_size);
status = ocfs2_journal_access(handle, local_alloc_inode, bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto bail;
}
ocfs2_clear_local_alloc(alloc);
status = ocfs2_journal_dirty(handle, bh);
if (status < 0) {
mlog_errno(status);
goto bail;
}
brelse(bh);
osb->local_alloc_bh = NULL;
osb->local_alloc_state = OCFS2_LA_UNUSED;
status = ocfs2_sync_local_to_main(osb, handle, alloc_copy,
main_bm_inode, main_bm_bh);
if (status < 0)
mlog_errno(status);
bail:
if (handle)
ocfs2_commit_trans(handle);
if (main_bm_bh)
brelse(main_bm_bh);
if (main_bm_inode)
iput(main_bm_inode);
if (local_alloc_inode)
iput(local_alloc_inode);
if (alloc_copy)
kfree(alloc_copy);
mlog_exit_void();
}
/*
* Using one journal handle to guarantee the data consistency in case
* crash happens anywhere.
*
* XXX: defrag can end up with finishing partial extent as requested,
* due to not enough contiguous clusters can be found in allocator.
*/
static int ocfs2_defrag_extent(struct ocfs2_move_extents_context *context,
u32 cpos, u32 phys_cpos, u32 *len, int ext_flags)
{
int ret, credits = 0, extra_blocks = 0, partial = context->partial;
handle_t *handle;
struct inode *inode = context->inode;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct inode *tl_inode = osb->osb_tl_inode;
struct ocfs2_refcount_tree *ref_tree = NULL;
u32 new_phys_cpos, new_len;
u64 phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, phys_cpos);
if ((ext_flags & OCFS2_EXT_REFCOUNTED) && *len) {
BUG_ON(!(OCFS2_I(inode)->ip_dyn_features &
OCFS2_HAS_REFCOUNT_FL));
BUG_ON(!context->refcount_loc);
ret = ocfs2_lock_refcount_tree(osb, context->refcount_loc, 1,
&ref_tree, NULL);
if (ret) {
mlog_errno(ret);
return ret;
}
ret = ocfs2_prepare_refcount_change_for_del(inode,
context->refcount_loc,
phys_blkno,
*len,
&credits,
&extra_blocks);
if (ret) {
mlog_errno(ret);
goto out;
}
}
ret = ocfs2_lock_allocators_move_extents(inode, &context->et, *len, 1,
&context->meta_ac,
&context->data_ac,
extra_blocks, &credits);
if (ret) {
mlog_errno(ret);
goto out;
}
/*
* should be using allocation reservation strategy there?
*
* if (context->data_ac)
* context->data_ac->ac_resv = &OCFS2_I(inode)->ip_la_data_resv;
*/
inode_lock(tl_inode);
if (ocfs2_truncate_log_needs_flush(osb)) {
ret = __ocfs2_flush_truncate_log(osb);
if (ret < 0) {
mlog_errno(ret);
goto out_unlock_mutex;
}
}
handle = ocfs2_start_trans(osb, credits);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
mlog_errno(ret);
goto out_unlock_mutex;
}
ret = __ocfs2_claim_clusters(handle, context->data_ac, 1, *len,
&new_phys_cpos, &new_len);
if (ret) {
mlog_errno(ret);
goto out_commit;
}
/*
* allowing partial extent moving is kind of 'pros and cons', it makes
* whole defragmentation less likely to fail, on the contrary, the bad
* thing is it may make the fs even more fragmented after moving, let
* userspace make a good decision here.
*/
if (new_len != *len) {
mlog(0, "len_claimed: %u, len: %u\n", new_len, *len);
if (!partial) {
context->range->me_flags &= ~OCFS2_MOVE_EXT_FL_COMPLETE;
ret = -ENOSPC;
goto out_commit;
}
}
mlog(0, "cpos: %u, phys_cpos: %u, new_phys_cpos: %u\n", cpos,
phys_cpos, new_phys_cpos);
ret = __ocfs2_move_extent(handle, context, cpos, new_len, phys_cpos,
new_phys_cpos, ext_flags);
if (ret)
mlog_errno(ret);
if (partial && (new_len != *len))
*len = new_len;
/*
* Here we should write the new page out first if we are
* in write-back mode.
*/
ret = ocfs2_cow_sync_writeback(inode->i_sb, context->inode, cpos, *len);
if (ret)
mlog_errno(ret);
out_commit:
ocfs2_commit_trans(osb, handle);
out_unlock_mutex:
inode_unlock(tl_inode);
if (context->data_ac) {
ocfs2_free_alloc_context(context->data_ac);
context->data_ac = NULL;
}
if (context->meta_ac) {
ocfs2_free_alloc_context(context->meta_ac);
context->meta_ac = NULL;
}
out:
if (ref_tree)
ocfs2_unlock_refcount_tree(osb, ref_tree, 1);
return ret;
}
static int ocfs2_move_extent(struct ocfs2_move_extents_context *context,
u32 cpos, u32 phys_cpos, u32 *new_phys_cpos,
u32 len, int ext_flags)
{
int ret, credits = 0, extra_blocks = 0, goal_bit = 0;
handle_t *handle;
struct inode *inode = context->inode;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct inode *tl_inode = osb->osb_tl_inode;
struct inode *gb_inode = NULL;
struct buffer_head *gb_bh = NULL;
struct buffer_head *gd_bh = NULL;
struct ocfs2_group_desc *gd;
struct ocfs2_refcount_tree *ref_tree = NULL;
u32 move_max_hop = ocfs2_blocks_to_clusters(inode->i_sb,
context->range->me_threshold);
u64 phys_blkno, new_phys_blkno;
phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, phys_cpos);
if ((ext_flags & OCFS2_EXT_REFCOUNTED) && len) {
BUG_ON(!(OCFS2_I(inode)->ip_dyn_features &
OCFS2_HAS_REFCOUNT_FL));
BUG_ON(!context->refcount_loc);
ret = ocfs2_lock_refcount_tree(osb, context->refcount_loc, 1,
&ref_tree, NULL);
if (ret) {
mlog_errno(ret);
return ret;
}
ret = ocfs2_prepare_refcount_change_for_del(inode,
context->refcount_loc,
phys_blkno,
len,
&credits,
&extra_blocks);
if (ret) {
mlog_errno(ret);
goto out;
}
}
ret = ocfs2_lock_allocators_move_extents(inode, &context->et, len, 1,
&context->meta_ac,
NULL, extra_blocks, &credits);
if (ret) {
mlog_errno(ret);
goto out;
}
/*
* need to count 2 extra credits for global_bitmap inode and
* group descriptor.
*/
credits += OCFS2_INODE_UPDATE_CREDITS + 1;
/*
* ocfs2_move_extent() didn't reserve any clusters in lock_allocators()
* logic, while we still need to lock the global_bitmap.
*/
gb_inode = ocfs2_get_system_file_inode(osb, GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT);
if (!gb_inode) {
mlog(ML_ERROR, "unable to get global_bitmap inode\n");
ret = -EIO;
goto out;
}
mutex_lock(&gb_inode->i_mutex);
ret = ocfs2_inode_lock(gb_inode, &gb_bh, 1);
if (ret) {
mlog_errno(ret);
goto out_unlock_gb_mutex;
}
mutex_lock(&tl_inode->i_mutex);
handle = ocfs2_start_trans(osb, credits);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
mlog_errno(ret);
goto out_unlock_tl_inode;
}
new_phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, *new_phys_cpos);
ret = ocfs2_find_victim_alloc_group(inode, new_phys_blkno,
GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT,
&goal_bit, &gd_bh);
if (ret) {
mlog_errno(ret);
goto out_commit;
}
/*
* probe the victim cluster group to find a proper
* region to fit wanted movement, it even will perfrom
* a best-effort attempt by compromising to a threshold
* around the goal.
*/
ocfs2_probe_alloc_group(inode, gd_bh, &goal_bit, len, move_max_hop,
new_phys_cpos);
if (!*new_phys_cpos) {
ret = -ENOSPC;
goto out_commit;
}
ret = __ocfs2_move_extent(handle, context, cpos, len, phys_cpos,
*new_phys_cpos, ext_flags);
if (ret) {
mlog_errno(ret);
goto out_commit;
}
gd = (struct ocfs2_group_desc *)gd_bh->b_data;
ret = ocfs2_alloc_dinode_update_counts(gb_inode, handle, gb_bh, len,
le16_to_cpu(gd->bg_chain));
if (ret) {
mlog_errno(ret);
goto out_commit;
}
ret = ocfs2_block_group_set_bits(handle, gb_inode, gd, gd_bh,
goal_bit, len);
if (ret)
mlog_errno(ret);
/*
* Here we should write the new page out first if we are
* in write-back mode.
*/
ret = ocfs2_cow_sync_writeback(inode->i_sb, context->inode, cpos, len);
if (ret)
mlog_errno(ret);
out_commit:
ocfs2_commit_trans(osb, handle);
brelse(gd_bh);
out_unlock_tl_inode:
mutex_unlock(&tl_inode->i_mutex);
ocfs2_inode_unlock(gb_inode, 1);
out_unlock_gb_mutex:
mutex_unlock(&gb_inode->i_mutex);
brelse(gb_bh);
iput(gb_inode);
out:
if (context->meta_ac) {
ocfs2_free_alloc_context(context->meta_ac);
context->meta_ac = NULL;
}
if (ref_tree)
ocfs2_unlock_refcount_tree(osb, ref_tree, 1);
return ret;
}
/*
* Step 2: By now, we've completed the journal recovery, we've stamped
* a clean local alloc on disk and dropped the node out of the
* recovery map. Dlm locks will no longer stall, so lets clear out the
* main bitmap.
*/
int ocfs2_complete_local_alloc_recovery(struct ocfs2_super *osb,
struct ocfs2_dinode *alloc)
{
int status;
struct ocfs2_journal_handle *handle = NULL;
struct buffer_head *main_bm_bh = NULL;
struct inode *main_bm_inode = NULL;
mlog_entry_void();
handle = ocfs2_alloc_handle(osb);
if (!handle) {
status = -ENOMEM;
mlog_errno(status);
goto bail;
}
main_bm_inode = ocfs2_get_system_file_inode(osb,
GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT);
if (!main_bm_inode) {
status = -EINVAL;
mlog_errno(status);
goto bail;
}
ocfs2_handle_add_inode(handle, main_bm_inode);
status = ocfs2_meta_lock(main_bm_inode, handle, &main_bm_bh, 1);
if (status < 0) {
mlog_errno(status);
goto bail;
}
handle = ocfs2_start_trans(osb, handle, OCFS2_WINDOW_MOVE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
handle = NULL;
mlog_errno(status);
goto bail;
}
/* we want the bitmap change to be recorded on disk asap */
ocfs2_handle_set_sync(handle, 1);
status = ocfs2_sync_local_to_main(osb, handle, alloc,
main_bm_inode, main_bm_bh);
if (status < 0)
mlog_errno(status);
bail:
if (handle)
ocfs2_commit_trans(handle);
if (main_bm_bh)
brelse(main_bm_bh);
if (main_bm_inode)
iput(main_bm_inode);
mlog_exit(status);
return status;
}
int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
{
int status = 0, size_change;
struct inode *inode = dentry->d_inode;
struct super_block *sb = inode->i_sb;
struct ocfs2_super *osb = OCFS2_SB(sb);
struct buffer_head *bh = NULL;
handle_t *handle = NULL;
mlog_entry("(0x%p, '%.*s')\n", dentry,
dentry->d_name.len, dentry->d_name.name);
if (attr->ia_valid & ATTR_MODE)
mlog(0, "mode change: %d\n", attr->ia_mode);
if (attr->ia_valid & ATTR_UID)
mlog(0, "uid change: %d\n", attr->ia_uid);
if (attr->ia_valid & ATTR_GID)
mlog(0, "gid change: %d\n", attr->ia_gid);
if (attr->ia_valid & ATTR_SIZE)
mlog(0, "size change...\n");
if (attr->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_CTIME))
mlog(0, "time change...\n");
#define OCFS2_VALID_ATTRS (ATTR_ATIME | ATTR_MTIME | ATTR_CTIME | ATTR_SIZE \
| ATTR_GID | ATTR_UID | ATTR_MODE)
if (!(attr->ia_valid & OCFS2_VALID_ATTRS)) {
mlog(0, "can't handle attrs: 0x%x\n", attr->ia_valid);
return 0;
}
status = inode_change_ok(inode, attr);
if (status)
return status;
size_change = S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE;
if (size_change) {
status = ocfs2_rw_lock(inode, 1);
if (status < 0) {
mlog_errno(status);
goto bail;
}
}
status = ocfs2_meta_lock(inode, &bh, 1);
if (status < 0) {
if (status != -ENOENT)
mlog_errno(status);
goto bail_unlock_rw;
}
if (size_change && attr->ia_size != i_size_read(inode)) {
if (i_size_read(inode) > attr->ia_size)
status = ocfs2_truncate_file(inode, bh, attr->ia_size);
else
status = ocfs2_extend_file(inode, bh, attr->ia_size, 0);
if (status < 0) {
if (status != -ENOSPC)
mlog_errno(status);
status = -ENOSPC;
goto bail_unlock;
}
}
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
goto bail_unlock;
}
status = inode_setattr(inode, attr);
if (status < 0) {
mlog_errno(status);
goto bail_commit;
}
status = ocfs2_mark_inode_dirty(handle, inode, bh);
if (status < 0)
mlog_errno(status);
bail_commit:
ocfs2_commit_trans(osb, handle);
bail_unlock:
ocfs2_meta_unlock(inode, 1);
bail_unlock_rw:
if (size_change)
ocfs2_rw_unlock(inode, 1);
bail:
if (bh)
brelse(bh);
mlog_exit(status);
return status;
}
/* Note that we do *NOT* lock the local alloc inode here as
* it's been locked already for us. */
static int ocfs2_local_alloc_slide_window(struct ocfs2_super *osb,
struct inode *local_alloc_inode)
{
int status = 0;
struct buffer_head *main_bm_bh = NULL;
struct inode *main_bm_inode = NULL;
struct ocfs2_journal_handle *handle = NULL;
struct ocfs2_dinode *alloc;
struct ocfs2_dinode *alloc_copy = NULL;
struct ocfs2_alloc_context *ac = NULL;
mlog_entry_void();
handle = ocfs2_alloc_handle(osb);
if (!handle) {
status = -ENOMEM;
mlog_errno(status);
goto bail;
}
/* This will lock the main bitmap for us. */
status = ocfs2_local_alloc_reserve_for_window(osb,
handle,
&ac,
&main_bm_inode,
&main_bm_bh);
if (status < 0) {
if (status != -ENOSPC)
mlog_errno(status);
goto bail;
}
handle = ocfs2_start_trans(osb, handle, OCFS2_WINDOW_MOVE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
handle = NULL;
mlog_errno(status);
goto bail;
}
alloc = (struct ocfs2_dinode *) osb->local_alloc_bh->b_data;
/* We want to clear the local alloc before doing anything
* else, so that if we error later during this operation,
* local alloc shutdown won't try to double free main bitmap
* bits. Make a copy so the sync function knows which bits to
* free. */
alloc_copy = kmalloc(osb->local_alloc_bh->b_size, GFP_KERNEL);
if (!alloc_copy) {
status = -ENOMEM;
mlog_errno(status);
goto bail;
}
memcpy(alloc_copy, alloc, osb->local_alloc_bh->b_size);
status = ocfs2_journal_access(handle, local_alloc_inode,
osb->local_alloc_bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto bail;
}
ocfs2_clear_local_alloc(alloc);
status = ocfs2_journal_dirty(handle, osb->local_alloc_bh);
if (status < 0) {
mlog_errno(status);
goto bail;
}
status = ocfs2_sync_local_to_main(osb, handle, alloc_copy,
main_bm_inode, main_bm_bh);
if (status < 0) {
mlog_errno(status);
goto bail;
}
status = ocfs2_local_alloc_new_window(osb, handle, ac);
if (status < 0) {
if (status != -ENOSPC)
mlog_errno(status);
goto bail;
}
atomic_inc(&osb->alloc_stats.moves);
status = 0;
bail:
if (handle)
ocfs2_commit_trans(handle);
if (main_bm_bh)
brelse(main_bm_bh);
if (main_bm_inode)
iput(main_bm_inode);
if (alloc_copy)
kfree(alloc_copy);
if (ac)
ocfs2_free_alloc_context(ac);
mlog_exit(status);
return status;
}
/*
* We expect the block group allocator to already be locked.
*/
static int ocfs2_block_group_alloc(struct ocfs2_super *osb,
struct inode *alloc_inode,
struct buffer_head *bh,
u64 max_block,
u64 *last_alloc_group,
int flags)
{
int status, credits;
struct ocfs2_dinode *fe = (struct ocfs2_dinode *) bh->b_data;
struct ocfs2_chain_list *cl;
struct ocfs2_alloc_context *ac = NULL;
handle_t *handle = NULL;
u16 alloc_rec;
struct buffer_head *bg_bh = NULL;
struct ocfs2_group_desc *bg;
BUG_ON(ocfs2_is_cluster_bitmap(alloc_inode));
cl = &fe->id2.i_chain;
status = ocfs2_reserve_clusters_with_limit(osb,
le16_to_cpu(cl->cl_cpg),
max_block, flags, &ac);
if (status < 0) {
if (status != -ENOSPC)
mlog_errno(status);
goto bail;
}
credits = ocfs2_calc_group_alloc_credits(osb->sb,
le16_to_cpu(cl->cl_cpg));
handle = ocfs2_start_trans(osb, credits);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
handle = NULL;
mlog_errno(status);
goto bail;
}
if (last_alloc_group && *last_alloc_group != 0) {
trace_ocfs2_block_group_alloc(
(unsigned long long)*last_alloc_group);
ac->ac_last_group = *last_alloc_group;
}
bg_bh = ocfs2_block_group_alloc_contig(osb, handle, alloc_inode,
ac, cl);
if (IS_ERR(bg_bh) && (PTR_ERR(bg_bh) == -ENOSPC))
bg_bh = ocfs2_block_group_alloc_discontig(handle,
alloc_inode,
ac, cl);
if (IS_ERR(bg_bh)) {
status = PTR_ERR(bg_bh);
bg_bh = NULL;
if (status != -ENOSPC)
mlog_errno(status);
goto bail;
}
bg = (struct ocfs2_group_desc *) bg_bh->b_data;
status = ocfs2_journal_access_di(handle, INODE_CACHE(alloc_inode),
bh, OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto bail;
}
alloc_rec = le16_to_cpu(bg->bg_chain);
le32_add_cpu(&cl->cl_recs[alloc_rec].c_free,
le16_to_cpu(bg->bg_free_bits_count));
le32_add_cpu(&cl->cl_recs[alloc_rec].c_total,
le16_to_cpu(bg->bg_bits));
cl->cl_recs[alloc_rec].c_blkno = bg->bg_blkno;
if (le16_to_cpu(cl->cl_next_free_rec) < le16_to_cpu(cl->cl_count))
le16_add_cpu(&cl->cl_next_free_rec, 1);
le32_add_cpu(&fe->id1.bitmap1.i_used, le16_to_cpu(bg->bg_bits) -
le16_to_cpu(bg->bg_free_bits_count));
le32_add_cpu(&fe->id1.bitmap1.i_total, le16_to_cpu(bg->bg_bits));
le32_add_cpu(&fe->i_clusters, le16_to_cpu(cl->cl_cpg));
ocfs2_journal_dirty(handle, bh);
spin_lock(&OCFS2_I(alloc_inode)->ip_lock);
OCFS2_I(alloc_inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
fe->i_size = cpu_to_le64(ocfs2_clusters_to_bytes(alloc_inode->i_sb,
le32_to_cpu(fe->i_clusters)));
spin_unlock(&OCFS2_I(alloc_inode)->ip_lock);
i_size_write(alloc_inode, le64_to_cpu(fe->i_size));
alloc_inode->i_blocks = ocfs2_inode_sector_count(alloc_inode);
ocfs2_update_inode_fsync_trans(handle, alloc_inode, 0);
status = 0;
/* save the new last alloc group so that the caller can cache it. */
if (last_alloc_group)
*last_alloc_group = ac->ac_last_group;
bail:
if (handle)
ocfs2_commit_trans(osb, handle);
if (ac)
ocfs2_free_alloc_context(ac);
brelse(bg_bh);
if (status)
mlog_errno(status);
return status;
}
/*
* Extend the filesystem to the new number of clusters specified. This entry
* point is only used to extend the current filesystem to the end of the last
* existing group.
*/
int ocfs2_group_extend(struct inode * inode, int new_clusters)
{
int ret;
handle_t *handle;
struct buffer_head *main_bm_bh = NULL;
struct buffer_head *group_bh = NULL;
struct inode *main_bm_inode = NULL;
struct ocfs2_dinode *fe = NULL;
struct ocfs2_group_desc *group = NULL;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
u16 cl_bpc;
u32 first_new_cluster;
u64 lgd_blkno;
if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb))
return -EROFS;
if (new_clusters < 0)
return -EINVAL;
else if (new_clusters == 0)
return 0;
main_bm_inode = ocfs2_get_system_file_inode(osb,
GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT);
if (!main_bm_inode) {
ret = -EINVAL;
mlog_errno(ret);
goto out;
}
mutex_lock(&main_bm_inode->i_mutex);
ret = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1);
if (ret < 0) {
mlog_errno(ret);
goto out_mutex;
}
fe = (struct ocfs2_dinode *)main_bm_bh->b_data;
/* main_bm_bh is validated by inode read inside ocfs2_inode_lock(),
* so any corruption is a code bug. */
BUG_ON(!OCFS2_IS_VALID_DINODE(fe));
if (le16_to_cpu(fe->id2.i_chain.cl_cpg) !=
ocfs2_group_bitmap_size(osb->sb, 0,
osb->s_feature_incompat) * 8) {
mlog(ML_ERROR, "The disk is too old and small. "
"Force to do offline resize.");
ret = -EINVAL;
goto out_unlock;
}
first_new_cluster = le32_to_cpu(fe->i_clusters);
lgd_blkno = ocfs2_which_cluster_group(main_bm_inode,
first_new_cluster - 1);
ret = ocfs2_read_group_descriptor(main_bm_inode, fe, lgd_blkno,
&group_bh);
if (ret < 0) {
mlog_errno(ret);
goto out_unlock;
}
group = (struct ocfs2_group_desc *)group_bh->b_data;
cl_bpc = le16_to_cpu(fe->id2.i_chain.cl_bpc);
if (le16_to_cpu(group->bg_bits) / cl_bpc + new_clusters >
le16_to_cpu(fe->id2.i_chain.cl_cpg)) {
ret = -EINVAL;
goto out_unlock;
}
trace_ocfs2_group_extend(
(unsigned long long)le64_to_cpu(group->bg_blkno), new_clusters);
handle = ocfs2_start_trans(osb, OCFS2_GROUP_EXTEND_CREDITS);
if (IS_ERR(handle)) {
mlog_errno(PTR_ERR(handle));
ret = -EINVAL;
goto out_unlock;
}
/* update the last group descriptor and inode. */
ret = ocfs2_update_last_group_and_inode(handle, main_bm_inode,
main_bm_bh, group_bh,
first_new_cluster,
new_clusters);
if (ret) {
mlog_errno(ret);
goto out_commit;
}
ocfs2_update_super_and_backups(main_bm_inode, new_clusters);
out_commit:
ocfs2_commit_trans(osb, handle);
out_unlock:
brelse(group_bh);
brelse(main_bm_bh);
ocfs2_inode_unlock(main_bm_inode, 1);
out_mutex:
mutex_unlock(&main_bm_inode->i_mutex);
iput(main_bm_inode);
out:
return ret;
}
/*
* We expect the block group allocator to already be locked.
*/
static int ocfs2_block_group_alloc(struct ocfs2_super *osb,
struct inode *alloc_inode,
struct buffer_head *bh)
{
int status, credits;
struct ocfs2_dinode *fe = (struct ocfs2_dinode *) bh->b_data;
struct ocfs2_chain_list *cl;
struct ocfs2_alloc_context *ac = NULL;
handle_t *handle = NULL;
u32 bit_off, num_bits;
u16 alloc_rec;
u64 bg_blkno;
struct buffer_head *bg_bh = NULL;
struct ocfs2_group_desc *bg;
BUG_ON(ocfs2_is_cluster_bitmap(alloc_inode));
mlog_entry_void();
cl = &fe->id2.i_chain;
status = ocfs2_reserve_clusters(osb,
le16_to_cpu(cl->cl_cpg),
&ac);
if (status < 0) {
if (status != -ENOSPC)
mlog_errno(status);
goto bail;
}
credits = ocfs2_calc_group_alloc_credits(osb->sb,
le16_to_cpu(cl->cl_cpg));
handle = ocfs2_start_trans(osb, credits);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
handle = NULL;
mlog_errno(status);
goto bail;
}
status = ocfs2_claim_clusters(osb,
handle,
ac,
le16_to_cpu(cl->cl_cpg),
&bit_off,
&num_bits);
if (status < 0) {
if (status != -ENOSPC)
mlog_errno(status);
goto bail;
}
alloc_rec = ocfs2_find_smallest_chain(cl);
/* setup the group */
bg_blkno = ocfs2_clusters_to_blocks(osb->sb, bit_off);
mlog(0, "new descriptor, record %u, at block %llu\n",
alloc_rec, (unsigned long long)bg_blkno);
bg_bh = sb_getblk(osb->sb, bg_blkno);
if (!bg_bh) {
status = -EIO;
mlog_errno(status);
goto bail;
}
ocfs2_set_new_buffer_uptodate(alloc_inode, bg_bh);
status = ocfs2_block_group_fill(handle,
alloc_inode,
bg_bh,
bg_blkno,
alloc_rec,
cl);
if (status < 0) {
mlog_errno(status);
goto bail;
}
bg = (struct ocfs2_group_desc *) bg_bh->b_data;
status = ocfs2_journal_access(handle, alloc_inode,
bh, OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto bail;
}
le32_add_cpu(&cl->cl_recs[alloc_rec].c_free,
le16_to_cpu(bg->bg_free_bits_count));
le32_add_cpu(&cl->cl_recs[alloc_rec].c_total, le16_to_cpu(bg->bg_bits));
cl->cl_recs[alloc_rec].c_blkno = cpu_to_le64(bg_blkno);
if (le16_to_cpu(cl->cl_next_free_rec) < le16_to_cpu(cl->cl_count))
le16_add_cpu(&cl->cl_next_free_rec, 1);
le32_add_cpu(&fe->id1.bitmap1.i_used, le16_to_cpu(bg->bg_bits) -
le16_to_cpu(bg->bg_free_bits_count));
le32_add_cpu(&fe->id1.bitmap1.i_total, le16_to_cpu(bg->bg_bits));
le32_add_cpu(&fe->i_clusters, le16_to_cpu(cl->cl_cpg));
status = ocfs2_journal_dirty(handle, bh);
if (status < 0) {
mlog_errno(status);
goto bail;
}
spin_lock(&OCFS2_I(alloc_inode)->ip_lock);
OCFS2_I(alloc_inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
fe->i_size = cpu_to_le64(ocfs2_clusters_to_bytes(alloc_inode->i_sb,
le32_to_cpu(fe->i_clusters)));
spin_unlock(&OCFS2_I(alloc_inode)->ip_lock);
i_size_write(alloc_inode, le64_to_cpu(fe->i_size));
alloc_inode->i_blocks = ocfs2_inode_sector_count(alloc_inode);
status = 0;
bail:
if (handle)
ocfs2_commit_trans(osb, handle);
if (ac)
ocfs2_free_alloc_context(ac);
if (bg_bh)
brelse(bg_bh);
mlog_exit(status);
return status;
}
/* Add a new group descriptor to global_bitmap. */
int ocfs2_group_add(struct inode *inode, struct ocfs2_new_group_input *input)
{
int ret;
handle_t *handle;
struct buffer_head *main_bm_bh = NULL;
struct inode *main_bm_inode = NULL;
struct ocfs2_dinode *fe = NULL;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct buffer_head *group_bh = NULL;
struct ocfs2_group_desc *group = NULL;
struct ocfs2_chain_list *cl;
struct ocfs2_chain_rec *cr;
u16 cl_bpc;
if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb))
return -EROFS;
main_bm_inode = ocfs2_get_system_file_inode(osb,
GLOBAL_BITMAP_SYSTEM_INODE,
OCFS2_INVALID_SLOT);
if (!main_bm_inode) {
ret = -EINVAL;
mlog_errno(ret);
goto out;
}
mutex_lock(&main_bm_inode->i_mutex);
ret = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1);
if (ret < 0) {
mlog_errno(ret);
goto out_mutex;
}
fe = (struct ocfs2_dinode *)main_bm_bh->b_data;
if (le16_to_cpu(fe->id2.i_chain.cl_cpg) !=
ocfs2_group_bitmap_size(osb->sb, 0,
osb->s_feature_incompat) * 8) {
mlog(ML_ERROR, "The disk is too old and small."
" Force to do offline resize.");
ret = -EINVAL;
goto out_unlock;
}
ret = ocfs2_read_blocks_sync(osb, input->group, 1, &group_bh);
if (ret < 0) {
mlog(ML_ERROR, "Can't read the group descriptor # %llu "
"from the device.", (unsigned long long)input->group);
goto out_unlock;
}
ocfs2_set_new_buffer_uptodate(INODE_CACHE(inode), group_bh);
ret = ocfs2_verify_group_and_input(main_bm_inode, fe, input, group_bh);
if (ret) {
mlog_errno(ret);
goto out_unlock;
}
trace_ocfs2_group_add((unsigned long long)input->group,
input->chain, input->clusters, input->frees);
handle = ocfs2_start_trans(osb, OCFS2_GROUP_ADD_CREDITS);
if (IS_ERR(handle)) {
mlog_errno(PTR_ERR(handle));
ret = -EINVAL;
goto out_unlock;
}
cl_bpc = le16_to_cpu(fe->id2.i_chain.cl_bpc);
cl = &fe->id2.i_chain;
cr = &cl->cl_recs[input->chain];
ret = ocfs2_journal_access_gd(handle, INODE_CACHE(main_bm_inode),
group_bh, OCFS2_JOURNAL_ACCESS_WRITE);
if (ret < 0) {
mlog_errno(ret);
goto out_commit;
}
group = (struct ocfs2_group_desc *)group_bh->b_data;
group->bg_next_group = cr->c_blkno;
ocfs2_journal_dirty(handle, group_bh);
ret = ocfs2_journal_access_di(handle, INODE_CACHE(main_bm_inode),
main_bm_bh, OCFS2_JOURNAL_ACCESS_WRITE);
if (ret < 0) {
mlog_errno(ret);
goto out_commit;
}
if (input->chain == le16_to_cpu(cl->cl_next_free_rec)) {
le16_add_cpu(&cl->cl_next_free_rec, 1);
memset(cr, 0, sizeof(struct ocfs2_chain_rec));
}
cr->c_blkno = cpu_to_le64(input->group);
le32_add_cpu(&cr->c_total, input->clusters * cl_bpc);
le32_add_cpu(&cr->c_free, input->frees * cl_bpc);
le32_add_cpu(&fe->id1.bitmap1.i_total, input->clusters *cl_bpc);
le32_add_cpu(&fe->id1.bitmap1.i_used,
(input->clusters - input->frees) * cl_bpc);
le32_add_cpu(&fe->i_clusters, input->clusters);
ocfs2_journal_dirty(handle, main_bm_bh);
spin_lock(&OCFS2_I(main_bm_inode)->ip_lock);
OCFS2_I(main_bm_inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
le64_add_cpu(&fe->i_size, input->clusters << osb->s_clustersize_bits);
spin_unlock(&OCFS2_I(main_bm_inode)->ip_lock);
i_size_write(main_bm_inode, le64_to_cpu(fe->i_size));
ocfs2_update_super_and_backups(main_bm_inode, input->clusters);
out_commit:
ocfs2_commit_trans(osb, handle);
out_unlock:
brelse(group_bh);
brelse(main_bm_bh);
ocfs2_inode_unlock(main_bm_inode, 1);
out_mutex:
mutex_unlock(&main_bm_inode->i_mutex);
iput(main_bm_inode);
out:
return ret;
}
static int ocfs2_defrag_extent(struct ocfs2_move_extents_context *context,
u32 cpos, u32 phys_cpos, u32 *len, int ext_flags)
{
int ret, credits = 0, extra_blocks = 0, partial = context->partial;
handle_t *handle;
struct inode *inode = context->inode;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct inode *tl_inode = osb->osb_tl_inode;
struct ocfs2_refcount_tree *ref_tree = NULL;
u32 new_phys_cpos, new_len;
u64 phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, phys_cpos);
if ((ext_flags & OCFS2_EXT_REFCOUNTED) && *len) {
BUG_ON(!(OCFS2_I(inode)->ip_dyn_features &
OCFS2_HAS_REFCOUNT_FL));
BUG_ON(!context->refcount_loc);
ret = ocfs2_lock_refcount_tree(osb, context->refcount_loc, 1,
&ref_tree, NULL);
if (ret) {
mlog_errno(ret);
return ret;
}
ret = ocfs2_prepare_refcount_change_for_del(inode,
context->refcount_loc,
phys_blkno,
*len,
&credits,
&extra_blocks);
if (ret) {
mlog_errno(ret);
goto out;
}
}
ret = ocfs2_lock_allocators_move_extents(inode, &context->et, *len, 1,
&context->meta_ac,
&context->data_ac,
extra_blocks, &credits);
if (ret) {
mlog_errno(ret);
goto out;
}
mutex_lock(&tl_inode->i_mutex);
if (ocfs2_truncate_log_needs_flush(osb)) {
ret = __ocfs2_flush_truncate_log(osb);
if (ret < 0) {
mlog_errno(ret);
goto out_unlock_mutex;
}
}
handle = ocfs2_start_trans(osb, credits);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
mlog_errno(ret);
goto out_unlock_mutex;
}
ret = __ocfs2_claim_clusters(handle, context->data_ac, 1, *len,
&new_phys_cpos, &new_len);
if (ret) {
mlog_errno(ret);
goto out_commit;
}
if (new_len != *len) {
mlog(0, "len_claimed: %u, len: %u\n", new_len, *len);
if (!partial) {
context->range->me_flags &= ~OCFS2_MOVE_EXT_FL_COMPLETE;
ret = -ENOSPC;
goto out_commit;
}
}
mlog(0, "cpos: %u, phys_cpos: %u, new_phys_cpos: %u\n", cpos,
phys_cpos, new_phys_cpos);
ret = __ocfs2_move_extent(handle, context, cpos, new_len, phys_cpos,
new_phys_cpos, ext_flags);
if (ret)
mlog_errno(ret);
if (partial && (new_len != *len))
*len = new_len;
ret = ocfs2_cow_sync_writeback(inode->i_sb, context->inode, cpos, *len);
if (ret)
mlog_errno(ret);
out_commit:
ocfs2_commit_trans(osb, handle);
out_unlock_mutex:
mutex_unlock(&tl_inode->i_mutex);
if (context->data_ac) {
ocfs2_free_alloc_context(context->data_ac);
context->data_ac = NULL;
}
if (context->meta_ac) {
ocfs2_free_alloc_context(context->meta_ac);
context->meta_ac = NULL;
}
out:
if (ref_tree)
ocfs2_unlock_refcount_tree(osb, ref_tree, 1);
return ret;
}