summaryrefslogtreecommitdiffstats
path: root/fs/ocfs2
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ocfs2')
-rw-r--r--fs/ocfs2/alloc.c47
-rw-r--r--fs/ocfs2/aops.c69
-rw-r--r--fs/ocfs2/dcache.c35
-rw-r--r--fs/ocfs2/dcache.h3
-rw-r--r--fs/ocfs2/dlm/dlmast.c1
-rw-r--r--fs/ocfs2/dlm/dlmrecovery.c2
-rw-r--r--fs/ocfs2/dlmglue.c123
-rw-r--r--fs/ocfs2/dlmglue.h24
-rw-r--r--fs/ocfs2/file.c11
-rw-r--r--fs/ocfs2/inode.c11
-rw-r--r--fs/ocfs2/ioctl.c1
-rw-r--r--fs/ocfs2/journal.c49
-rw-r--r--fs/ocfs2/journal.h21
-rw-r--r--fs/ocfs2/namei.c15
-rw-r--r--fs/ocfs2/ocfs2.h32
-rw-r--r--fs/ocfs2/quota.h1
-rw-r--r--fs/ocfs2/quota_global.c134
-rw-r--r--fs/ocfs2/quota_local.c110
-rw-r--r--fs/ocfs2/stack_o2cb.c14
-rw-r--r--fs/ocfs2/stack_user.c8
-rw-r--r--fs/ocfs2/stackglue.c13
-rw-r--r--fs/ocfs2/stackglue.h6
-rw-r--r--fs/ocfs2/suballoc.c28
-rw-r--r--fs/ocfs2/super.c97
-rw-r--r--fs/ocfs2/sysfile.c19
-rw-r--r--fs/ocfs2/xattr.c3
26 files changed, 638 insertions, 239 deletions
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 9edcde4..f9a3e89 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -1914,7 +1914,8 @@ static void ocfs2_adjust_adjacent_records(struct ocfs2_extent_rec *left_rec,
* immediately to their right.
*/
left_clusters = le32_to_cpu(right_child_el->l_recs[0].e_cpos);
- if (ocfs2_is_empty_extent(&right_child_el->l_recs[0])) {
+ if (!ocfs2_rec_clusters(right_child_el, &right_child_el->l_recs[0])) {
+ BUG_ON(right_child_el->l_tree_depth);
BUG_ON(le16_to_cpu(right_child_el->l_next_free_rec) <= 1);
left_clusters = le32_to_cpu(right_child_el->l_recs[1].e_cpos);
}
@@ -2476,15 +2477,37 @@ out_ret_path:
return ret;
}
-static void ocfs2_update_edge_lengths(struct inode *inode, handle_t *handle,
- struct ocfs2_path *path)
+static int ocfs2_update_edge_lengths(struct inode *inode, handle_t *handle,
+ int subtree_index, struct ocfs2_path *path)
{
- int i, idx;
+ int i, idx, ret;
struct ocfs2_extent_rec *rec;
struct ocfs2_extent_list *el;
struct ocfs2_extent_block *eb;
u32 range;
+ /*
+ * In normal tree rotation process, we will never touch the
+ * tree branch above subtree_index and ocfs2_extend_rotate_transaction
+ * doesn't reserve the credits for them either.
+ *
+ * But we do have a special case here which will update the rightmost
+ * records for all the bh in the path.
+ * So we have to allocate extra credits and access them.
+ */
+ ret = ocfs2_extend_trans(handle,
+ handle->h_buffer_credits + subtree_index);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
+
+ ret = ocfs2_journal_access_path(inode, handle, path);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
+
/* Path should always be rightmost. */
eb = (struct ocfs2_extent_block *)path_leaf_bh(path)->b_data;
BUG_ON(eb->h_next_leaf_blk != 0ULL);
@@ -2505,6 +2528,8 @@ static void ocfs2_update_edge_lengths(struct inode *inode, handle_t *handle,
ocfs2_journal_dirty(handle, path->p_node[i].bh);
}
+out:
+ return ret;
}
static void ocfs2_unlink_path(struct inode *inode, handle_t *handle,
@@ -2717,7 +2742,12 @@ static int ocfs2_rotate_subtree_left(struct inode *inode, handle_t *handle,
if (del_right_subtree) {
ocfs2_unlink_subtree(inode, handle, left_path, right_path,
subtree_index, dealloc);
- ocfs2_update_edge_lengths(inode, handle, left_path);
+ ret = ocfs2_update_edge_lengths(inode, handle, subtree_index,
+ left_path);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
eb = (struct ocfs2_extent_block *)path_leaf_bh(left_path)->b_data;
ocfs2_et_set_last_eb_blk(et, le64_to_cpu(eb->h_blkno));
@@ -3034,7 +3064,12 @@ static int ocfs2_remove_rightmost_path(struct inode *inode, handle_t *handle,
ocfs2_unlink_subtree(inode, handle, left_path, path,
subtree_index, dealloc);
- ocfs2_update_edge_lengths(inode, handle, left_path);
+ ret = ocfs2_update_edge_lengths(inode, handle, subtree_index,
+ left_path);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
eb = (struct ocfs2_extent_block *)path_leaf_bh(left_path)->b_data;
ocfs2_et_set_last_eb_blk(et, le64_to_cpu(eb->h_blkno));
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index b2c52b3..b401654 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -193,6 +193,7 @@ static int ocfs2_get_block(struct inode *inode, sector_t iblock,
(unsigned long long)OCFS2_I(inode)->ip_blkno);
mlog(ML_ERROR, "Size %llu, clusters %u\n", (unsigned long long)i_size_read(inode), OCFS2_I(inode)->ip_clusters);
dump_stack();
+ goto bail;
}
past_eof = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode));
@@ -894,18 +895,17 @@ struct ocfs2_write_cluster_desc {
*/
unsigned c_new;
unsigned c_unwritten;
+ unsigned c_needs_zero;
};
-static inline int ocfs2_should_zero_cluster(struct ocfs2_write_cluster_desc *d)
-{
- return d->c_new || d->c_unwritten;
-}
-
struct ocfs2_write_ctxt {
/* Logical cluster position / len of write */
u32 w_cpos;
u32 w_clen;
+ /* First cluster allocated in a nonsparse extend */
+ u32 w_first_new_cpos;
+
struct ocfs2_write_cluster_desc w_desc[OCFS2_MAX_CLUSTERS_PER_PAGE];
/*
@@ -983,6 +983,7 @@ static int ocfs2_alloc_write_ctxt(struct ocfs2_write_ctxt **wcp,
return -ENOMEM;
wc->w_cpos = pos >> osb->s_clustersize_bits;
+ wc->w_first_new_cpos = UINT_MAX;
cend = (pos + len - 1) >> osb->s_clustersize_bits;
wc->w_clen = cend - wc->w_cpos + 1;
get_bh(di_bh);
@@ -1217,20 +1218,18 @@ out:
*/
static int ocfs2_write_cluster(struct address_space *mapping,
u32 phys, unsigned int unwritten,
+ unsigned int should_zero,
struct ocfs2_alloc_context *data_ac,
struct ocfs2_alloc_context *meta_ac,
struct ocfs2_write_ctxt *wc, u32 cpos,
loff_t user_pos, unsigned user_len)
{
- int ret, i, new, should_zero = 0;
+ int ret, i, new;
u64 v_blkno, p_blkno;
struct inode *inode = mapping->host;
struct ocfs2_extent_tree et;
new = phys == 0 ? 1 : 0;
- if (new || unwritten)
- should_zero = 1;
-
if (new) {
u32 tmp_pos;
@@ -1301,7 +1300,7 @@ static int ocfs2_write_cluster(struct address_space *mapping,
if (tmpret) {
mlog_errno(tmpret);
if (ret == 0)
- tmpret = ret;
+ ret = tmpret;
}
}
@@ -1341,7 +1340,9 @@ static int ocfs2_write_cluster_by_desc(struct address_space *mapping,
local_len = osb->s_clustersize - cluster_off;
ret = ocfs2_write_cluster(mapping, desc->c_phys,
- desc->c_unwritten, data_ac, meta_ac,
+ desc->c_unwritten,
+ desc->c_needs_zero,
+ data_ac, meta_ac,
wc, desc->c_cpos, pos, local_len);
if (ret) {
mlog_errno(ret);
@@ -1391,14 +1392,14 @@ static void ocfs2_set_target_boundaries(struct ocfs2_super *osb,
* newly allocated cluster.
*/
desc = &wc->w_desc[0];
- if (ocfs2_should_zero_cluster(desc))
+ if (desc->c_needs_zero)
ocfs2_figure_cluster_boundaries(osb,
desc->c_cpos,
&wc->w_target_from,
NULL);
desc = &wc->w_desc[wc->w_clen - 1];
- if (ocfs2_should_zero_cluster(desc))
+ if (desc->c_needs_zero)
ocfs2_figure_cluster_boundaries(osb,
desc->c_cpos,
NULL,
@@ -1466,13 +1467,28 @@ static int ocfs2_populate_write_desc(struct inode *inode,
phys++;
}
+ /*
+ * If w_first_new_cpos is < UINT_MAX, we have a non-sparse
+ * file that got extended. w_first_new_cpos tells us
+ * where the newly allocated clusters are so we can
+ * zero them.
+ */
+ if (desc->c_cpos >= wc->w_first_new_cpos) {
+ BUG_ON(phys == 0);
+ desc->c_needs_zero = 1;
+ }
+
desc->c_phys = phys;
if (phys == 0) {
desc->c_new = 1;
+ desc->c_needs_zero = 1;
*clusters_to_alloc = *clusters_to_alloc + 1;
}
- if (ext_flags & OCFS2_EXT_UNWRITTEN)
+
+ if (ext_flags & OCFS2_EXT_UNWRITTEN) {
desc->c_unwritten = 1;
+ desc->c_needs_zero = 1;
+ }
num_clusters--;
}
@@ -1632,10 +1648,13 @@ static int ocfs2_expand_nonsparse_inode(struct inode *inode, loff_t pos,
if (newsize <= i_size_read(inode))
return 0;
- ret = ocfs2_extend_no_holes(inode, newsize, newsize - len);
+ ret = ocfs2_extend_no_holes(inode, newsize, pos);
if (ret)
mlog_errno(ret);
+ wc->w_first_new_cpos =
+ ocfs2_clusters_for_bytes(inode->i_sb, i_size_read(inode));
+
return ret;
}
@@ -1644,7 +1663,7 @@ int ocfs2_write_begin_nolock(struct address_space *mapping,
struct page **pagep, void **fsdata,
struct buffer_head *di_bh, struct page *mmap_page)
{
- int ret, credits = OCFS2_INODE_UPDATE_CREDITS;
+ int ret, cluster_of_pages, credits = OCFS2_INODE_UPDATE_CREDITS;
unsigned int clusters_to_alloc, extents_to_split;
struct ocfs2_write_ctxt *wc;
struct inode *inode = mapping->host;
@@ -1722,8 +1741,19 @@ int ocfs2_write_begin_nolock(struct address_space *mapping,
}
- ocfs2_set_target_boundaries(osb, wc, pos, len,
- clusters_to_alloc + extents_to_split);
+ /*
+ * We have to zero sparse allocated clusters, unwritten extent clusters,
+ * and non-sparse clusters we just extended. For non-sparse writes,
+ * we know zeros will only be needed in the first and/or last cluster.
+ */
+ if (clusters_to_alloc || extents_to_split ||
+ wc->w_desc[0].c_needs_zero ||
+ wc->w_desc[wc->w_clen - 1].c_needs_zero)
+ cluster_of_pages = 1;
+ else
+ cluster_of_pages = 0;
+
+ ocfs2_set_target_boundaries(osb, wc, pos, len, cluster_of_pages);
handle = ocfs2_start_trans(osb, credits);
if (IS_ERR(handle)) {
@@ -1756,8 +1786,7 @@ int ocfs2_write_begin_nolock(struct address_space *mapping,
* extent.
*/
ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos,
- clusters_to_alloc + extents_to_split,
- mmap_page);
+ cluster_of_pages, mmap_page);
if (ret) {
mlog_errno(ret);
goto out_quota;
diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c
index b574431..2f28b7d 100644
--- a/fs/ocfs2/dcache.c
+++ b/fs/ocfs2/dcache.c
@@ -310,22 +310,19 @@ out_attach:
return ret;
}
-static DEFINE_SPINLOCK(dentry_list_lock);
+DEFINE_SPINLOCK(dentry_list_lock);
/* We limit the number of dentry locks to drop in one go. We have
* this limit so that we don't starve other users of ocfs2_wq. */
#define DL_INODE_DROP_COUNT 64
/* Drop inode references from dentry locks */
-void ocfs2_drop_dl_inodes(struct work_struct *work)
+static void __ocfs2_drop_dl_inodes(struct ocfs2_super *osb, int drop_count)
{
- struct ocfs2_super *osb = container_of(work, struct ocfs2_super,
- dentry_lock_work);
struct ocfs2_dentry_lock *dl;
- int drop_count = DL_INODE_DROP_COUNT;
spin_lock(&dentry_list_lock);
- while (osb->dentry_lock_list && drop_count--) {
+ while (osb->dentry_lock_list && (drop_count < 0 || drop_count--)) {
dl = osb->dentry_lock_list;
osb->dentry_lock_list = dl->dl_next;
spin_unlock(&dentry_list_lock);
@@ -333,11 +330,32 @@ void ocfs2_drop_dl_inodes(struct work_struct *work)
kfree(dl);
spin_lock(&dentry_list_lock);
}
- if (osb->dentry_lock_list)
+ spin_unlock(&dentry_list_lock);
+}
+
+void ocfs2_drop_dl_inodes(struct work_struct *work)
+{
+ struct ocfs2_super *osb = container_of(work, struct ocfs2_super,
+ dentry_lock_work);
+
+ __ocfs2_drop_dl_inodes(osb, DL_INODE_DROP_COUNT);
+ /*
+ * Don't queue dropping if umount is in progress. We flush the
+ * list in ocfs2_dismount_volume
+ */
+ spin_lock(&dentry_list_lock);
+ if (osb->dentry_lock_list &&
+ !ocfs2_test_osb_flag(osb, OCFS2_OSB_DROP_DENTRY_LOCK_IMMED))
queue_work(ocfs2_wq, &osb->dentry_lock_work);
spin_unlock(&dentry_list_lock);
}
+/* Flush the whole work queue */
+void ocfs2_drop_all_dl_inodes(struct ocfs2_super *osb)
+{
+ __ocfs2_drop_dl_inodes(osb, -1);
+}
+
/*
* ocfs2_dentry_iput() and friends.
*
@@ -368,7 +386,8 @@ static void ocfs2_drop_dentry_lock(struct ocfs2_super *osb,
/* We leave dropping of inode reference to ocfs2_wq as that can
* possibly lead to inode deletion which gets tricky */
spin_lock(&dentry_list_lock);
- if (!osb->dentry_lock_list)
+ if (!osb->dentry_lock_list &&
+ !ocfs2_test_osb_flag(osb, OCFS2_OSB_DROP_DENTRY_LOCK_IMMED))
queue_work(ocfs2_wq, &osb->dentry_lock_work);
dl->dl_next = osb->dentry_lock_list;
osb->dentry_lock_list = dl;
diff --git a/fs/ocfs2/dcache.h b/fs/ocfs2/dcache.h
index faa12e7..f5dd178 100644
--- a/fs/ocfs2/dcache.h
+++ b/fs/ocfs2/dcache.h
@@ -49,10 +49,13 @@ struct ocfs2_dentry_lock {
int ocfs2_dentry_attach_lock(struct dentry *dentry, struct inode *inode,
u64 parent_blkno);
+extern spinlock_t dentry_list_lock;
+
void ocfs2_dentry_lock_put(struct ocfs2_super *osb,
struct ocfs2_dentry_lock *dl);
void ocfs2_drop_dl_inodes(struct work_struct *work);
+void ocfs2_drop_all_dl_inodes(struct ocfs2_super *osb);
struct dentry *ocfs2_find_local_alias(struct inode *inode, u64 parent_blkno,
int skip_unhashed);
diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c
index d07ddbe..81eff8e 100644
--- a/fs/ocfs2/dlm/dlmast.c
+++ b/fs/ocfs2/dlm/dlmast.c
@@ -103,7 +103,6 @@ static void __dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock)
lock->ast_pending, lock->ml.type);
BUG();
}
- BUG_ON(!list_empty(&lock->ast_list));
if (lock->ast_pending)
mlog(0, "lock has an ast getting flushed right now\n");
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index bcb9260..43e6e32 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -1118,7 +1118,7 @@ static int dlm_send_mig_lockres_msg(struct dlm_ctxt *dlm,
mlog(0, "%s:%.*s: sending mig lockres (%s) to %u\n",
dlm->name, res->lockname.len, res->lockname.name,
- orig_flags & DLM_MRES_MIGRATION ? "migrate" : "recovery",
+ orig_flags & DLM_MRES_MIGRATION ? "migration" : "recovery",
send_to);
/* send it */
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 6cdeaa7..110bb57 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -92,6 +92,9 @@ struct ocfs2_unblock_ctl {
enum ocfs2_unblock_action unblock_action;
};
+/* Lockdep class keys */
+struct lock_class_key lockdep_keys[OCFS2_NUM_LOCK_TYPES];
+
static int ocfs2_check_meta_downconvert(struct ocfs2_lock_res *lockres,
int new_level);
static void ocfs2_set_meta_lvb(struct ocfs2_lock_res *lockres);
@@ -317,9 +320,16 @@ static int ocfs2_lock_create(struct ocfs2_super *osb,
u32 dlm_flags);
static inline int ocfs2_may_continue_on_blocked_lock(struct ocfs2_lock_res *lockres,
int wanted);
-static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
- struct ocfs2_lock_res *lockres,
- int level);
+static void __ocfs2_cluster_unlock(struct ocfs2_super *osb,
+ struct ocfs2_lock_res *lockres,
+ int level, unsigned long caller_ip);
+static inline void ocfs2_cluster_unlock(struct ocfs2_super *osb,
+ struct ocfs2_lock_res *lockres,
+ int level)
+{
+ __ocfs2_cluster_unlock(osb, lockres, level, _RET_IP_);
+}
+
static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res *lockres);
static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lockres);
static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *lockres);
@@ -489,6 +499,13 @@ static void ocfs2_lock_res_init_common(struct ocfs2_super *osb,
ocfs2_add_lockres_tracking(res, osb->osb_dlm_debug);
ocfs2_init_lock_stats(res);
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+ if (type != OCFS2_LOCK_TYPE_OPEN)
+ lockdep_init_map(&res->l_lockdep_map, ocfs2_lock_type_strings[type],
+ &lockdep_keys[type], 0);
+ else
+ res->l_lockdep_map.key = NULL;
+#endif
}
void ocfs2_lock_res_init_once(struct ocfs2_lock_res *res)
@@ -644,14 +661,10 @@ static void ocfs2_nfs_sync_lock_res_init(struct ocfs2_lock_res *res,
static void ocfs2_orphan_scan_lock_res_init(struct ocfs2_lock_res *res,
struct ocfs2_super *osb)
{
- struct ocfs2_orphan_scan_lvb *lvb;
-
ocfs2_lock_res_init_once(res);
ocfs2_build_lock_name(OCFS2_LOCK_TYPE_ORPHAN_SCAN, 0, 0, res->l_name);
ocfs2_lock_res_init_common(osb, res, OCFS2_LOCK_TYPE_ORPHAN_SCAN,
&ocfs2_orphan_scan_lops, osb);
- lvb = ocfs2_dlm_lvb(&res->l_lksb);
- lvb->lvb_version = OCFS2_ORPHAN_LVB_VERSION;
}
void ocfs2_file_lock_res_init(struct ocfs2_lock_res *lockres,
@@ -1256,11 +1269,13 @@ static int ocfs2_wait_for_mask_interruptible(struct ocfs2_mask_waiter *mw,
return ret;
}
-static int ocfs2_cluster_lock(struct ocfs2_super *osb,
- struct ocfs2_lock_res *lockres,
- int level,
- u32 lkm_flags,
- int arg_flags)
+static int __ocfs2_cluster_lock(struct ocfs2_super *osb,
+ struct ocfs2_lock_res *lockres,
+ int level,
+ u32 lkm_flags,
+ int arg_flags,
+ int l_subclass,
+ unsigned long caller_ip)
{
struct ocfs2_mask_waiter mw;
int wait, catch_signals = !(osb->s_mount_opt & OCFS2_MOUNT_NOINTR);
@@ -1403,13 +1418,37 @@ out:
}
ocfs2_update_lock_stats(lockres, level, &mw, ret);
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+ if (!ret && lockres->l_lockdep_map.key != NULL) {
+ if (level == DLM_LOCK_PR)
+ rwsem_acquire_read(&lockres->l_lockdep_map, l_subclass,
+ !!(arg_flags & OCFS2_META_LOCK_NOQUEUE),
+ caller_ip);
+ else
+ rwsem_acquire(&lockres->l_lockdep_map, l_subclass,
+ !!(arg_flags & OCFS2_META_LOCK_NOQUEUE),
+ caller_ip);
+ }
+#endif
mlog_exit(ret);
return ret;
}
-static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
- struct ocfs2_lock_res *lockres,
- int level)
+static inline int ocfs2_cluster_lock(struct ocfs2_super *osb,
+ struct ocfs2_lock_res *lockres,
+ int level,
+ u32 lkm_flags,
+ int arg_flags)
+{
+ return __ocfs2_cluster_lock(osb, lockres, level, lkm_flags, arg_flags,
+ 0, _RET_IP_);
+}
+
+
+static void __ocfs2_cluster_unlock(struct ocfs2_super *osb,
+ struct ocfs2_lock_res *lockres,
+ int level,
+ unsigned long caller_ip)
{
unsigned long flags;
@@ -1418,6 +1457,10 @@ static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
ocfs2_dec_holders(lockres, level);
ocfs2_downconvert_on_unlock(osb, lockres);
spin_unlock_irqrestore(&lockres->l_lock, flags);
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+ if (lockres->l_lockdep_map.key != NULL)
+ rwsem_release(&lockres->l_lockdep_map, 1, caller_ip);
+#endif
mlog_exit_void();
}
@@ -1989,7 +2032,8 @@ static inline int ocfs2_meta_lvb_is_trustable(struct inode *inode,
{
struct ocfs2_meta_lvb *lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
- if (lvb->lvb_version == OCFS2_LVB_VERSION
+ if (ocfs2_dlm_lvb_valid(&lockres->l_lksb)
+ && lvb->lvb_version == OCFS2_LVB_VERSION
&& be32_to_cpu(lvb->lvb_igeneration) == inode->i_generation)
return 1;
return 0;
@@ -2162,10 +2206,11 @@ static int ocfs2_assign_bh(struct inode *inode,
* returns < 0 error if the callback will never be called, otherwise
* the result of the lock will be communicated via the callback.
*/
-int ocfs2_inode_lock_full(struct inode *inode,
- struct buffer_head **ret_bh,
- int ex,
- int arg_flags)
+int ocfs2_inode_lock_full_nested(struct inode *inode,
+ struct buffer_head **ret_bh,
+ int ex,
+ int arg_flags,
+ int subclass)
{
int status, level, acquired;
u32 dlm_flags;
@@ -2203,7 +2248,8 @@ int ocfs2_inode_lock_full(struct inode *inode,
if (arg_flags & OCFS2_META_LOCK_NOQUEUE)
dlm_flags |= DLM_LKF_NOQUEUE;
- status = ocfs2_cluster_lock(osb, lockres, level, dlm_flags, arg_flags);
+ status = __ocfs2_cluster_lock(osb, lockres, level, dlm_flags,
+ arg_flags, subclass, _RET_IP_);
if (status < 0) {
if (status != -EAGAIN && status != -EIOCBRETRY)
mlog_errno(status);
@@ -2369,35 +2415,45 @@ void ocfs2_inode_unlock(struct inode *inode,
mlog_exit_void();
}
-int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
+int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno)
{
struct ocfs2_lock_res *lockres;
struct ocfs2_orphan_scan_lvb *lvb;
- int level = ex ? DLM_LOCK_EX : DLM_LOCK_PR;
int status = 0;
+ if (ocfs2_is_hard_readonly(osb))
+ return -EROFS;
+
+ if (ocfs2_mount_local(osb))
+ return 0;
+
lockres = &osb->osb_orphan_scan.os_lockres;
- status = ocfs2_cluster_lock(osb, lockres, level, 0, 0);
+ status = ocfs2_cluster_lock(osb, lockres, DLM_LOCK_EX, 0, 0);
if (status < 0)
return status;
lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
- if (lvb->lvb_version == OCFS2_ORPHAN_LVB_VERSION)
+ if (ocfs2_dlm_lvb_valid(&lockres->l_lksb) &&
+ lvb->lvb_version == OCFS2_ORPHAN_LVB_VERSION)
*seqno = be32_to_cpu(lvb->lvb_os_seqno);
+ else
+ *seqno = osb->osb_orphan_scan.os_seqno + 1;
+
return status;
}
-void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno, int ex)
+void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno)
{
struct ocfs2_lock_res *lockres;
struct ocfs2_orphan_scan_lvb *lvb;
- int level = ex ? DLM_LOCK_EX : DLM_LOCK_PR;
- lockres = &osb->osb_orphan_scan.os_lockres;
- lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
- lvb->lvb_version = OCFS2_ORPHAN_LVB_VERSION;
- lvb->lvb_os_seqno = cpu_to_be32(seqno);
- ocfs2_cluster_unlock(osb, lockres, level);
+ if (!ocfs2_is_hard_readonly(osb) && !ocfs2_mount_local(osb)) {
+ lockres = &osb->osb_orphan_scan.os_lockres;
+ lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
+ lvb->lvb_version = OCFS2_ORPHAN_LVB_VERSION;
+ lvb->lvb_os_seqno = cpu_to_be32(seqno);
+ ocfs2_cluster_unlock(osb, lockres, DLM_LOCK_EX);
+ }
}
int ocfs2_super_lock(struct ocfs2_super *osb,
@@ -3627,7 +3683,8 @@ static int ocfs2_refresh_qinfo(struct ocfs2_mem_dqinfo *oinfo)
struct ocfs2_global_disk_dqinfo *gdinfo;
int status = 0;
- if (lvb->lvb_version == OCFS2_QINFO_LVB_VERSION) {
+ if (ocfs2_dlm_lvb_valid(&lockres->l_lksb) &&
+ lvb->lvb_version == OCFS2_QINFO_LVB_VERSION) {
info->dqi_bgrace = be32_to_cpu(lvb->lvb_bgrace);
info->dqi_igrace = be32_to_cpu(lvb->lvb_igrace);
oinfo->dqi_syncms = be32_to_cpu(lvb->lvb_syncms);
diff --git a/fs/ocfs2/dlmglue.h b/fs/ocfs2/dlmglue.h
index 31b90d7..7553836 100644
--- a/fs/ocfs2/dlmglue.h
+++ b/fs/ocfs2/dlmglue.h
@@ -78,6 +78,14 @@ struct ocfs2_orphan_scan_lvb {
/* don't block waiting for the downconvert thread, instead return -EAGAIN */
#define OCFS2_LOCK_NONBLOCK (0x04)
+/* Locking subclasses of inode cluster lock */
+enum {
+ OI_LS_NORMAL = 0,
+ OI_LS_PARENT,
+ OI_LS_RENAME1,
+ OI_LS_RENAME2,
+};
+
int ocfs2_dlm_init(struct ocfs2_super *osb);
void ocfs2_dlm_shutdown(struct ocfs2_super *osb, int hangup_pending);
void ocfs2_lock_res_init_once(struct ocfs2_lock_res *res);
@@ -104,25 +112,31 @@ void ocfs2_open_unlock(struct inode *inode);
int ocfs2_inode_lock_atime(struct inode *inode,
struct vfsmount *vfsmnt,
int *level);
-int ocfs2_inode_lock_full(struct inode *inode,
+int ocfs2_inode_lock_full_nested(struct inode *inode,
struct buffer_head **ret_bh,
int ex,
- int arg_flags);
+ int arg_flags,
+ int subclass);
int ocfs2_inode_lock_with_page(struct inode *inode,
struct buffer_head **ret_bh,
int ex,
struct page *page);
+/* Variants without special locking class or flags */
+#define ocfs2_inode_lock_full(i, r, e, f)\
+ ocfs2_inode_lock_full_nested(i, r, e, f, OI_LS_NORMAL)
+#define ocfs2_inode_lock_nested(i, b, e, s)\
+ ocfs2_inode_lock_full_nested(i, b, e, 0, s)
/* 99% of the time we don't want to supply any additional flags --
* those are for very specific cases only. */
-#define ocfs2_inode_lock(i, b, e) ocfs2_inode_lock_full(i, b, e, 0)
+#define ocfs2_inode_lock(i, b, e) ocfs2_inode_lock_full_nested(i, b, e, 0, OI_LS_NORMAL)
void ocfs2_inode_unlock(struct inode *inode,
int ex);
int ocfs2_super_lock(struct ocfs2_super *osb,
int ex);
void ocfs2_super_unlock(struct ocfs2_super *osb,
int ex);
-int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex);
-void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno, int ex);
+int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno);
+void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno);
int ocfs2_rename_lock(struct ocfs2_super *osb);
void ocfs2_rename_unlock(struct ocfs2_super *osb);
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 07267e0..aa501d3 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1851,6 +1851,7 @@ relock:
if (ret)
goto out_dio;
+ count = ocount;
ret = generic_write_checks(file, ppos, &count,
S_ISBLK(inode->i_mode));
if (ret)
@@ -1918,8 +1919,10 @@ out_sems:
mutex_unlock(&inode->i_mutex);
+ if (written)
+ ret = written;
mlog_exit(ret);
- return written ? written : ret;
+ return ret;
}
static int ocfs2_splice_to_file(struct pipe_inode_info *pipe,
@@ -2026,7 +2029,7 @@ static ssize_t ocfs2_file_splice_read(struct file *in,
size_t len,
unsigned int flags)
{
- int ret = 0;
+ int ret = 0, lock_level = 0;
struct inode *inode = in->f_path.dentry->d_inode;
mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", in, pipe,
@@ -2037,12 +2040,12 @@ static ssize_t ocfs2_file_splice_read(struct file *in,
/*
* See the comment in ocfs2_file_aio_read()
*/
- ret = ocfs2_inode_lock(inode, NULL, 0);
+ ret = ocfs2_inode_lock_atime(inode, in->f_vfsmnt, &lock_level);
if (ret < 0) {
mlog_errno(ret);
goto bail;
}
- ocfs2_inode_unlock(inode, 0);
+ ocfs2_inode_unlock(inode, lock_level);
ret = generic_file_splice_read(in, ppos, pipe, len, flags);
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 10e1fa87..4dc8890 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -215,6 +215,8 @@ bail:
static int ocfs2_init_locked_inode(struct inode *inode, void *opaque)
{
struct ocfs2_find_inode_args *args = opaque;
+ static struct lock_class_key ocfs2_quota_ip_alloc_sem_key,
+ ocfs2_file_ip_alloc_sem_key;
mlog_entry("inode = %p, opaque = %p\n", inode, opaque);
@@ -223,6 +225,15 @@ static int ocfs2_init_locked_inode(struct inode *inode, void *opaque)
if (args->fi_sysfile_type != 0)
lockdep_set_class(&inode->i_mutex,
&ocfs2_sysfile_lock_key[args->fi_sysfile_type]);
+ if (args->fi_sysfile_type == USER_QUOTA_SYSTEM_INODE ||
+ args->fi_sysfile_type == GROUP_QUOTA_SYSTEM_INODE ||
+ args->fi_sysfile_type == LOCAL_USER_QUOTA_SYSTEM_INODE ||
+ args->fi_sysfile_type == LOCAL_GROUP_QUOTA_SYSTEM_INODE)
+ lockdep_set_class(&OCFS2_I(inode)->ip_alloc_sem,
+ &ocfs2_quota_ip_alloc_sem_key);
+ else
+ lockdep_set_class(&OCFS2_I(inode)->ip_alloc_sem,
+ &ocfs2_file_ip_alloc_sem_key);
mlog_exit(0);
return 0;
diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c
index 9fcd36d..467b413 100644
--- a/fs/ocfs2/ioctl.c
+++ b/fs/ocfs2/ioctl.c
@@ -7,7 +7,6 @@
#include <linux/fs.h>
#include <linux/mount.h>
-#include <linux/smp_lock.h>
#define MLOG_MASK_PREFIX ML_INODE
#include <cluster/masklog.h>
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 4a3b9e6..c48b93a 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1880,13 +1880,20 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
os = &osb->osb_orphan_scan;
- status = ocfs2_orphan_scan_lock(osb, &seqno, DLM_LOCK_EX);
+ if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
+ goto out;
+
+ status = ocfs2_orphan_scan_lock(osb, &seqno);
if (status < 0) {
if (status != -EAGAIN)
mlog_errno(status);
goto out;
}
+ /* Do no queue the tasks if the volume is being umounted */
+ if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
+ goto unlock;
+
if (os->os_seqno != seqno) {
os->os_seqno = seqno;
goto unlock;
@@ -1903,7 +1910,7 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
os->os_count++;
os->os_scantime = CURRENT_TIME;
unlock:
- ocfs2_orphan_scan_unlock(osb, seqno, DLM_LOCK_EX);
+ ocfs2_orphan_scan_unlock(osb, seqno);
out:
return;
}
@@ -1920,8 +1927,9 @@ void ocfs2_orphan_scan_work(struct work_struct *work)
mutex_lock(&os->os_lock);
ocfs2_queue_orphan_scan(osb);
- schedule_delayed_work(&os->os_orphan_scan_work,
- ocfs2_orphan_scan_timeout());
+ if (atomic_read(&os->os_state) == ORPHAN_SCAN_ACTIVE)
+ schedule_delayed_work(&os->os_orphan_scan_work,
+ ocfs2_orphan_scan_timeout());
mutex_unlock(&os->os_lock);
}
@@ -1930,26 +1938,39 @@ void ocfs2_orphan_scan_stop(struct ocfs2_super *osb)
struct ocfs2_orphan_scan *os;
os = &osb->osb_orphan_scan;
- mutex_lock(&os->os_lock);
- cancel_delayed_work(&os->os_orphan_scan_work);
- mutex_unlock(&os->os_lock);
+ if (atomic_read(&os->os_state) == ORPHAN_SCAN_ACTIVE) {
+ atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE);
+ mutex_lock(&os->os_lock);
+ cancel_delayed_work(&os->os_orphan_scan_work);
+ mutex_unlock(&os->os_lock);
+ }
}
-int ocfs2_orphan_scan_init(struct ocfs2_super *osb)
+void ocfs2_orphan_scan_init(struct ocfs2_super *osb)
{
struct ocfs2_orphan_scan *os;
os = &osb->osb_orphan_scan;
os->os_osb = osb;
os->os_count = 0;
- os->os_scantime = CURRENT_TIME;
+ os->os_seqno = 0;
mutex_init(&os->os_lock);
+ INIT_DELAYED_WORK(&os->os_orphan_scan_work, ocfs2_orphan_scan_work);
+}
- INIT_DELAYED_WORK(&os->os_orphan_scan_work,
- ocfs2_orphan_scan_work);
- schedule_delayed_work(&os->os_orphan_scan_work,
- ocfs2_orphan_scan_timeout());
- return 0;
+void ocfs2_orphan_scan_start(struct ocfs2_super *osb)
+{
+ struct ocfs2_orphan_scan *os;
+
+ os = &osb->osb_orphan_scan;
+ os->os_scantime = CURRENT_TIME;
+ if (ocfs2_is_hard_readonly(osb) || ocfs2_mount_local(osb))
+ atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE);
+ else {
+ atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
+ schedule_delayed_work(&os->os_orphan_scan_work,
+ ocfs2_orphan_scan_timeout());
+ }
}
struct ocfs2_orphan_filldir_priv {
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h
index 61045ee..2c3222a 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -144,7 +144,8 @@ static inline void ocfs2_inode_set_new(struct ocfs2_super *osb,
}
/* Exported only for the journal struct init code in super.c. Do not call. */
-int ocfs2_orphan_scan_init(struct ocfs2_super *osb);
+void ocfs2_orphan_scan_init(struct ocfs2_super *osb);
+void ocfs2_orphan_scan_start(struct ocfs2_super *osb);
void ocfs2_orphan_scan_stop(struct ocfs2_super *osb);
void ocfs2_orphan_scan_exit(struct ocfs2_super *osb);
@@ -329,20 +330,27 @@ int ocfs2_journal_dirty(handle_t *handle,
/* extended attribute block update */
#define OCFS2_XATTR_BLOCK_UPDATE_CREDITS 1
+/* Update of a single quota block */
+#define OCFS2_QUOTA_BLOCK_UPDATE_CREDITS 1
+
/* global quotafile inode update, data block */
-#define OCFS2_QINFO_WRITE_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 1)
+#define OCFS2_QINFO_WRITE_CREDITS (OCFS2_INODE_UPDATE_CREDITS + \
+ OCFS2_QUOTA_BLOCK_UPDATE_CREDITS)
+#define OCFS2_LOCAL_QINFO_WRITE_CREDITS OCFS2_QUOTA_BLOCK_UPDATE_CREDITS
/*
* The two writes below can accidentally see global info dirty due
* to set_info() quotactl so make them prepared for the writes.
*/
/* quota data block, global info */
/* Write to local quota file */
-#define OCFS2_QWRITE_CREDITS (OCFS2_QINFO_WRITE_CREDITS + 1)
+#define OCFS2_QWRITE_CREDITS (OCFS2_QINFO_WRITE_CREDITS + \
+ OCFS2_QUOTA_BLOCK_UPDATE_CREDITS)
/* global quota data block, local quota data block, global quota inode,
* global quota info */
-#define OCFS2_QSYNC_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 3)
+#define OCFS2_QSYNC_CREDITS (OCFS2_QINFO_WRITE_CREDITS + \
+ 2 * OCFS2_QUOTA_BLOCK_UPDATE_CREDITS)
static inline int ocfs2_quota_trans_credits(struct super_block *sb)
{
@@ -355,11 +363,6 @@ static inline int ocfs2_quota_trans_credits(struct super_block *sb)
return credits;
}
-/* Number of credits needed for removing quota structure from file */
-int ocfs2_calc_qdel_credits(struct super_block *sb, int type);
-/* Number of credits needed for initialization of new quota structure */
-int ocfs2_calc_qinit_credits(struct super_block *sb, int type);
-
/* group extend. inode update and last group update. */
#define OCFS2_GROUP_EXTEND_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 1)
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 33464c6..8601f93 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -118,7 +118,7 @@ static struct dentry *ocfs2_lookup(struct inode *dir, struct dentry *dentry,
mlog(0, "find name %.*s in directory %llu\n", dentry->d_name.len,
dentry->d_name.name, (unsigned long long)OCFS2_I(dir)->ip_blkno);
- status = ocfs2_inode_lock(dir, NULL, 0);
+ status = ocfs2_inode_lock_nested(dir, NULL, 0, OI_LS_PARENT);
if (status < 0) {
if (status != -ENOENT)
mlog_errno(status);
@@ -636,7 +636,7 @@ static int ocfs2_link(struct dentry *old_dentry,
if (S_ISDIR(inode->i_mode))
return -EPERM;
- err = ocfs2_inode_lock(dir, &parent_fe_bh, 1);
+ err = ocfs2_inode_lock_nested(dir, &parent_fe_bh, 1, OI_LS_PARENT);
if (err < 0) {
if (err != -ENOENT)
mlog_errno(err);
@@ -800,7 +800,8 @@ static int ocfs2_unlink(struct inode *dir,
return -EPERM;
}
- status = ocfs2_inode_lock(dir, &parent_node_bh, 1);
+ status = ocfs2_inode_lock_nested(dir, &parent_node_bh, 1,
+ OI_LS_PARENT);
if (status < 0) {
if (status != -ENOENT)
mlog_errno(status);
@@ -978,7 +979,8 @@ static int ocfs2_double_lock(struct ocfs2_super *osb,
inode1 = tmpinode;
}
/* lock id2 */
- status = ocfs2_inode_lock(inode2, bh2, 1);
+ status = ocfs2_inode_lock_nested(inode2, bh2, 1,
+ OI_LS_RENAME1);
if (status < 0) {
if (status != -ENOENT)
mlog_errno(status);
@@ -987,7 +989,7 @@ static int ocfs2_double_lock(struct ocfs2_super *osb,
}
/* lock id1 */
- status = ocfs2_inode_lock(inode1, bh1, 1);
+ status = ocfs2_inode_lock_nested(inode1, bh1, 1, OI_LS_RENAME2);
if (status < 0) {
/*
* An error return must mean that no cluster locks
@@ -1103,7 +1105,8 @@ static int ocfs2_rename(struct inode *old_dir,
* won't have to concurrently downconvert the inode and the
* dentry locks.
*/
- status = ocfs2_inode_lock(old_inode, &old_inode_bh, 1);
+ status = ocfs2_inode_lock_nested(old_inode, &old_inode_bh, 1,
+ OI_LS_PARENT);
if (status < 0) {
if (status != -ENOENT)
mlog_errno(status);
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index 18c1d9e..39e1d5a 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -34,6 +34,7 @@
#include <linux/workqueue.h>
#include <linux/kref.h>
#include <linux/mutex.h>
+#include <linux/lockdep.h>
#ifndef CONFIG_OCFS2_COMPAT_JBD
# include <linux/jbd2.h>
#else
@@ -152,6 +153,14 @@ struct ocfs2_lock_res {
unsigned int l_lock_max_exmode; /* Max wait for EX */
unsigned int l_lock_refresh; /* Disk refreshes */
#endif
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+ struct lockdep_map l_lockdep_map;
+#endif
+};
+
+enum ocfs2_orphan_scan_state {
+ ORPHAN_SCAN_ACTIVE,
+ ORPHAN_SCAN_INACTIVE
};
struct ocfs2_orphan_scan {
@@ -162,6 +171,7 @@ struct ocfs2_orphan_scan {
struct timespec os_scantime; /* time this node ran the scan */
u32 os_count; /* tracks node specific scans */
u32 os_seqno; /* tracks cluster wide scans */
+ atomic_t os_state; /* ACTIVE or INACTIVE */
};
struct ocfs2_dlm_debug {
@@ -214,10 +224,12 @@ enum ocfs2_mount_options
OCFS2_MOUNT_GRPQUOTA = 1 << 10, /* We support group quotas */
};
-#define OCFS2_OSB_SOFT_RO 0x0001
-#define OCFS2_OSB_HARD_RO 0x0002
-#define OCFS2_OSB_ERROR_FS 0x0004
-#define OCFS2_DEFAULT_ATIME_QUANTUM 60
+#define OCFS2_OSB_SOFT_RO 0x0001
+#define OCFS2_OSB_HARD_RO 0x0002
+#define OCFS2_OSB_ERROR_FS 0x0004
+#define OCFS2_OSB_DROP_DENTRY_LOCK_IMMED 0x0008
+
+#define OCFS2_DEFAULT_ATIME_QUANTUM 60
struct ocfs2_journal;
struct ocfs2_slot_info;
@@ -480,6 +492,18 @@ static inline void ocfs2_set_osb_flag(struct ocfs2_super *osb,
spin_unlock(&osb->osb_lock);
}
+
+static inline unsigned long ocfs2_test_osb_flag(struct ocfs2_super *osb,
+ unsigned long flag)
+{
+ unsigned long ret;
+
+ spin_lock(&osb->osb_lock);
+ ret = osb->osb_flags & flag;
+ spin_unlock(&osb->osb_lock);
+ return ret;
+}
+
static inline void ocfs2_set_ro_flag(struct ocfs2_super *osb,
int hard)
{
diff --git a/fs/ocfs2/quota.h b/fs/ocfs2/quota.h
index 7365e2e..3fb96fcd 100644
--- a/fs/ocfs2/quota.h
+++ b/fs/ocfs2/quota.h
@@ -50,7 +50,6 @@ struct ocfs2_mem_dqinfo {
unsigned int dqi_chunks; /* Number of chunks in local quota file */
unsigned int dqi_blocks; /* Number of blocks allocated for local quota file */
unsigned int dqi_syncms; /* How often should we sync with other nodes */
- unsigned int dqi_syncjiff; /* Precomputed dqi_syncms in jiffies */
struct list_head dqi_chunk; /* List of chunks */
struct inode *dqi_gqinode; /* Global quota file inode */
struct ocfs2_lock_res dqi_gqlock; /* Lock protecting quota information structure */
diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c
index edfa60c..bf7742d 100644
--- a/fs/ocfs2/quota_global.c
+++ b/fs/ocfs2/quota_global.c
@@ -69,6 +69,7 @@ static void ocfs2_global_mem2diskdqb(void *dp, struct dquot *dquot)
d->dqb_curspace = cpu_to_le64(m->dqb_curspace);
d->dqb_btime = cpu_to_le64(m->dqb_btime);
d->dqb_itime = cpu_to_le64(m->dqb_itime);
+ d->dqb_pad1 = d->dqb_pad2 = 0;
}
static int ocfs2_global_is_id(void *dp, struct dquot *dquot)
@@ -211,14 +212,13 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
mutex_lock_nested(&gqinode->i_mutex, I_MUTEX_QUOTA);
if (gqinode->i_size < off + len) {
- down_write(&OCFS2_I(gqinode)->ip_alloc_sem);
- err = ocfs2_extend_no_holes(gqinode, off + len, off);
- up_write(&OCFS2_I(gqinode)->ip_alloc_sem);
- if (err < 0)
- goto out;
+ loff_t rounded_end =
+ ocfs2_align_bytes_to_blocks(sb, off + len);
+
+ /* Space is already allocated in ocfs2_global_read_dquot() */
err = ocfs2_simple_size_update(gqinode,
oinfo->dqi_gqi_bh,
- off + len);
+ rounded_end);
if (err < 0)
goto out;
new = 1;
@@ -234,7 +234,7 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
}
if (err) {
mlog_errno(err);
- return err;
+ goto out;
}
lock_buffer(bh);
if (new)
@@ -342,7 +342,6 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace);
info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace);
oinfo->dqi_syncms = le32_to_cpu(dinfo.dqi_syncms);
- oinfo->dqi_syncjiff = msecs_to_jiffies(oinfo->dqi_syncms);
oinfo->dqi_gi.dqi_blocks = le32_to_cpu(dinfo.dqi_blocks);
oinfo->dqi_gi.dqi_free_blk = le32_to_cpu(dinfo.dqi_free_blk);
oinfo->dqi_gi.dqi_free_entry = le32_to_cpu(dinfo.dqi_free_entry);
@@ -352,7 +351,7 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
oinfo->dqi_gi.dqi_qtree_depth = qtree_depth(&oinfo->dqi_gi);
INIT_DELAYED_WORK(&oinfo->dqi_sync_work, qsync_work_fn);
queue_delayed_work(ocfs2_quota_wq, &oinfo->dqi_sync_work,
- oinfo->dqi_syncjiff);
+ msecs_to_jiffies(oinfo->dqi_syncms));
out_err:
mlog_exit(status);
@@ -402,13 +401,36 @@ int ocfs2_global_write_info(struct super_block *sb, int type)
return err;
}
+static int ocfs2_global_qinit_alloc(struct super_block *sb, int type)
+{
+ struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(sb, type)->dqi_priv;
+
+ /*
+ * We may need to allocate tree blocks and a leaf block but not the
+ * root block
+ */
+ return oinfo->dqi_gi.dqi_qtree_depth;
+}
+
+static int ocfs2_calc_global_qinit_credits(struct super_block *sb, int type)
+{
+ /* We modify all the allocated blocks, tree root, and info block */
+ return (ocfs2_global_qinit_alloc(sb, type) + 2) *
+ OCFS2_QUOTA_BLOCK_UPDATE_CREDITS;
+}
+
/* Read in information from global quota file and acquire a reference to it.
* dquot_acquire() has already started the transaction and locked quota file */
int ocfs2_global_read_dquot(struct dquot *dquot)
{
int err, err2, ex = 0;
- struct ocfs2_mem_dqinfo *info =
- sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
+ struct super_block *sb = dquot->dq_sb;
+ int type = dquot->dq_type;
+ struct ocfs2_mem_dqinfo *info = sb_dqinfo(sb, type)->dqi_priv;
+ struct ocfs2_super *osb = OCFS2_SB(sb);
+ struct inode *gqinode = info->dqi_gqinode;
+ int need_alloc = ocfs2_global_qinit_alloc(sb, type);
+ handle_t *handle = NULL;
err = ocfs2_qinfo_lock(info, 0);
if (err < 0)
@@ -419,14 +441,33 @@ int ocfs2_global_read_dquot(struct dquot *dquot)
OCFS2_DQUOT(dquot)->dq_use_count++;
OCFS2_DQUOT(dquot)->dq_origspace = dquot->dq_dqb.dqb_curspace;
OCFS2_DQUOT(dquot)->dq_originodes = dquot->dq_dqb.dqb_curinodes;
+ ocfs2_qinfo_unlock(info, 0);
+
if (!dquot->dq_off) { /* No real quota entry? */
- /* Upgrade to exclusive lock for allocation */
- ocfs2_qinfo_unlock(info, 0);
- err = ocfs2_qinfo_lock(info, 1);
- if (err < 0)
- goto out_qlock;
ex = 1;
+ /*
+ * Add blocks to quota file before we start a transaction since
+ * locking allocators ranks above a transaction start
+ */
+ WARN_ON(journal_current_handle());
+ down_write(&OCFS2_I(gqinode)->ip_alloc_sem);
+ err = ocfs2_extend_no_holes(gqinode,
+ gqinode->i_size + (need_alloc << sb->s_blocksize_bits),
+ gqinode->i_size);
+ up_write(&OCFS2_I(gqinode)->ip_alloc_sem);
+ if (err < 0)
+ goto out;
}
+
+ handle = ocfs2_start_trans(osb,
+ ocfs2_calc_global_qinit_credits(sb, type));
+ if (IS_ERR(handle)) {
+ err = PTR_ERR(handle);
+ goto out;
+ }
+ err = ocfs2_qinfo_lock(info, ex);
+ if (err < 0)
+ goto out_trans;
err = qtree_write_dquot(&info->dqi_gi, dquot);
if (ex && info_dirty(sb_dqinfo(dquot->dq_sb, dquot->dq_type))) {
err2 = __ocfs2_global_write_info(dquot->dq_sb, dquot->dq_type);
@@ -438,6 +479,9 @@ out_qlock:
ocfs2_qinfo_unlock(info, 1);
else
ocfs2_qinfo_unlock(info, 0);
+out_trans:
+ if (handle)
+ ocfs2_commit_trans(osb, handle);
out:
if (err < 0)
mlog_errno(err);
@@ -607,7 +651,7 @@ static void qsync_work_fn(struct work_struct *work)
dquot_scan_active(sb, ocfs2_sync_dquot_helper, oinfo->dqi_type);
queue_delayed_work(ocfs2_quota_wq, &oinfo->dqi_sync_work,
- oinfo->dqi_syncjiff);
+ msecs_to_jiffies(oinfo->dqi_syncms));
}
/*
@@ -635,20 +679,18 @@ out:
return status;
}
-int ocfs2_calc_qdel_credits(struct super_block *sb, int type)
+static int ocfs2_calc_qdel_credits(struct super_block *sb, int type)
{
- struct ocfs2_mem_dqinfo *oinfo;
- int features[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA,
- OCFS2_FEATURE_RO_COMPAT_GRPQUOTA };
-
- if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, features[type]))
- return 0;
-
- oinfo = sb_dqinfo(sb, type)->dqi_priv;
- /* We modify tree, leaf block, global info, local chunk header,
- * global and local inode */
- return oinfo->dqi_gi.dqi_qtree_depth + 2 + 1 +
- 2 * OCFS2_INODE_UPDATE_CREDITS;
+ struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(sb, type)->dqi_priv;
+ /*
+ * We modify tree, leaf block, global info, local chunk header,
+ * global and local inode; OCFS2_QINFO_WRITE_CREDITS already
+ * accounts for inode update
+ */
+ return (oinfo->dqi_gi.dqi_qtree_depth + 2) *
+ OCFS2_QUOTA_BLOCK_UPDATE_CREDITS +
+ OCFS2_QINFO_WRITE_CREDITS +
+ OCFS2_INODE_UPDATE_CREDITS;
}
static int ocfs2_release_dquot(struct dquot *dquot)
@@ -680,33 +722,10 @@ out:
return status;
}
-int ocfs2_calc_qinit_credits(struct super_block *sb, int type)
-{
- struct ocfs2_mem_dqinfo *oinfo;
- int features[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA,
- OCFS2_FEATURE_RO_COMPAT_GRPQUOTA };
- struct ocfs2_dinode *lfe, *gfe;
-
- if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, features[type]))
- return 0;
-
- oinfo = sb_dqinfo(sb, type)->dqi_priv;
- gfe = (struct ocfs2_dinode *)oinfo->dqi_gqi_bh->b_data;
- lfe = (struct ocfs2_dinode *)oinfo->dqi_lqi_bh->b_data;
- /* We can extend local file + global file. In local file we
- * can modify info, chunk header block and dquot block. In
- * global file we can modify info, tree and leaf block */
- return ocfs2_calc_extend_credits(sb, &lfe->id2.i_list, 0) +
- ocfs2_calc_extend_credits(sb, &gfe->id2.i_list, 0) +
- 3 + oinfo->dqi_gi.dqi_qtree_depth + 2;
-}
-
static int ocfs2_acquire_dquot(struct dquot *dquot)
{
- handle_t *handle;
struct ocfs2_mem_dqinfo *oinfo =
sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
- struct ocfs2_super *osb = OCFS2_SB(dquot->dq_sb);
int status = 0;
mlog_entry("id=%u, type=%d", dquot->dq_id, dquot->dq_type);
@@ -715,16 +734,7 @@ static int ocfs2_acquire_dquot(struct dquot *dquot)
status = ocfs2_lock_global_qf(oinfo, 1);
if (status < 0)
goto out;
- handle = ocfs2_start_trans(osb,
- ocfs2_calc_qinit_credits(dquot->dq_sb, dquot->dq_type));
- if (IS_ERR(handle)) {
- status = PTR_ERR(handle);
- mlog_errno(status);
- goto out_ilock;
- }
status = dquot_acquire(dquot);
- ocfs2_commit_trans(osb, handle);
-out_ilock:
ocfs2_unlock_global_qf(oinfo, 1);
out:
mlog_exit(status);
diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
index 5a460fa..bdb09cb 100644
--- a/fs/ocfs2/quota_local.c
+++ b/fs/ocfs2/quota_local.c
@@ -20,6 +20,7 @@
#include "sysfile.h"
#include "dlmglue.h"
#include "quota.h"
+#include "uptodate.h"
/* Number of local quota structures per block */
static inline unsigned int ol_quota_entries_per_block(struct super_block *sb)
@@ -100,7 +101,8 @@ static int ocfs2_modify_bh(struct inode *inode, struct buffer_head *bh,
handle_t *handle;
int status;
- handle = ocfs2_start_trans(OCFS2_SB(sb), 1);
+ handle = ocfs2_start_trans(OCFS2_SB(sb),
+ OCFS2_QUOTA_BLOCK_UPDATE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
@@ -610,7 +612,8 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb,
goto out_bh;
/* Mark quota file as clean if we are recovering quota file of
* some other node. */
- handle = ocfs2_start_trans(osb, 1);
+ handle = ocfs2_start_trans(osb,
+ OCFS2_LOCAL_QINFO_WRITE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
@@ -940,7 +943,7 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk(
struct ocfs2_local_disk_chunk *dchunk;
int status;
handle_t *handle;
- struct buffer_head *bh = NULL;
+ struct buffer_head *bh = NULL, *dbh = NULL;
u64 p_blkno;
/* We are protected by dqio_sem so no locking needed */
@@ -964,32 +967,35 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk(
mlog_errno(status);
goto out;
}
+ /* Local quota info and two new blocks we initialize */
+ handle = ocfs2_start_trans(OCFS2_SB(sb),
+ OCFS2_LOCAL_QINFO_WRITE_CREDITS +
+ 2 * OCFS2_QUOTA_BLOCK_UPDATE_CREDITS);
+ if (IS_ERR(handle)) {
+ status = PTR_ERR(handle);
+ mlog_errno(status);
+ goto out;
+ }
+ /* Initialize chunk header */
down_read(&OCFS2_I(lqinode)->ip_alloc_sem);
status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks,
&p_blkno, NULL, NULL);
up_read(&OCFS2_I(lqinode)->ip_alloc_sem);
if (status < 0) {
mlog_errno(status);
- goto out;
+ goto out_trans;
}
bh = sb_getblk(sb, p_blkno);
if (!bh) {
status = -ENOMEM;
mlog_errno(status);
- goto out;
+ goto out_trans;
}
dchunk = (struct ocfs2_local_disk_chunk *)bh->b_data;
-
- handle = ocfs2_start_trans(OCFS2_SB(sb), 2);
- if (IS_ERR(handle)) {
- status = PTR_ERR(handle);
- mlog_errno(status);
- goto out;
- }
-
+ ocfs2_set_new_buffer_uptodate(lqinode, bh);
status = ocfs2_journal_access_dq(handle, lqinode, bh,
- OCFS2_JOURNAL_ACCESS_WRITE);
+ OCFS2_JOURNAL_ACCESS_CREATE);
if (status < 0) {
mlog_errno(status);
goto out_trans;
@@ -999,7 +1005,6 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk(
memset(dchunk->dqc_bitmap, 0,
sb->s_blocksize - sizeof(struct ocfs2_local_disk_chunk) -
OCFS2_QBLK_RESERVED_SPACE);
- set_buffer_uptodate(bh);
unlock_buffer(bh);
status = ocfs2_journal_dirty(handle, bh);
if (status < 0) {
@@ -1007,6 +1012,38 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk(
goto out_trans;
}
+ /* Initialize new block with structures */
+ down_read(&OCFS2_I(lqinode)->ip_alloc_sem);
+ status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks + 1,
+ &p_blkno, NULL, NULL);
+ up_read(&OCFS2_I(lqinode)->ip_alloc_sem);
+ if (status < 0) {
+ mlog_errno(status);
+ goto out_trans;
+ }
+ dbh = sb_getblk(sb, p_blkno);
+ if (!dbh) {
+ status = -ENOMEM;
+ mlog_errno(status);
+ goto out_trans;
+ }
+ ocfs2_set_new_buffer_uptodate(lqinode, dbh);
+ status = ocfs2_journal_access_dq(handle, lqinode, dbh,
+ OCFS2_JOURNAL_ACCESS_CREATE);
+ if (status < 0) {
+ mlog_errno(status);
+ goto out_trans;
+ }
+ lock_buffer(dbh);
+ memset(dbh->b_data, 0, sb->s_blocksize - OCFS2_QBLK_RESERVED_SPACE);
+ unlock_buffer(dbh);
+ status = ocfs2_journal_dirty(handle, dbh);
+ if (status < 0) {
+ mlog_errno(status);
+ goto out_trans;
+ }
+
+ /* Update local quotafile info */
oinfo->dqi_blocks += 2;
oinfo->dqi_chunks++;
status = ocfs2_local_write_info(sb, type);
@@ -1031,6 +1068,7 @@ out_trans:
ocfs2_commit_trans(OCFS2_SB(sb), handle);
out:
brelse(bh);
+ brelse(dbh);
kmem_cache_free(ocfs2_qf_chunk_cachep, chunk);
return ERR_PTR(status);
}
@@ -1048,6 +1086,8 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file(
struct ocfs2_local_disk_chunk *dchunk;
int epb = ol_quota_entries_per_block(sb);
unsigned int chunk_blocks;
+ struct buffer_head *bh;
+ u64 p_blkno;
int status;
handle_t *handle;
@@ -1075,12 +1115,49 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file(
mlog_errno(status);
goto out;
}
- handle = ocfs2_start_trans(OCFS2_SB(sb), 2);
+
+ /* Get buffer from the just added block */
+ down_read(&OCFS2_I(lqinode)->ip_alloc_sem);
+ status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks,
+ &p_blkno, NULL, NULL);
+ up_read(&OCFS2_I(lqinode)->ip_alloc_sem);
+ if (status < 0) {
+ mlog_errno(status);
+ goto out;
+ }
+ bh = sb_getblk(sb, p_blkno);
+ if (!bh) {
+ status = -ENOMEM;
+ mlog_errno(status);
+ goto out;
+ }
+ ocfs2_set_new_buffer_uptodate(lqinode, bh);
+
+ /* Local quota info, chunk header and the new block we initialize */
+ handle = ocfs2_start_trans(OCFS2_SB(sb),
+ OCFS2_LOCAL_QINFO_WRITE_CREDITS +
+ 2 * OCFS2_QUOTA_BLOCK_UPDATE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
goto out;
}
+ /* Zero created block */
+ status = ocfs2_journal_access_dq(handle, lqinode, bh,
+ OCFS2_JOURNAL_ACCESS_CREATE);
+ if (status < 0) {
+ mlog_errno(status);
+ goto out_trans;
+ }
+ lock_buffer(bh);
+ memset(bh->b_data, 0, sb->s_blocksize);
+ unlock_buffer(bh);
+ status = ocfs2_journal_dirty(handle, bh);
+ if (status < 0) {
+ mlog_errno(status);
+ goto out_trans;
+ }
+ /* Update chunk header */
status = ocfs2_journal_access_dq(handle, lqinode, chunk->qc_headerbh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
@@ -1097,6 +1174,7 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file(
mlog_errno(status);
goto out_trans;
}
+ /* Update file header */
oinfo->dqi_blocks++;
status = ocfs2_local_write_info(sb, type);
if (status < 0) {
diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c
index fcd120f..e49c410 100644
--- a/fs/ocfs2/stack_o2cb.c
+++ b/fs/ocfs2/stack_o2cb.c
@@ -17,6 +17,7 @@
* General Public License for more details.
*/
+#include <linux/kernel.h>
#include <linux/crc32.h>
#include <linux/module.h>
@@ -153,7 +154,7 @@ static int status_map[] = {
static int dlm_status_to_errno(enum dlm_status status)
{
- BUG_ON(status > (sizeof(status_map) / sizeof(status_map[0])));
+ BUG_ON(status < 0 || status >= ARRAY_SIZE(status_map));
return status_map[status];
}
@@ -236,6 +237,16 @@ static int o2cb_dlm_lock_status(union ocfs2_dlm_lksb *lksb)
return dlm_status_to_errno(lksb->lksb_o2dlm.status);
}
+/*
+ * o2dlm aways has a "valid" LVB. If the dlm loses track of the LVB
+ * contents, it will zero out the LVB. Thus the caller can always trust
+ * the contents.
+ */
+static int o2cb_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb)
+{
+ return 1;
+}
+
static void *o2cb_dlm_lvb(union ocfs2_dlm_lksb *lksb)
{
return (void *)(lksb->lksb_o2dlm.lvb);
@@ -354,6 +365,7 @@ static struct ocfs2_stack_operations o2cb_stack_ops = {
.dlm_lock = o2cb_dlm_lock,
.dlm_unlock = o2cb_dlm_unlock,
.lock_status = o2cb_dlm_lock_status,
+ .lvb_valid = o2cb_dlm_lvb_valid,
.lock_lvb = o2cb_dlm_lvb,
.dump_lksb = o2cb_dump_lksb,
};
diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c
index 9b76d41..ff4c798 100644
--- a/fs/ocfs2/stack_user.c
+++ b/fs/ocfs2/stack_user.c
@@ -738,6 +738,13 @@ static int user_dlm_lock_status(union ocfs2_dlm_lksb *lksb)
return lksb->lksb_fsdlm.sb_status;
}
+static int user_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb)
+{
+ int invalid = lksb->lksb_fsdlm.sb_flags & DLM_SBF_VALNOTVALID;
+
+ return !invalid;
+}
+
static void *user_dlm_lvb(union ocfs2_dlm_lksb *lksb)
{
if (!lksb->lksb_fsdlm.sb_lvbptr)
@@ -873,6 +880,7 @@ static struct ocfs2_stack_operations ocfs2_user_plugin_ops = {
.dlm_lock = user_dlm_lock,
.dlm_unlock = user_dlm_unlock,
.lock_status = user_dlm_lock_status,
+ .lvb_valid = user_dlm_lvb_valid,
.lock_lvb = user_dlm_lvb,
.plock = user_plock,
.dump_lksb = user_dlm_dump_lksb,
diff --git a/fs/ocfs2/stackglue.c b/fs/ocfs2/stackglue.c
index 68b668b..3f2f1c4 100644
--- a/fs/ocfs2/stackglue.c
+++ b/fs/ocfs2/stackglue.c
@@ -6,7 +6,7 @@
* Code which implements an OCFS2 specific interface to underlying
* cluster stacks.
*
- * Copyright (C) 2007 Oracle. All rights reserved.
+ * Copyright (C) 2007, 2009 Oracle. All rights reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public
@@ -271,11 +271,12 @@ int ocfs2_dlm_lock_status(union ocfs2_dlm_lksb *lksb)
}
EXPORT_SYMBOL_GPL(ocfs2_dlm_lock_status);
-/*
- * Why don't we cast to ocfs2_meta_lvb? The "clean" answer is that we
- * don't cast at the glue level. The real answer is that the header
- * ordering is nigh impossible.
- */
+int ocfs2_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb)
+{
+ return active_stack->sp_ops->lvb_valid(lksb);
+}
+EXPORT_SYMBOL_GPL(ocfs2_dlm_lvb_valid);
+
void *ocfs2_dlm_lvb(union ocfs2_dlm_lksb *lksb)
{
return active_stack->sp_ops->lock_lvb(lksb);
diff --git a/fs/ocfs2/stackglue.h b/fs/ocfs2/stackglue.h
index c571af3..03a44d6 100644
--- a/fs/ocfs2/stackglue.h
+++ b/fs/ocfs2/stackglue.h
@@ -186,6 +186,11 @@ struct ocfs2_stack_operations {
int (*lock_status)(union ocfs2_dlm_lksb *lksb);
/*
+ * Return non-zero if the LVB is valid.
+ */
+ int (*lvb_valid)(union ocfs2_dlm_lksb *lksb);
+
+ /*
* Pull the lvb pointer off of the stack-specific lksb.
*/
void *(*lock_lvb)(union ocfs2_dlm_lksb *lksb);
@@ -252,6 +257,7 @@ int ocfs2_dlm_unlock(struct ocfs2_cluster_connection *conn,
struct ocfs2_lock_res *astarg);
int ocfs2_dlm_lock_status(union ocfs2_dlm_lksb *lksb);
+int ocfs2_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb);
void *ocfs2_dlm_lvb(union ocfs2_dlm_lksb *lksb);
void ocfs2_dlm_dump_lksb(union ocfs2_dlm_lksb *lksb);
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index 8439f6b..73a16d4 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -923,14 +923,23 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh,
int nr)
{
struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data;
+ int ret;
if (ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap))
return 0;
- if (!buffer_jbd(bg_bh) || !bh2jh(bg_bh)->b_committed_data)
+
+ if (!buffer_jbd(bg_bh))
return 1;
+ jbd_lock_bh_state(bg_bh);
bg = (struct ocfs2_group_desc *) bh2jh(bg_bh)->b_committed_data;
- return !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
+ if (bg)
+ ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
+ else
+ ret = 1;
+ jbd_unlock_bh_state(bg_bh);
+
+ return ret;
}
static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb,
@@ -1885,6 +1894,7 @@ static inline int ocfs2_block_group_clear_bits(handle_t *handle,
unsigned int tmp;
int journal_type = OCFS2_JOURNAL_ACCESS_WRITE;
struct ocfs2_group_desc *undo_bg = NULL;
+ int cluster_bitmap = 0;
mlog_entry_void();
@@ -1905,18 +1915,28 @@ static inline int ocfs2_block_group_clear_bits(handle_t *handle,
}
if (ocfs2_is_cluster_bitmap(alloc_inode))
- undo_bg = (struct ocfs2_group_desc *) bh2jh(group_bh)->b_committed_data;
+ cluster_bitmap = 1;
+
+ if (cluster_bitmap) {
+ jbd_lock_bh_state(group_bh);
+ undo_bg = (struct ocfs2_group_desc *)
+ bh2jh(group_bh)->b_committed_data;
+ BUG_ON(!undo_bg);
+ }
tmp = num_bits;
while(tmp--) {
ocfs2_clear_bit((bit_off + tmp),
(unsigned long *) bg->bg_bitmap);
- if (ocfs2_is_cluster_bitmap(alloc_inode))
+ if (cluster_bitmap)
ocfs2_set_bit(bit_off + tmp,
(unsigned long *) undo_bg->bg_bitmap);
}
le16_add_cpu(&bg->bg_free_bits_count, num_bits);
+ if (cluster_bitmap)
+ jbd_unlock_bh_state(group_bh);
+
status = ocfs2_journal_dirty(handle, group_bh);
if (status < 0)
mlog_errno(status);
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 0d3ed74..b0ee0fd 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -205,11 +205,10 @@ static const match_table_t tokens = {
#ifdef CONFIG_DEBUG_FS
static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
{
- int out = 0;
- int i;
struct ocfs2_cluster_connection *cconn = osb->cconn;
struct ocfs2_recovery_map *rm = osb->recovery_map;
- struct ocfs2_orphan_scan *os;
+ struct ocfs2_orphan_scan *os = &osb->osb_orphan_scan;
+ int i, out = 0;
out += snprintf(buf + out, len - out,
"%10s => Id: %-s Uuid: %-s Gen: 0x%X Label: %-s\n",
@@ -234,20 +233,24 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
"%10s => Opts: 0x%lX AtimeQuanta: %u\n", "Mount",
osb->s_mount_opt, osb->s_atime_quantum);
- out += snprintf(buf + out, len - out,
- "%10s => Stack: %s Name: %*s Version: %d.%d\n",
- "Cluster",
- (*osb->osb_cluster_stack == '\0' ?
- "o2cb" : osb->osb_cluster_stack),
- cconn->cc_namelen, cconn->cc_name,
- cconn->cc_version.pv_major, cconn->cc_version.pv_minor);
+ if (cconn) {
+ out += snprintf(buf + out, len - out,
+ "%10s => Stack: %s Name: %*s "
+ "Version: %d.%d\n", "Cluster",
+ (*osb->osb_cluster_stack == '\0' ?
+ "o2cb" : osb->osb_cluster_stack),
+ cconn->cc_namelen, cconn->cc_name,
+ cconn->cc_version.pv_major,
+ cconn->cc_version.pv_minor);
+ }
spin_lock(&osb->dc_task_lock);
out += snprintf(buf + out, len - out,
"%10s => Pid: %d Count: %lu WakeSeq: %lu "
"WorkSeq: %lu\n", "DownCnvt",
- task_pid_nr(osb->dc_task), osb->blocked_lock_count,
- osb->dc_wake_sequence, osb->dc_work_sequence);
+ (osb->dc_task ? task_pid_nr(osb->dc_task) : -1),
+ osb->blocked_lock_count, osb->dc_wake_sequence,
+ osb->dc_work_sequence);
spin_unlock(&osb->dc_task_lock);
spin_lock(&osb->osb_lock);
@@ -267,14 +270,15 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
out += snprintf(buf + out, len - out,
"%10s => Pid: %d Interval: %lu Needs: %d\n", "Commit",
- task_pid_nr(osb->commit_task), osb->osb_commit_interval,
+ (osb->commit_task ? task_pid_nr(osb->commit_task) : -1),
+ osb->osb_commit_interval,
atomic_read(&osb->needs_checkpoint));
out += snprintf(buf + out, len - out,
- "%10s => State: %d NumTxns: %d TxnId: %lu\n",
+ "%10s => State: %d TxnId: %lu NumTxns: %d\n",
"Journal", osb->journal->j_state,
- atomic_read(&osb->journal->j_num_trans),
- osb->journal->j_trans_id);
+ osb->journal->j_trans_id,
+ atomic_read(&osb->journal->j_num_trans));
out += snprintf(buf + out, len - out,
"%10s => GlobalAllocs: %d LocalAllocs: %d "
@@ -300,9 +304,18 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
atomic_read(&osb->s_num_inodes_stolen));
spin_unlock(&osb->osb_lock);
+ out += snprintf(buf + out, len - out, "OrphanScan => ");
+ out += snprintf(buf + out, len - out, "Local: %u Global: %u ",
+ os->os_count, os->os_seqno);
+ out += snprintf(buf + out, len - out, " Last Scan: ");
+ if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
+ out += snprintf(buf + out, len - out, "Disabled\n");
+ else
+ out += snprintf(buf + out, len - out, "%lu seconds ago\n",
+ (get_seconds() - os->os_scantime.tv_sec));
+
out += snprintf(buf + out, len - out, "%10s => %3s %10s\n",
"Slots", "Num", "RecoGen");
-
for (i = 0; i < osb->max_slots; ++i) {
out += snprintf(buf + out, len - out,
"%10s %c %3d %10d\n",
@@ -311,13 +324,6 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
i, osb->slot_recovery_generations[i]);
}
- os = &osb->osb_orphan_scan;
- out += snprintf(buf + out, len - out, "Orphan Scan=> ");
- out += snprintf(buf + out, len - out, "Local: %u Global: %u ",
- os->os_count, os->os_seqno);
- out += snprintf(buf + out, len - out, " Last Scan: %lu seconds ago\n",
- (get_seconds() - os->os_scantime.tv_sec));
-
return out;
}
@@ -771,6 +777,7 @@ static int ocfs2_sb_probe(struct super_block *sb,
}
di = (struct ocfs2_dinode *) (*bh)->b_data;
memset(stats, 0, sizeof(struct ocfs2_blockcheck_stats));
+ spin_lock_init(&stats->b_lock);
status = ocfs2_verify_volume(di, *bh, blksize, stats);
if (status >= 0)
goto bail;
@@ -1175,6 +1182,9 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
atomic_set(&osb->vol_state, VOLUME_MOUNTED_QUOTAS);
wake_up(&osb->osb_mount_event);
+ /* Start this when the mount is almost sure of being successful */
+ ocfs2_orphan_scan_start(osb);
+
mlog_exit(status);
return status;
@@ -1204,14 +1214,27 @@ static int ocfs2_get_sb(struct file_system_type *fs_type,
mnt);
}
+static void ocfs2_kill_sb(struct super_block *sb)
+{
+ struct ocfs2_super *osb = OCFS2_SB(sb);
+
+ /* Prevent further queueing of inode drop events */
+ spin_lock(&dentry_list_lock);
+ ocfs2_set_osb_flag(osb, OCFS2_OSB_DROP_DENTRY_LOCK_IMMED);
+ spin_unlock(&dentry_list_lock);
+ /* Wait for work to finish and/or remove it */
+ cancel_work_sync(&osb->dentry_lock_work);
+
+ kill_block_super(sb);
+}
+
static struct file_system_type ocfs2_fs_type = {
.owner = THIS_MODULE,
.name = "ocfs2",
.get_sb = ocfs2_get_sb, /* is this called when we mount
* the fs? */
- .kill_sb = kill_block_super, /* set to the generic one
- * right now, but do we
- * need to change that? */
+ .kill_sb = ocfs2_kill_sb,
+
.fs_flags = FS_REQUIRES_DEV|FS_RENAME_DOES_D_MOVE,
.next = NULL
};
@@ -1810,14 +1833,21 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
debugfs_remove(osb->osb_ctxt);
+ /*
+ * Flush inode dropping work queue so that deletes are
+ * performed while the filesystem is still working
+ */
+ ocfs2_drop_all_dl_inodes(osb);
+
+ /* Orphan scan should be stopped as early as possible */
+ ocfs2_orphan_scan_stop(osb);
+
ocfs2_disable_quotas(osb);
ocfs2_shutdown_local_alloc(osb);
ocfs2_truncate_log_shutdown(osb);
- ocfs2_orphan_scan_stop(osb);
-
/* This will disable recovery and flush any recovery work. */
ocfs2_recovery_exit(osb);
@@ -1971,6 +2001,8 @@ static int ocfs2_initialize_super(struct super_block *sb,
snprintf(osb->dev_str, sizeof(osb->dev_str), "%u,%u",
MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev));
+ ocfs2_orphan_scan_init(osb);
+
status = ocfs2_recovery_init(osb);
if (status) {
mlog(ML_ERROR, "Unable to initialize recovery state\n");
@@ -1978,13 +2010,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
goto bail;
}
- status = ocfs2_orphan_scan_init(osb);
- if (status) {
- mlog(ML_ERROR, "Unable to initialize delayed orphan scan\n");
- mlog_errno(status);
- goto bail;
- }
-
init_waitqueue_head(&osb->checkpoint_event);
atomic_set(&osb->needs_checkpoint, 0);
diff --git a/fs/ocfs2/sysfile.c b/fs/ocfs2/sysfile.c
index ab713eb..40e5370 100644
--- a/fs/ocfs2/sysfile.c
+++ b/fs/ocfs2/sysfile.c
@@ -50,6 +50,10 @@ static inline int is_in_system_inode_array(struct ocfs2_super *osb,
int type,
u32 slot);
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+static struct lock_class_key ocfs2_sysfile_cluster_lock_key[NUM_SYSTEM_INODES];
+#endif
+
static inline int is_global_system_inode(int type)
{
return type >= OCFS2_FIRST_ONLINE_SYSTEM_INODE &&
@@ -118,6 +122,21 @@ static struct inode * _ocfs2_get_system_file_inode(struct ocfs2_super *osb,
inode = NULL;
goto bail;
}
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+ if (type == LOCAL_USER_QUOTA_SYSTEM_INODE ||
+ type == LOCAL_GROUP_QUOTA_SYSTEM_INODE ||
+ type == JOURNAL_SYSTEM_INODE) {
+ /* Ignore inode lock on these inodes as the lock does not
+ * really belong to any process and lockdep cannot handle
+ * that */
+ OCFS2_I(inode)->ip_inode_lockres.l_lockdep_map.key = NULL;
+ } else {
+ lockdep_init_map(&OCFS2_I(inode)->ip_inode_lockres.
+ l_lockdep_map,
+ ocfs2_system_inodes[type].si_name,
+ &ocfs2_sysfile_cluster_lock_key[type], 0);
+ }
+#endif
bail:
return inode;
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index ba320e2..d1a27cd 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -1052,7 +1052,8 @@ static int ocfs2_xattr_block_get(struct inode *inode,
struct ocfs2_xattr_block *xb;
struct ocfs2_xattr_value_root *xv;
size_t size;
- int ret = -ENODATA, name_offset, name_len, block_off, i;
+ int ret = -ENODATA, name_offset, name_len, i;
+ int uninitialized_var(block_off);
xs->bucket = ocfs2_xattr_bucket_new(inode);
if (!xs->bucket) {
OpenPOWER on IntegriCloud