From 1c520dfbf391e1617ef61553f815b8006a066c44 Mon Sep 17 00:00:00 2001
From: Joel Becker <joel.becker@oracle.com>
Date: Fri, 19 Jun 2009 15:14:13 -0700
Subject: [PATCH 01/10] ocfs2: Provide the ocfs2_dlm_lvb_valid() stack API.

The Lock Value Block (LVB) of a DLM lock can be lost when nodes die and
the DLM cannot reconstruct its state.  Clients of the DLM need to know
this.

ocfs2's internal DLM, o2dlm, explicitly zeroes out the LVB when it loses
track of the state.  This is not a standard behavior, but ocfs2 has
always relied on it.  Thus, an o2dlm LVB is always "valid".

ocfs2 now supports both o2dlm and fs/dlm via the stack glue.  When
fs/dlm loses track of an LVBs state, it sets a flag
(DLM_SBF_VALNOTVALID) on the Lock Status Block (LKSB).  The contents of
the LVB may be garbage or merely stale.

ocfs2 doesn't want to try to guess at the validity of the stale LVB.
Instead, it should be checking the VALNOTVALID flag.  As this is the
'standard' way of treating LVBs, we will promote this behavior.

We add a stack glue API ocfs2_dlm_lvb_valid().  It returns non-zero when
the LVB is valid.  o2dlm will always return valid, while fs/dlm will
check VALNOTVALID.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Acked-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/ocfs2/dlmglue.c    |  9 ++++++---
 fs/ocfs2/stack_o2cb.c | 11 +++++++++++
 fs/ocfs2/stack_user.c |  8 ++++++++
 fs/ocfs2/stackglue.c  | 13 +++++++------
 fs/ocfs2/stackglue.h  |  6 ++++++
 5 files changed, 38 insertions(+), 9 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 6cdeaa76f27f..83d2ddb27186 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -1989,7 +1989,8 @@ 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;
@@ -2382,7 +2383,8 @@ int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
 		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);
 	return status;
 }
@@ -3627,7 +3629,8 @@ 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/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c
index fcd120f1493a..3f661376a2de 100644
--- a/fs/ocfs2/stack_o2cb.c
+++ b/fs/ocfs2/stack_o2cb.c
@@ -236,6 +236,16 @@ 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 +364,7 @@ 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 9b76d41a8ac6..ff4c798a5635 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 68b668b0e60a..3f2f1c45b7b6 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 c571af375ef8..03a44d60eac9 100644
--- a/fs/ocfs2/stackglue.h
+++ b/fs/ocfs2/stackglue.h
@@ -185,6 +185,11 @@ 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.
 	 */
@@ -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);
 

From 1962f39abbb2d5643a7d59169422661a2d58793d Mon Sep 17 00:00:00 2001
From: Tao Ma <tao.ma@oracle.com>
Date: Fri, 19 Jun 2009 15:36:52 +0800
Subject: [PATCH 02/10] ocfs2: Update atime in splice read if necessary.

We should call ocfs2_inode_lock_atime instead of ocfs2_inode_lock
in ocfs2_file_splice_read like we do in ocfs2_file_aio_read so
that we can update atime in splice read if necessary.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/file.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 07267e0da909..62442e413a00 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -2026,7 +2026,7 @@ 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 +2037,12 @@ 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);
 

From 94e41ecfe0f202df948fdbb19a53308a58cf2184 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 14:45:54 -0700
Subject: [PATCH 03/10] ocfs2: Pin journal head before accessing
 jh->b_committed_data

This patch adds jbd_lock_bh_state() and jbd_unlock_bh_state() around accessses
to jh->b_committed_data.

Fixes oss bugzilla#1131
http://oss.oracle.com/bugzilla/show_bug.cgi?id=1131

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/suballoc.c | 28 ++++++++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)

diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index 8439f6b324b9..73a16d4666dc 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);

From c3d38840abaa45c1c5a5fabbb8ffc9a0d1a764d1 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 14:45:55 -0700
Subject: [PATCH 04/10] ocfs2: Fix ocfs2_osb_dump()

Skip printing information that is not valid for local mounts.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/super.c | 32 ++++++++++++++++++--------------
 1 file changed, 18 insertions(+), 14 deletions(-)

diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index d33767f17ba3..d64739b593e9 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -234,20 +234,24 @@ 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 +271,15 @@ 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  "
@@ -302,7 +307,6 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 
 	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",

From 692684e19e317a374c18e70a44d6413e51f71c11 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 16:53:17 -0700
Subject: [PATCH 05/10] ocfs2: Stop orphan scan as early as possible during
 umount

Currently if the orphan scan fires a tick before the user issues the umount,
the umount will wait for the queued orphan scan tasks to complete.

This patch makes the umount stop the orphan scan as early as possible so as
to reduce the probability of the queued tasks slowing down the umount.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/journal.c | 14 ++++++++++++--
 fs/ocfs2/ocfs2.h   |  6 ++++++
 fs/ocfs2/super.c   |  5 +++--
 3 files changed, 21 insertions(+), 4 deletions(-)

diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 4a3b9e6b31ad..70215a21fb20 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1880,6 +1880,9 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 
 	os = &osb->osb_orphan_scan;
 
+	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
+		goto out;
+
 	status = ocfs2_orphan_scan_lock(osb, &seqno, DLM_LOCK_EX);
 	if (status < 0) {
 		if (status != -EAGAIN)
@@ -1887,6 +1890,10 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 		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;
@@ -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);
 }
 
@@ -1930,6 +1938,7 @@ void ocfs2_orphan_scan_stop(struct ocfs2_super *osb)
 	struct ocfs2_orphan_scan *os;
 
 	os = &osb->osb_orphan_scan;
+	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);
@@ -1940,6 +1949,7 @@ int ocfs2_orphan_scan_init(struct ocfs2_super *osb)
 	struct ocfs2_orphan_scan *os;
 
 	os = &osb->osb_orphan_scan;
+	atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
 	os->os_osb = osb;
 	os->os_count = 0;
 	os->os_scantime = CURRENT_TIME;
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index 18c1d9ec1c93..60e89503ce5a 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -154,6 +154,11 @@ struct ocfs2_lock_res {
 #endif
 };
 
+enum ocfs2_orphan_scan_state {
+	ORPHAN_SCAN_ACTIVE,
+	ORPHAN_SCAN_INACTIVE
+};
+
 struct ocfs2_orphan_scan {
 	struct mutex 		os_lock;
 	struct ocfs2_super 	*os_osb;
@@ -162,6 +167,7 @@ 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 {
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index d64739b593e9..3e8a68b103ab 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -1814,14 +1814,15 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
 
 	debugfs_remove(osb->osb_ctxt);
 
+	/* 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);
 

From 3211949f8998dde71d9fe2e063de045ece5e0473 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 16:53:18 -0700
Subject: [PATCH 06/10] ocfs2: Do not initialize lvb in
 ocfs2_orphan_scan_lock_res_init()

We don't access the LVB in our ocfs2_*_lock_res_init() functions.

Since the LVB can become invalid during some cluster recovery
operations, the dlmglue must be able to handle an uninitialized
LVB.

For the orphan scan lock, we initialized an uninitialzed LVB with our
scan sequence number plus one.  This starts a normal orphan scan
cycle.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/dlmglue.c | 7 +++----
 fs/ocfs2/journal.c | 1 +
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 83d2ddb27186..d3d1f9372f7e 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -644,14 +644,10 @@ 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,
@@ -2386,6 +2382,9 @@ int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
 	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;
 }
 
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 70215a21fb20..0b2c27a9485e 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1952,6 +1952,7 @@ int ocfs2_orphan_scan_init(struct ocfs2_super *osb)
 	atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
 	os->os_osb = osb;
 	os->os_count = 0;
+	os->os_seqno = 0;
 	os->os_scantime = CURRENT_TIME;
 	mutex_init(&os->os_lock);
 

From df152c241df9e9d2b9a65d37bd02961abe7f591a Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Mon, 22 Jun 2009 11:40:07 -0700
Subject: [PATCH 07/10] ocfs2: Disable orphan scanning for local and hard-ro
 mounts

Local and Hard-RO mounts do not need orphan scanning.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/dlmglue.c | 26 ++++++++++++++++----------
 fs/ocfs2/dlmglue.h |  4 ++--
 fs/ocfs2/journal.c | 30 +++++++++++++++++-------------
 fs/ocfs2/journal.h |  2 +-
 fs/ocfs2/super.c   | 32 +++++++++++++++-----------------
 5 files changed, 51 insertions(+), 43 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index d3d1f9372f7e..1841bbb49cb6 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -2366,15 +2366,20 @@ 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;
 
@@ -2388,17 +2393,18 @@ int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
 	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,
diff --git a/fs/ocfs2/dlmglue.h b/fs/ocfs2/dlmglue.h
index 31b90d7b8f51..30f683107f1e 100644
--- a/fs/ocfs2/dlmglue.h
+++ b/fs/ocfs2/dlmglue.h
@@ -121,8 +121,8 @@ 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/journal.c b/fs/ocfs2/journal.c
index 0b2c27a9485e..f033760ecbea 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1883,7 +1883,7 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
 		goto out;
 
-	status = ocfs2_orphan_scan_lock(osb, &seqno, DLM_LOCK_EX);
+	status = ocfs2_orphan_scan_lock(osb, &seqno);
 	if (status < 0) {
 		if (status != -EAGAIN)
 			mlog_errno(status);
@@ -1910,7 +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;
 }
@@ -1938,29 +1938,33 @@ void ocfs2_orphan_scan_stop(struct ocfs2_super *osb)
 	struct ocfs2_orphan_scan *os;
 
 	os = &osb->osb_orphan_scan;
-	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);
+	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;
-	atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
 	os->os_osb = osb;
 	os->os_count = 0;
 	os->os_seqno = 0;
 	os->os_scantime = CURRENT_TIME;
 	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;
+	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 61045eeb3f6e..5432c7f79cc6 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -144,7 +144,7 @@ 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_stop(struct ocfs2_super *osb);
 void ocfs2_orphan_scan_exit(struct ocfs2_super *osb);
 
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 3e8a68b103ab..0746e1abdfc7 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",
@@ -305,6 +304,16 @@ 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) {
@@ -315,13 +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;
 }
 
@@ -1179,6 +1181,9 @@ 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_init(osb);
+
 	mlog_exit(status);
 	return status;
 
@@ -1983,13 +1988,6 @@ 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);
 

From 9a7aa12f3911853a3574d47d567b81a2a5df7208 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 4 Jun 2009 15:26:49 +0200
Subject: [PATCH 08/10] vfs: Set special lockdep map for dirs only if not set
 by fs

Some filesystems need to set lockdep map for i_mutex differently for
different directories. For example OCFS2 has system directories (for
orphan inode tracking and for gathering all system files like journal
or quota files into a single place) which have different locking
locking rules than standard directories. For a filesystem setting
lockdep map is naturaly done when the inode is read but we have to
modify unlock_new_inode() not to overwrite the lockdep map the filesystem
has set.

Acked-by: peterz@infradead.org
CC: mingo@redhat.com
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/inode.c              | 17 +++++++++++------
 include/linux/lockdep.h | 15 +++++++++++++++
 2 files changed, 26 insertions(+), 6 deletions(-)

diff --git a/fs/inode.c b/fs/inode.c
index f643be565df8..04c785bb63c3 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -665,12 +665,17 @@ void unlock_new_inode(struct inode *inode)
 	if (inode->i_mode & S_IFDIR) {
 		struct file_system_type *type = inode->i_sb->s_type;
 
-		/*
-		 * ensure nobody is actually holding i_mutex
-		 */
-		mutex_destroy(&inode->i_mutex);
-		mutex_init(&inode->i_mutex);
-		lockdep_set_class(&inode->i_mutex, &type->i_mutex_dir_key);
+		/* Set new key only if filesystem hasn't already changed it */
+		if (!lockdep_match_class(&inode->i_mutex,
+		    &type->i_mutex_key)) {
+			/*
+			 * ensure nobody is actually holding i_mutex
+			 */
+			mutex_destroy(&inode->i_mutex);
+			mutex_init(&inode->i_mutex);
+			lockdep_set_class(&inode->i_mutex,
+					  &type->i_mutex_dir_key);
+		}
 	}
 #endif
 	/*
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index da5a5a1f4cd2..b25d1b53df0d 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -258,6 +258,16 @@ extern void lockdep_init_map(struct lockdep_map *lock, const char *name,
 #define lockdep_set_subclass(lock, sub)	\
 		lockdep_init_map(&(lock)->dep_map, #lock, \
 				 (lock)->dep_map.key, sub)
+/*
+ * Compare locking classes
+ */
+#define lockdep_match_class(lock, key) lockdep_match_key(&(lock)->dep_map, key)
+
+static inline int lockdep_match_key(struct lockdep_map *lock,
+				    struct lock_class_key *key)
+{
+	return lock->key == key;
+}
 
 /*
  * Acquire a lock.
@@ -326,6 +336,11 @@ static inline void lockdep_on(void)
 #define lockdep_set_class_and_subclass(lock, key, sub) \
 		do { (void)(key); } while (0)
 #define lockdep_set_subclass(lock, sub)		do { } while (0)
+/*
+ * We don't define lockdep_match_class() and lockdep_match_key() for !LOCKDEP
+ * case since the result is not well defined and the caller should rather
+ * #ifdef the call himself.
+ */
 
 # define INIT_LOCKDEP
 # define lockdep_reset()		do { debug_locks = 1; } while (0)

From cb25797d451dc774d9dbc402a65f16a0e32199fe Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 4 Jun 2009 15:26:50 +0200
Subject: [PATCH 09/10] ocfs2: Add lockdep annotations

Add lockdep support to OCFS2. The support also covers all of the cluster
locks except for open locks, journal locks, and local quotafile locks. These
are special because they are acquired for a node, not for a particular process
and lockdep cannot deal with such type of locking.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/dlmglue.c | 81 +++++++++++++++++++++++++++++++++++++---------
 fs/ocfs2/dlmglue.h | 20 ++++++++++--
 fs/ocfs2/inode.c   | 11 +++++++
 fs/ocfs2/namei.c   | 15 +++++----
 fs/ocfs2/ocfs2.h   |  4 +++
 fs/ocfs2/sysfile.c | 17 ++++++++++
 6 files changed, 123 insertions(+), 25 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 1841bbb49cb6..110bb57c46ab 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)
@@ -1252,11 +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);
@@ -1399,13 +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;
 
@@ -1414,6 +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();
 }
 
@@ -2159,10 +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;
@@ -2200,7 +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);
diff --git a/fs/ocfs2/dlmglue.h b/fs/ocfs2/dlmglue.h
index 30f683107f1e..7553836931de 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);
@@ -104,17 +112,23 @@ 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,
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 10e1fa87396a..4dc8890ba316 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/namei.c b/fs/ocfs2/namei.c
index 33464c6b60a2..8601f934010b 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 60e89503ce5a..c9345ebb8493 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 +153,9 @@ 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 {
diff --git a/fs/ocfs2/sysfile.c b/fs/ocfs2/sysfile.c
index ab713ebdd546..6f53f5e7256a 100644
--- a/fs/ocfs2/sysfile.c
+++ b/fs/ocfs2/sysfile.c
@@ -50,6 +50,8 @@ static inline int is_in_system_inode_array(struct ocfs2_super *osb,
 					   int type,
 					   u32 slot);
 
+static struct lock_class_key ocfs2_sysfile_cluster_lock_key[NUM_SYSTEM_INODES];
+
 static inline int is_global_system_inode(int type)
 {
 	return type >= OCFS2_FIRST_ONLINE_SYSTEM_INODE &&
@@ -118,6 +120,21 @@ 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;

From d246ab307d1d003c80fe279897dea22bf52b6e41 Mon Sep 17 00:00:00 2001
From: Tao Ma <tao.ma@oracle.com>
Date: Thu, 18 Jun 2009 13:12:06 +0800
Subject: [PATCH 10/10] ocfs2/trivial: Wrap ocfs2_sysfile_cluster_lock_key
 within define.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Actually ocfs2_sysfile_cluster_lock_key is only used if we enable
CONFIG_DEBUG_LOCK_ALLOC. Wrap it so that we can avoid a building
warning.
fs/ocfs2/sysfile.c:53: warning: ‘ocfs2_sysfile_cluster_lock_key’
defined but not used

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/sysfile.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/ocfs2/sysfile.c b/fs/ocfs2/sysfile.c
index 6f53f5e7256a..40e53702948c 100644
--- a/fs/ocfs2/sysfile.c
+++ b/fs/ocfs2/sysfile.c
@@ -50,7 +50,9 @@ 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)
 {