ocfs2: Add new cluster lock type

Replace the dentry vote mechanism with a cluster lock which covers a set
of dentries. This allows us to force d_delete() only on nodes which actually
care about an unlink.

Every node that does a ->lookup() gets a read only lock on the dentry, until
an unlink during which the unlinking node, will request an exclusive lock,
forcing the other nodes who care about that dentry to d_delete() it. The
effect is that we retain a very lightweight ->d_revalidate(), and at the
same time get to make large improvements to the average case performance of
the ocfs2 unlink and rename operations.

This patch adds the cluster lock type which OCFS2 can attach to
dentries.  A small number of fs/ocfs2/dcache.c functions are stubbed
out so that this change can compile.
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>

fs/ocfs2/dcache.c

@@ -87,6 +87,8 @@ static int ocfs2_dentry_revalidate(struct dentry *dentry,
 	return ret;
 }
 
+DEFINE_SPINLOCK(dentry_attach_lock);
+
 struct dentry_operations ocfs2_dentry_ops = {
 	.d_revalidate		= ocfs2_dentry_revalidate,
 };

fs/ocfs2/dcache.h

@@ -28,4 +28,31 @@
 
 extern struct dentry_operations ocfs2_dentry_ops;
 
+struct ocfs2_dentry_lock {
+	unsigned int		dl_count;
+	u64			dl_parent_blkno;
+
+	/*
+	 * The ocfs2_dentry_lock keeps an inode reference until
+	 * dl_lockres has been destroyed. This is usually done in
+	 * ->d_iput() anyway, so there should be minimal impact.
+	 */
+	struct inode		*dl_inode;
+	struct ocfs2_lock_res	dl_lockres;
+};
+
+static inline void ocfs2_dentry_lock_put(struct ocfs2_super *osb,
+					 struct ocfs2_dentry_lock *dl)
+{
+}
+
+static inline struct dentry *ocfs2_find_local_alias(struct inode *inode,
+						    u64 parent_blkno,
+						    int skip_unhashed)
+{
+	return NULL;
+}
+
+extern spinlock_t dentry_attach_lock;
+
 #endif /* OCFS2_DCACHE_H */

fs/ocfs2/dlmglue.h

@@ -27,6 +27,8 @@
 #ifndef DLMGLUE_H
 #define DLMGLUE_H
 
+#include "dcache.h"
+
 #define OCFS2_LVB_VERSION 3
 
 struct ocfs2_meta_lvb {
@@ -58,8 +60,12 @@ void ocfs2_lock_res_init_once(struct ocfs2_lock_res *res);
 void ocfs2_inode_lock_res_init(struct ocfs2_lock_res *res,
 			       enum ocfs2_lock_type type,
 			       struct inode *inode);
+void ocfs2_dentry_lock_res_init(struct ocfs2_dentry_lock *dl,
+				u64 parent, struct inode *inode);
 void ocfs2_lock_res_free(struct ocfs2_lock_res *res);
 int ocfs2_create_new_inode_locks(struct inode *inode);
+int ocfs2_create_new_lock(struct ocfs2_super *osb,
+			  struct ocfs2_lock_res *lockres, int ex);
 int ocfs2_drop_inode_locks(struct inode *inode);
 int ocfs2_data_lock_full(struct inode *inode,
 			 int write,
@@ -93,7 +99,12 @@ void ocfs2_super_unlock(struct ocfs2_super *osb,
 			int ex);
 int ocfs2_rename_lock(struct ocfs2_super *osb);
 void ocfs2_rename_unlock(struct ocfs2_super *osb);
+int ocfs2_dentry_lock(struct dentry *dentry, int ex);
+void ocfs2_dentry_unlock(struct dentry *dentry, int ex);
+
 void ocfs2_mark_lockres_freeing(struct ocfs2_lock_res *lockres);
+void ocfs2_simple_drop_lockres(struct ocfs2_super *osb,
+			       struct ocfs2_lock_res *lockres);
 
 /* for the vote thread */
 void ocfs2_process_blocked_lock(struct ocfs2_super *osb,

fs/ocfs2/ocfs2_lockid.h

@@ -35,12 +35,15 @@
 #define OCFS2_LOCK_ID_MAX_LEN  32
 #define OCFS2_LOCK_ID_PAD "000000"
 
+#define OCFS2_DENTRY_LOCK_INO_START 18
+
 enum ocfs2_lock_type {
 	OCFS2_LOCK_TYPE_META = 0,
 	OCFS2_LOCK_TYPE_DATA,
 	OCFS2_LOCK_TYPE_SUPER,
 	OCFS2_LOCK_TYPE_RENAME,
 	OCFS2_LOCK_TYPE_RW,
+	OCFS2_LOCK_TYPE_DENTRY,
 	OCFS2_NUM_LOCK_TYPES
 };
 
@@ -63,6 +66,9 @@ static inline char ocfs2_lock_type_char(enum ocfs2_lock_type type)
 		case OCFS2_LOCK_TYPE_RW:
 			c = 'W';
 			break;
+		case OCFS2_LOCK_TYPE_DENTRY:
+			c = 'N';
+			break;
 		default:
 			c = '\0';
 	}
@@ -70,4 +76,23 @@ static inline char ocfs2_lock_type_char(enum ocfs2_lock_type type)
 	return c;
 }
 
+static char *ocfs2_lock_type_strings[] = {
+	[OCFS2_LOCK_TYPE_META] = "Meta",
+	[OCFS2_LOCK_TYPE_DATA] = "Data",
+	[OCFS2_LOCK_TYPE_SUPER] = "Super",
+	[OCFS2_LOCK_TYPE_RENAME] = "Rename",
+	/* Need to differntiate from [R]ename.. serializing writes is the
+	 * important job it does, anyway. */
+	[OCFS2_LOCK_TYPE_RW] = "Write/Read",
+	[OCFS2_LOCK_TYPE_DENTRY] = "Dentry",
+};
+
+static inline const char *ocfs2_lock_type_string(enum ocfs2_lock_type type)
+{
+#ifdef __KERNEL__
+	mlog_bug_on_msg(type >= OCFS2_NUM_LOCK_TYPES, "%d\n", type);
+#endif
+	return ocfs2_lock_type_strings[type];
+}
+
 #endif  /* OCFS2_LOCKID_H */