X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=fs%2Fdcache.c;h=cbf76e692013281de5852d569896ac1625bd96fb;hb=d4c1b5f91280f0f15d82ddf2156117c300c490b9;hp=a56398970d1e7868eba5dc070faa7f736986a7e3;hpb=ec9397bab20a628530ce3051167d3d0fcc2c1af7;p=linux-2.6.git

diff --git a/fs/dcache.c b/fs/dcache.c
index a56398970..cbf76e692 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -14,10 +14,11 @@
  * the dcache entry is deleted or garbage collected.
  */
 
-#include <linux/config.h>
+#include <linux/syscalls.h>
 #include <linux/string.h>
 #include <linux/mm.h>
 #include <linux/fs.h>
+#include <linux/fsnotify.h>
 #include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/smp_lock.h>
@@ -32,16 +33,16 @@
 #include <linux/swap.h>
 #include <linux/bootmem.h>
 
-/* #define DCACHE_DEBUG 1 */
 
-int sysctl_vfs_cache_pressure = 100;
+int sysctl_vfs_cache_pressure __read_mostly = 100;
+EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure);
 
-spinlock_t dcache_lock __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
-seqlock_t rename_lock __cacheline_aligned_in_smp = SEQLOCK_UNLOCKED;
+ __cacheline_aligned_in_smp DEFINE_SPINLOCK(dcache_lock);
+static __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock);
 
 EXPORT_SYMBOL(dcache_lock);
 
-static kmem_cache_t *dentry_cache; 
+static kmem_cache_t *dentry_cache __read_mostly;
 
 #define DNAME_INLINE_LEN (sizeof(struct dentry)-offsetof(struct dentry,d_iname))
 
@@ -56,9 +57,9 @@ static kmem_cache_t *dentry_cache;
 #define D_HASHBITS     d_hash_shift
 #define D_HASHMASK     d_hash_mask
 
-static unsigned int d_hash_mask;
-static unsigned int d_hash_shift;
-static struct hlist_head *dentry_hashtable;
+static unsigned int d_hash_mask __read_mostly;
+static unsigned int d_hash_shift __read_mostly;
+static struct hlist_head *dentry_hashtable __read_mostly;
 static LIST_HEAD(dentry_unused);
 
 /* Statistics gathering. */
@@ -68,7 +69,7 @@ struct dentry_stat_t dentry_stat = {
 
 static void d_callback(struct rcu_head *head)
 {
-	struct dentry * dentry = container_of(head, struct dentry, d_rcu);
+	struct dentry * dentry = container_of(head, struct dentry, d_u.d_rcu);
 
 	if (dname_external(dentry))
 		kfree(dentry->d_name.name);
@@ -83,11 +84,11 @@ static void d_free(struct dentry *dentry)
 {
 	if (dentry->d_op && dentry->d_op->d_release)
 		dentry->d_op->d_release(dentry);
-	if (dentry->d_extra_attributes) {
-		kfree(dentry->d_extra_attributes);
-		dentry->d_extra_attributes = NULL;
-	}
- 	call_rcu(&dentry->d_rcu, d_callback);
+ 	if (dentry->d_extra_attributes) {
+ 		kfree(dentry->d_extra_attributes);
+ 		dentry->d_extra_attributes = NULL;
+ 	}
+ 	call_rcu(&dentry->d_u.d_rcu, d_callback);
 }
 
 /*
@@ -95,7 +96,7 @@ static void d_free(struct dentry *dentry)
  * d_iput() operation if defined.
  * Called with dcache_lock and per dentry lock held, drops both.
  */
-static inline void dentry_iput(struct dentry * dentry)
+static void dentry_iput(struct dentry * dentry)
 {
 	struct inode *inode = dentry->d_inode;
 	if (inode) {
@@ -103,6 +104,8 @@ static inline void dentry_iput(struct dentry * dentry)
 		list_del_init(&dentry->d_alias);
 		spin_unlock(&dentry->d_lock);
 		spin_unlock(&dcache_lock);
+		if (!inode->i_nlink)
+			fsnotify_inoderemove(inode);
 		if (dentry->d_op && dentry->d_op->d_iput)
 			dentry->d_op->d_iput(dentry, inode);
 		else
@@ -159,7 +162,7 @@ repeat:
 		spin_unlock(&dcache_lock);
 		return;
 	}
-			
+
 	/*
 	 * AV: ->d_delete() is _NOT_ allowed to block now.
 	 */
@@ -192,7 +195,7 @@ kill_it: {
   			list_del(&dentry->d_lru);
   			dentry_stat.nr_unused--;
   		}
-  		list_del(&dentry->d_child);
+  		list_del(&dentry->d_u.d_child);
 		dentry_stat.nr_dentry--;	/* For d_free, below */
 		/*drops the locks, at that point nobody can reach this dentry */
 		dentry_iput(dentry);
@@ -324,10 +327,13 @@ static struct dentry * __d_find_alias(struct inode *inode, int want_discon)
 
 struct dentry * d_find_alias(struct inode *inode)
 {
-	struct dentry *de;
-	spin_lock(&dcache_lock);
-	de = __d_find_alias(inode, 0);
-	spin_unlock(&dcache_lock);
+	struct dentry *de = NULL;
+
+	if (!list_empty(&inode->i_dentry)) {
+		spin_lock(&dcache_lock);
+		de = __d_find_alias(inode, 0);
+		spin_unlock(&dcache_lock);
+	}
 	return de;
 }
 
@@ -337,35 +343,37 @@ struct dentry * d_find_alias(struct inode *inode)
  */
 void d_prune_aliases(struct inode *inode)
 {
-	struct list_head *tmp, *head = &inode->i_dentry;
+	struct dentry *dentry;
 restart:
 	spin_lock(&dcache_lock);
-	tmp = head;
-	while ((tmp = tmp->next) != head) {
-		struct dentry *dentry = list_entry(tmp, struct dentry, d_alias);
+	list_for_each_entry(dentry, &inode->i_dentry, d_alias) {
+		spin_lock(&dentry->d_lock);
 		if (!atomic_read(&dentry->d_count)) {
 			__dget_locked(dentry);
 			__d_drop(dentry);
+			spin_unlock(&dentry->d_lock);
 			spin_unlock(&dcache_lock);
 			dput(dentry);
 			goto restart;
 		}
+		spin_unlock(&dentry->d_lock);
 	}
 	spin_unlock(&dcache_lock);
 }
 
 /*
- * Throw away a dentry - free the inode, dput the parent.
- * This requires that the LRU list has already been
- * removed.
+ * Throw away a dentry - free the inode, dput the parent.  This requires that
+ * the LRU list has already been removed.
+ *
  * Called with dcache_lock, drops it and then regains.
+ * Called with dentry->d_lock held, drops it.
  */
-static inline void prune_one_dentry(struct dentry * dentry)
+static void prune_one_dentry(struct dentry * dentry)
 {
 	struct dentry * parent;
 
 	__d_drop(dentry);
-	list_del(&dentry->d_child);
+	list_del(&dentry->d_u.d_child);
 	dentry_stat.nr_dentry--;	/* For d_free, below */
 	dentry_iput(dentry);
 	parent = dentry->d_parent;
@@ -378,6 +386,8 @@ static inline void prune_one_dentry(struct dentry * dentry)
 /**
  * prune_dcache - shrink the dcache
  * @count: number of entries to try and free
+ * @sb: if given, ignore dentries for other superblocks
+ *         which are being unmounted.
  *
  * Shrink the dcache. This is done when we need
  * more memory, or simply when we need to unmount
@@ -388,16 +398,29 @@ static inline void prune_one_dentry(struct dentry * dentry)
  * all the dentries are in use.
  */
  
-static void prune_dcache(int count)
+static void prune_dcache(int count, struct super_block *sb)
 {
 	spin_lock(&dcache_lock);
 	for (; count ; count--) {
 		struct dentry *dentry;
 		struct list_head *tmp;
+		struct rw_semaphore *s_umount;
 
 		cond_resched_lock(&dcache_lock);
 
 		tmp = dentry_unused.prev;
+		if (sb) {
+			/* Try to find a dentry for this sb, but don't try
+			 * too hard, if they aren't near the tail they will
+			 * be moved down again soon
+			 */
+			int skip = count;
+			while (skip && tmp != &dentry_unused &&
+			    list_entry(tmp, struct dentry, d_lru)->d_sb != sb) {
+				skip--;
+				tmp = tmp->prev;
+			}
+		}
 		if (tmp == &dentry_unused)
 			break;
 		list_del_init(tmp);
@@ -423,7 +446,45 @@ static void prune_dcache(int count)
  			spin_unlock(&dentry->d_lock);
 			continue;
 		}
-		prune_one_dentry(dentry);
+		/*
+		 * If the dentry is not DCACHED_REFERENCED, it is time
+		 * to remove it from the dcache, provided the super block is
+		 * NULL (which means we are trying to reclaim memory)
+		 * or this dentry belongs to the same super block that
+		 * we want to shrink.
+		 */
+		/*
+		 * If this dentry is for "my" filesystem, then I can prune it
+		 * without taking the s_umount lock (I already hold it).
+		 */
+		if (sb && dentry->d_sb == sb) {
+			prune_one_dentry(dentry);
+			continue;
+		}
+		/*
+		 * ...otherwise we need to be sure this filesystem isn't being
+		 * unmounted, otherwise we could race with
+		 * generic_shutdown_super(), and end up holding a reference to
+		 * an inode while the filesystem is unmounted.
+		 * So we try to get s_umount, and make sure s_root isn't NULL.
+		 * (Take a local copy of s_umount to avoid a use-after-free of
+		 * `dentry').
+		 */
+		s_umount = &dentry->d_sb->s_umount;
+		if (down_read_trylock(s_umount)) {
+			if (dentry->d_sb->s_root != NULL) {
+				prune_one_dentry(dentry);
+				up_read(s_umount);
+				continue;
+			}
+			up_read(s_umount);
+		}
+		spin_unlock(&dentry->d_lock);
+		/* Cannot remove the first dentry, and it isn't appropriate
+		 * to move it to the head of the list, so give up, and try
+		 * later
+		 */
+		break;
 	}
 	spin_unlock(&dcache_lock);
 }
@@ -460,25 +521,18 @@ void shrink_dcache_sb(struct super_block * sb)
 	 * superblock to the most recent end of the unused list.
 	 */
 	spin_lock(&dcache_lock);
-	next = dentry_unused.next;
-	while (next != &dentry_unused) {
-		tmp = next;
-		next = tmp->next;
+	list_for_each_safe(tmp, next, &dentry_unused) {
 		dentry = list_entry(tmp, struct dentry, d_lru);
 		if (dentry->d_sb != sb)
 			continue;
-		list_del(tmp);
-		list_add(tmp, &dentry_unused);
+		list_move(tmp, &dentry_unused);
 	}
 
 	/*
 	 * Pass two ... free the dentries for this superblock.
 	 */
 repeat:
-	next = dentry_unused.next;
-	while (next != &dentry_unused) {
-		tmp = next;
-		next = tmp->next;
+	list_for_each_safe(tmp, next, &dentry_unused) {
 		dentry = list_entry(tmp, struct dentry, d_lru);
 		if (dentry->d_sb != sb)
 			continue;
@@ -490,11 +544,142 @@ repeat:
 			continue;
 		}
 		prune_one_dentry(dentry);
+		cond_resched_lock(&dcache_lock);
 		goto repeat;
 	}
 	spin_unlock(&dcache_lock);
 }
 
+/*
+ * destroy a single subtree of dentries for unmount
+ * - see the comments on shrink_dcache_for_umount() for a description of the
+ *   locking
+ */
+static void shrink_dcache_for_umount_subtree(struct dentry *dentry)
+{
+	struct dentry *parent;
+
+	BUG_ON(!IS_ROOT(dentry));
+
+	/* detach this root from the system */
+	spin_lock(&dcache_lock);
+	if (!list_empty(&dentry->d_lru)) {
+		dentry_stat.nr_unused--;
+		list_del_init(&dentry->d_lru);
+	}
+	__d_drop(dentry);
+	spin_unlock(&dcache_lock);
+
+	for (;;) {
+		/* descend to the first leaf in the current subtree */
+		while (!list_empty(&dentry->d_subdirs)) {
+			struct dentry *loop;
+
+			/* this is a branch with children - detach all of them
+			 * from the system in one go */
+			spin_lock(&dcache_lock);
+			list_for_each_entry(loop, &dentry->d_subdirs,
+					    d_u.d_child) {
+				if (!list_empty(&loop->d_lru)) {
+					dentry_stat.nr_unused--;
+					list_del_init(&loop->d_lru);
+				}
+
+				__d_drop(loop);
+				cond_resched_lock(&dcache_lock);
+			}
+			spin_unlock(&dcache_lock);
+
+			/* move to the first child */
+			dentry = list_entry(dentry->d_subdirs.next,
+					    struct dentry, d_u.d_child);
+		}
+
+		/* consume the dentries from this leaf up through its parents
+		 * until we find one with children or run out altogether */
+		do {
+			struct inode *inode;
+
+			if (atomic_read(&dentry->d_count) != 0) {
+				printk(KERN_ERR
+				       "BUG: Dentry %p{i=%lx,n=%s}"
+				       " still in use (%d)"
+				       " [unmount of %s %s]\n",
+				       dentry,
+				       dentry->d_inode ?
+				       dentry->d_inode->i_ino : 0UL,
+				       dentry->d_name.name,
+				       atomic_read(&dentry->d_count),
+				       dentry->d_sb->s_type->name,
+				       dentry->d_sb->s_id);
+				BUG();
+			}
+
+			parent = dentry->d_parent;
+			if (parent == dentry)
+				parent = NULL;
+			else
+				atomic_dec(&parent->d_count);
+
+			list_del(&dentry->d_u.d_child);
+			dentry_stat.nr_dentry--;	/* For d_free, below */
+
+			inode = dentry->d_inode;
+			if (inode) {
+				dentry->d_inode = NULL;
+				list_del_init(&dentry->d_alias);
+				if (dentry->d_op && dentry->d_op->d_iput)
+					dentry->d_op->d_iput(dentry, inode);
+				else
+					iput(inode);
+			}
+
+			d_free(dentry);
+
+			/* finished when we fall off the top of the tree,
+			 * otherwise we ascend to the parent and move to the
+			 * next sibling if there is one */
+			if (!parent)
+				return;
+
+			dentry = parent;
+
+		} while (list_empty(&dentry->d_subdirs));
+
+		dentry = list_entry(dentry->d_subdirs.next,
+				    struct dentry, d_u.d_child);
+	}
+}
+
+/*
+ * destroy the dentries attached to a superblock on unmounting
+ * - we don't need to use dentry->d_lock, and only need dcache_lock when
+ *   removing the dentry from the system lists and hashes because:
+ *   - the superblock is detached from all mountings and open files, so the
+ *     dentry trees will not be rearranged by the VFS
+ *   - s_umount is write-locked, so the memory pressure shrinker will ignore
+ *     any dentries belonging to this superblock that it comes across
+ *   - the filesystem itself is no longer permitted to rearrange the dentries
+ *     in this superblock
+ */
+void shrink_dcache_for_umount(struct super_block *sb)
+{
+	struct dentry *dentry;
+
+	if (down_read_trylock(&sb->s_umount))
+		BUG();
+
+	dentry = sb->s_root;
+	sb->s_root = NULL;
+	atomic_dec(&dentry->d_count);
+	shrink_dcache_for_umount_subtree(dentry);
+
+	while (!hlist_empty(&sb->s_anon)) {
+		dentry = hlist_entry(sb->s_anon.first, struct dentry, d_hash);
+		shrink_dcache_for_umount_subtree(dentry);
+	}
+}
+
 /*
  * Search for at least 1 mount point in the dentry's subdirs.
  * We descend to the next level whenever the d_subdirs
@@ -522,7 +707,7 @@ repeat:
 resume:
 	while (next != &this_parent->d_subdirs) {
 		struct list_head *tmp = next;
-		struct dentry *dentry = list_entry(tmp, struct dentry, d_child);
+		struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child);
 		next = tmp->next;
 		/* Have we found a mount point ? */
 		if (d_mountpoint(dentry))
@@ -536,7 +721,7 @@ resume:
 	 * All done at this level ... ascend and resume the search.
 	 */
 	if (this_parent != parent) {
-		next = this_parent->d_child.next; 
+		next = this_parent->d_u.d_child.next;
 		this_parent = this_parent->d_parent;
 		goto resume;
 	}
@@ -553,6 +738,13 @@ positive:
  * list for prune_dcache(). We descend to the next level
  * whenever the d_subdirs list is non-empty and continue
  * searching.
+ *
+ * It returns zero iff there are no unused children,
+ * otherwise  it returns the number of children moved to
+ * the end of the unused list. This may not be the total
+ * number of unused children, because select_parent can
+ * drop the lock and return early due to latency
+ * constraints.
  */
 static int select_parent(struct dentry * parent)
 {
@@ -566,7 +758,7 @@ repeat:
 resume:
 	while (next != &this_parent->d_subdirs) {
 		struct list_head *tmp = next;
-		struct dentry *dentry = list_entry(tmp, struct dentry, d_child);
+		struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child);
 		next = tmp->next;
 
 		if (!list_empty(&dentry->d_lru)) {
@@ -578,19 +770,24 @@ resume:
 		 * of the unused list for prune_dcache
 		 */
 		if (!atomic_read(&dentry->d_count)) {
-			list_add(&dentry->d_lru, dentry_unused.prev);
+			list_add_tail(&dentry->d_lru, &dentry_unused);
 			dentry_stat.nr_unused++;
 			found++;
 		}
+
+		/*
+		 * We can return to the caller if we have found some (this
+		 * ensures forward progress). We'll be coming back to find
+		 * the rest.
+		 */
+		if (found && need_resched())
+			goto out;
+
 		/*
 		 * Descend a level if the d_subdirs list is non-empty.
 		 */
 		if (!list_empty(&dentry->d_subdirs)) {
 			this_parent = dentry;
-#ifdef DCACHE_DEBUG
-printk(KERN_DEBUG "select_parent: descending to %s/%s, found=%d\n",
-dentry->d_parent->d_name.name, dentry->d_name.name, found);
-#endif
 			goto repeat;
 		}
 	}
@@ -598,14 +795,11 @@ dentry->d_parent->d_name.name, dentry->d_name.name, found);
 	 * All done at this level ... ascend and resume the search.
 	 */
 	if (this_parent != parent) {
-		next = this_parent->d_child.next; 
+		next = this_parent->d_u.d_child.next;
 		this_parent = this_parent->d_parent;
-#ifdef DCACHE_DEBUG
-printk(KERN_DEBUG "select_parent: ascending to %s/%s, found=%d\n",
-this_parent->d_parent->d_name.name, this_parent->d_name.name, found);
-#endif
 		goto resume;
 	}
+out:
 	spin_unlock(&dcache_lock);
 	return found;
 }
@@ -622,46 +816,7 @@ void shrink_dcache_parent(struct dentry * parent)
 	int found;
 
 	while ((found = select_parent(parent)) != 0)
-		prune_dcache(found);
-}
-
-/**
- * shrink_dcache_anon - further prune the cache
- * @head: head of d_hash list of dentries to prune
- *
- * Prune the dentries that are anonymous
- *
- * parsing d_hash list does not hlist_for_each_rcu() as it
- * done under dcache_lock.
- *
- */
-void shrink_dcache_anon(struct hlist_head *head)
-{
-	struct hlist_node *lp;
-	int found;
-	do {
-		found = 0;
-		spin_lock(&dcache_lock);
-		hlist_for_each(lp, head) {
-			struct dentry *this = hlist_entry(lp, struct dentry, d_hash);
-			if (!list_empty(&this->d_lru)) {
-				dentry_stat.nr_unused--;
-				list_del_init(&this->d_lru);
-			}
-
-			/* 
-			 * move only zero ref count dentries to the end 
-			 * of the unused list for prune_dcache
-			 */
-			if (!atomic_read(&this->d_count)) {
-				list_add_tail(&this->d_lru, &dentry_unused);
-				dentry_stat.nr_unused++;
-				found++;
-			}
-		}
-		spin_unlock(&dcache_lock);
-		prune_dcache(found);
-	} while(found);
+		prune_dcache(found, parent->d_sb);
 }
 
 /*
@@ -676,12 +831,12 @@ void shrink_dcache_anon(struct hlist_head *head)
  *
  * In this case we return -1 to tell the caller that we baled.
  */
-static int shrink_dcache_memory(int nr, unsigned int gfp_mask)
+static int shrink_dcache_memory(int nr, gfp_t gfp_mask)
 {
 	if (nr) {
 		if (!(gfp_mask & __GFP_FS))
 			return -1;
-		prune_dcache(nr);
+		prune_dcache(nr, NULL);
 	}
 	return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure;
 }
@@ -723,7 +878,7 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
 
 	atomic_set(&dentry->d_count, 1);
 	dentry->d_flags = DCACHE_UNHASHED;
-	dentry->d_lock = SPIN_LOCK_UNLOCKED;
+	spin_lock_init(&dentry->d_lock);
 	dentry->d_inode = NULL;
 	dentry->d_parent = NULL;
 	dentry->d_sb = NULL;
@@ -731,8 +886,9 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
 	dentry->d_fsdata = NULL;
 	dentry->d_extra_attributes = NULL;
 	dentry->d_mounted = 0;
+#ifdef CONFIG_PROFILING
 	dentry->d_cookie = NULL;
-	dentry->d_bucket = NULL;
+#endif
 	INIT_HLIST_NODE(&dentry->d_hash);
 	INIT_LIST_HEAD(&dentry->d_lru);
 	INIT_LIST_HEAD(&dentry->d_subdirs);
@@ -742,18 +898,28 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
 		dentry->d_parent = dget(parent);
 		dentry->d_sb = parent->d_sb;
 	} else {
-		INIT_LIST_HEAD(&dentry->d_child);
+		INIT_LIST_HEAD(&dentry->d_u.d_child);
 	}
 
 	spin_lock(&dcache_lock);
 	if (parent)
-		list_add(&dentry->d_child, &parent->d_subdirs);
+		list_add(&dentry->d_u.d_child, &parent->d_subdirs);
 	dentry_stat.nr_dentry++;
 	spin_unlock(&dcache_lock);
 
 	return dentry;
 }
 
+struct dentry *d_alloc_name(struct dentry *parent, const char *name)
+{
+	struct qstr q;
+
+	q.name = name;
+	q.len = strlen(name);
+	q.hash = full_name_hash(q.name, q.len);
+	return d_alloc(parent, &q);
+}
+
 /**
  * d_instantiate - fill in inode information for a dentry
  * @entry: dentry to complete
@@ -771,15 +937,88 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
  
 void d_instantiate(struct dentry *entry, struct inode * inode)
 {
-	if (!list_empty(&entry->d_alias)) BUG();
+	BUG_ON(!list_empty(&entry->d_alias));
 	spin_lock(&dcache_lock);
 	if (inode)
 		list_add(&entry->d_alias, &inode->i_dentry);
 	entry->d_inode = inode;
+	fsnotify_d_instantiate(entry, inode);
 	spin_unlock(&dcache_lock);
 	security_d_instantiate(entry, inode);
 }
 
+/**
+ * d_instantiate_unique - instantiate a non-aliased dentry
+ * @entry: dentry to instantiate
+ * @inode: inode to attach to this dentry
+ *
+ * Fill in inode information in the entry. On success, it returns NULL.
+ * If an unhashed alias of "entry" already exists, then we return the
+ * aliased dentry instead and drop one reference to inode.
+ *
+ * Note that in order to avoid conflicts with rename() etc, the caller
+ * had better be holding the parent directory semaphore.
+ *
+ * This also assumes that the inode count has been incremented
+ * (or otherwise set) by the caller to indicate that it is now
+ * in use by the dcache.
+ */
+static struct dentry *__d_instantiate_unique(struct dentry *entry,
+					     struct inode *inode)
+{
+	struct dentry *alias;
+	int len = entry->d_name.len;
+	const char *name = entry->d_name.name;
+	unsigned int hash = entry->d_name.hash;
+
+	if (!inode) {
+		entry->d_inode = NULL;
+		return NULL;
+	}
+
+	list_for_each_entry(alias, &inode->i_dentry, d_alias) {
+		struct qstr *qstr = &alias->d_name;
+
+		if (qstr->hash != hash)
+			continue;
+		if (alias->d_parent != entry->d_parent)
+			continue;
+		if (qstr->len != len)
+			continue;
+		if (memcmp(qstr->name, name, len))
+			continue;
+		dget_locked(alias);
+		return alias;
+	}
+
+	list_add(&entry->d_alias, &inode->i_dentry);
+	entry->d_inode = inode;
+	fsnotify_d_instantiate(entry, inode);
+	return NULL;
+}
+
+struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode)
+{
+	struct dentry *result;
+
+	BUG_ON(!list_empty(&entry->d_alias));
+
+	spin_lock(&dcache_lock);
+	result = __d_instantiate_unique(entry, inode);
+	spin_unlock(&dcache_lock);
+
+	if (!result) {
+		security_d_instantiate(entry, inode);
+		return NULL;
+	}
+
+	BUG_ON(!d_unhashed(result));
+	iput(inode);
+	return result;
+}
+
+EXPORT_SYMBOL(d_instantiate_unique);
+
 /**
  * d_alloc_root - allocate root dentry
  * @root_inode: inode to allocate the root for
@@ -861,12 +1100,6 @@ struct dentry * d_alloc_anon(struct inode *inode)
 		res->d_sb = inode->i_sb;
 		res->d_parent = res;
 		res->d_inode = inode;
-
-		/*
-		 * Set d_bucket to an "impossible" bucket address so
-		 * that d_move() doesn't get a false positive
-		 */
-		res->d_bucket = NULL;
 		res->d_flags |= DCACHE_DISCONNECTED;
 		res->d_flags &= ~DCACHE_UNHASHED;
 		list_add(&res->d_alias, &inode->i_dentry);
@@ -910,6 +1143,7 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
 		new = __d_find_alias(inode, 1);
 		if (new) {
 			BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED));
+			fsnotify_d_instantiate(new, inode);
 			spin_unlock(&dcache_lock);
 			security_d_instantiate(new, inode);
 			d_rehash(dentry);
@@ -919,6 +1153,7 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
 			/* d_instantiate takes dcache_lock, so we do it by hand */
 			list_add(&dentry->d_alias, &inode->i_dentry);
 			dentry->d_inode = inode;
+			fsnotify_d_instantiate(dentry, inode);
 			spin_unlock(&dcache_lock);
 			security_d_instantiate(dentry, inode);
 			d_rehash(dentry);
@@ -980,17 +1215,13 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name)
 	struct hlist_head *head = d_hash(parent,hash);
 	struct dentry *found = NULL;
 	struct hlist_node *node;
+	struct dentry *dentry;
 
 	rcu_read_lock();
 	
-	hlist_for_each_rcu(node, head) {
-		struct dentry *dentry; 
+	hlist_for_each_entry_rcu(dentry, node, head, d_hash) {
 		struct qstr *qstr;
 
-		dentry = hlist_entry(node, struct dentry, d_hash);
-
-		smp_rmb();
-
 		if (dentry->d_name.hash != hash)
 			continue;
 		if (dentry->d_parent != parent)
@@ -998,13 +1229,6 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name)
 
 		spin_lock(&dentry->d_lock);
 
-		/*
-		 * If lookup ends up in a different bucket due to concurrent
-		 * rename, fail it
-		 */
-		if (unlikely(dentry->d_bucket != head))
-			goto terminate;
-
 		/*
 		 * Recheck the dentry after taking the lock - d_move may have
 		 * changed things.  Don't bother checking the hash because we're
@@ -1013,7 +1237,11 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name)
 		if (dentry->d_parent != parent)
 			goto next;
 
-		qstr = rcu_dereference(&dentry->d_name);
+		/*
+		 * It is safe to compare names since d_move() cannot
+		 * change the qstr (protected by d_lock).
+		 */
+		qstr = &dentry->d_name;
 		if (parent->d_op && parent->d_op->d_compare) {
 			if (parent->d_op->d_compare(parent, qstr, name))
 				goto next;
@@ -1028,7 +1256,6 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name)
 			atomic_inc(&dentry->d_count);
 			found = dentry;
 		}
-terminate:
 		spin_unlock(&dentry->d_lock);
 		break;
 next:
@@ -1039,6 +1266,32 @@ next:
  	return found;
 }
 
+/**
+ * d_hash_and_lookup - hash the qstr then search for a dentry
+ * @dir: Directory to search in
+ * @name: qstr of name we wish to find
+ *
+ * On hash failure or on lookup failure NULL is returned.
+ */
+struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name)
+{
+	struct dentry *dentry = NULL;
+
+	/*
+	 * Check for a fs-specific hash function. Note that we must
+	 * calculate the standard hash first, as the d_op->d_hash()
+	 * routine may choose to leave the hash value unchanged.
+	 */
+	name->hash = full_name_hash(name->name, name->len);
+	if (dir->d_op && dir->d_op->d_hash) {
+		if (dir->d_op->d_hash(dir, name) < 0)
+			goto out;
+	}
+	dentry = d_lookup(dir, name);
+out:
+	return dentry;
+}
+
 /**
  * d_validate - verify dentry provided from insecure source
  * @dentry: The dentry alleged to be valid child of @dparent
@@ -1066,7 +1319,7 @@ int d_validate(struct dentry *dentry, struct dentry *dparent)
 	spin_lock(&dcache_lock);
 	base = d_hash(dparent, dentry->d_name.hash);
 	hlist_for_each(lhp,base) { 
-		/* hlist_for_each_rcu() not required for d_hash list
+		/* hlist_for_each_entry_rcu() not required for d_hash list
 		 * as it is parsed under dcache_lock
 		 */
 		if (dentry == hlist_entry(lhp, struct dentry, d_hash)) {
@@ -1103,13 +1356,19 @@ out:
  
 void d_delete(struct dentry * dentry)
 {
+	int isdir = 0;
 	/*
 	 * Are we the only user?
 	 */
 	spin_lock(&dcache_lock);
 	spin_lock(&dentry->d_lock);
+	isdir = S_ISDIR(dentry->d_inode->i_mode);
 	if (atomic_read(&dentry->d_count) == 1) {
 		dentry_iput(dentry);
+		fsnotify_nameremove(dentry, isdir);
+
+		/* remove this and other inotify debug checks after 2.6.18 */
+		dentry->d_flags &= ~DCACHE_INOTIFY_PARENT_WATCHED;
 		return;
 	}
 
@@ -1118,6 +1377,20 @@ void d_delete(struct dentry * dentry)
 
 	spin_unlock(&dentry->d_lock);
 	spin_unlock(&dcache_lock);
+
+	fsnotify_nameremove(dentry, isdir);
+}
+
+static void __d_rehash(struct dentry * entry, struct hlist_head *list)
+{
+
+ 	entry->d_flags &= ~DCACHE_UNHASHED;
+ 	hlist_add_head_rcu(&entry->d_hash, list);
+}
+
+static void _d_rehash(struct dentry * entry)
+{
+	__d_rehash(entry, d_hash(entry->d_parent, entry->d_name.hash));
 }
 
 /**
@@ -1129,14 +1402,10 @@ void d_delete(struct dentry * dentry)
  
 void d_rehash(struct dentry * entry)
 {
-	struct hlist_head *list = d_hash(entry->d_parent, entry->d_name.hash);
-
 	spin_lock(&dcache_lock);
 	spin_lock(&entry->d_lock);
- 	entry->d_flags &= ~DCACHE_UNHASHED;
+	_d_rehash(entry);
 	spin_unlock(&entry->d_lock);
-	entry->d_bucket = list;
- 	hlist_add_head_rcu(&entry->d_hash, list);
 	spin_unlock(&dcache_lock);
 }
 
@@ -1214,6 +1483,8 @@ static void switch_names(struct dentry *dentry, struct dentry *target)
 
 void d_move(struct dentry * dentry, struct dentry * target)
 {
+	struct hlist_head *list;
+
 	if (!dentry->d_inode)
 		printk(KERN_WARNING "VFS: moving negative dcache entry\n");
 
@@ -1224,42 +1495,40 @@ void d_move(struct dentry * dentry, struct dentry * target)
 	 */
 	if (target < dentry) {
 		spin_lock(&target->d_lock);
-		spin_lock(&dentry->d_lock);
+		spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED);
 	} else {
 		spin_lock(&dentry->d_lock);
-		spin_lock(&target->d_lock);
+		spin_lock_nested(&target->d_lock, DENTRY_D_LOCK_NESTED);
 	}
 
 	/* Move the dentry to the target hash queue, if on different bucket */
 	if (dentry->d_flags & DCACHE_UNHASHED)
 		goto already_unhashed;
-	if (dentry->d_bucket != target->d_bucket) {
-		hlist_del_rcu(&dentry->d_hash);
+
+	hlist_del_rcu(&dentry->d_hash);
+
 already_unhashed:
-		dentry->d_bucket = target->d_bucket;
-		hlist_add_head_rcu(&dentry->d_hash, target->d_bucket);
-		dentry->d_flags &= ~DCACHE_UNHASHED;
-	}
+	list = d_hash(target->d_parent, target->d_name.hash);
+	__d_rehash(dentry, list);
 
 	/* Unhash the target: dput() will then get rid of it */
 	__d_drop(target);
 
-	/* flush any possible attributes */
-	if (dentry->d_extra_attributes) {
-		kfree(dentry->d_extra_attributes);
-		dentry->d_extra_attributes = NULL;
-	}
-	if (target->d_extra_attributes) {
-		kfree(target->d_extra_attributes);
-		target->d_extra_attributes = NULL;
-	}
+ 	/* flush any possible attributes */
+ 	if (dentry->d_extra_attributes) {
+ 		kfree(dentry->d_extra_attributes);
+ 		dentry->d_extra_attributes = NULL;
+ 	}
+ 	if (target->d_extra_attributes) {
+ 		kfree(target->d_extra_attributes);
+ 		target->d_extra_attributes = NULL;
+ 	}
 
-	list_del(&dentry->d_child);
-	list_del(&target->d_child);
+	list_del(&dentry->d_u.d_child);
+	list_del(&target->d_u.d_child);
 
 	/* Switch the names.. */
 	switch_names(dentry, target);
-	smp_wmb();
 	do_switch(dentry->d_name.len, target->d_name.len);
 	do_switch(dentry->d_name.hash, target->d_name.hash);
 
@@ -1267,21 +1536,136 @@ already_unhashed:
 	if (IS_ROOT(dentry)) {
 		dentry->d_parent = target->d_parent;
 		target->d_parent = target;
-		INIT_LIST_HEAD(&target->d_child);
+		INIT_LIST_HEAD(&target->d_u.d_child);
 	} else {
 		do_switch(dentry->d_parent, target->d_parent);
 
 		/* And add them back to the (new) parent lists */
-		list_add(&target->d_child, &target->d_parent->d_subdirs);
+		list_add(&target->d_u.d_child, &target->d_parent->d_subdirs);
 	}
 
-	list_add(&dentry->d_child, &dentry->d_parent->d_subdirs);
+	list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs);
 	spin_unlock(&target->d_lock);
+	fsnotify_d_move(dentry);
 	spin_unlock(&dentry->d_lock);
 	write_sequnlock(&rename_lock);
 	spin_unlock(&dcache_lock);
 }
 
+/*
+ * Prepare an anonymous dentry for life in the superblock's dentry tree as a
+ * named dentry in place of the dentry to be replaced.
+ */
+static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon)
+{
+	struct dentry *dparent, *aparent;
+
+	switch_names(dentry, anon);
+	do_switch(dentry->d_name.len, anon->d_name.len);
+	do_switch(dentry->d_name.hash, anon->d_name.hash);
+
+	dparent = dentry->d_parent;
+	aparent = anon->d_parent;
+
+	dentry->d_parent = (aparent == anon) ? dentry : aparent;
+	list_del(&dentry->d_u.d_child);
+	if (!IS_ROOT(dentry))
+		list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs);
+	else
+		INIT_LIST_HEAD(&dentry->d_u.d_child);
+
+	anon->d_parent = (dparent == dentry) ? anon : dparent;
+	list_del(&anon->d_u.d_child);
+	if (!IS_ROOT(anon))
+		list_add(&anon->d_u.d_child, &anon->d_parent->d_subdirs);
+	else
+		INIT_LIST_HEAD(&anon->d_u.d_child);
+
+	anon->d_flags &= ~DCACHE_DISCONNECTED;
+}
+
+/**
+ * d_materialise_unique - introduce an inode into the tree
+ * @dentry: candidate dentry
+ * @inode: inode to bind to the dentry, to which aliases may be attached
+ *
+ * Introduces an dentry into the tree, substituting an extant disconnected
+ * root directory alias in its place if there is one
+ */
+struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode)
+{
+	struct dentry *alias, *actual;
+
+	BUG_ON(!d_unhashed(dentry));
+
+	spin_lock(&dcache_lock);
+
+	if (!inode) {
+		actual = dentry;
+		dentry->d_inode = NULL;
+		goto found_lock;
+	}
+
+	/* See if a disconnected directory already exists as an anonymous root
+	 * that we should splice into the tree instead */
+	if (S_ISDIR(inode->i_mode) && (alias = __d_find_alias(inode, 1))) {
+		spin_lock(&alias->d_lock);
+
+		/* Is this a mountpoint that we could splice into our tree? */
+		if (IS_ROOT(alias))
+			goto connect_mountpoint;
+
+		if (alias->d_name.len == dentry->d_name.len &&
+		    alias->d_parent == dentry->d_parent &&
+		    memcmp(alias->d_name.name,
+			   dentry->d_name.name,
+			   dentry->d_name.len) == 0)
+			goto replace_with_alias;
+
+		spin_unlock(&alias->d_lock);
+
+		/* Doh! Seem to be aliasing directories for some reason... */
+		dput(alias);
+	}
+
+	/* Add a unique reference */
+	actual = __d_instantiate_unique(dentry, inode);
+	if (!actual)
+		actual = dentry;
+	else if (unlikely(!d_unhashed(actual)))
+		goto shouldnt_be_hashed;
+
+found_lock:
+	spin_lock(&actual->d_lock);
+found:
+	_d_rehash(actual);
+	spin_unlock(&actual->d_lock);
+	spin_unlock(&dcache_lock);
+
+	if (actual == dentry) {
+		security_d_instantiate(dentry, inode);
+		return NULL;
+	}
+
+	iput(inode);
+	return actual;
+
+	/* Convert the anonymous/root alias into an ordinary dentry */
+connect_mountpoint:
+	__d_materialise_dentry(dentry, alias);
+
+	/* Replace the candidate dentry with the alias in the tree */
+replace_with_alias:
+	__d_drop(alias);
+	actual = alias;
+	goto found;
+
+shouldnt_be_hashed:
+	spin_unlock(&dcache_lock);
+	BUG();
+	goto shouldnt_be_hashed;
+}
+
 /**
  * d_path - return the path of a dentry
  * @dentry: dentry to report
@@ -1478,7 +1862,6 @@ int is_subdir(struct dentry * new_dentry, struct dentry * old_dentry)
 	struct dentry * saved = new_dentry;
 	unsigned long seq;
 
-	result = 0;
 	/* need rcu_readlock to protect against the d_parent trashing due to
 	 * d_move
 	 */
@@ -1486,6 +1869,7 @@ int is_subdir(struct dentry * new_dentry, struct dentry * old_dentry)
         do {
 		/* for restarting inner loop in case of seq retry */
 		new_dentry = saved;
+		result = 0;
 		seq = read_seqbegin(&rename_lock);
 		for (;;) {
 			if (new_dentry != old_dentry) {
@@ -1515,7 +1899,7 @@ repeat:
 resume:
 	while (next != &this_parent->d_subdirs) {
 		struct list_head *tmp = next;
-		struct dentry *dentry = list_entry(tmp, struct dentry, d_child);
+		struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child);
 		next = tmp->next;
 		if (d_unhashed(dentry)||!dentry->d_inode)
 			continue;
@@ -1526,7 +1910,7 @@ resume:
 		atomic_dec(&dentry->d_count);
 	}
 	if (this_parent != root) {
-		next = this_parent->d_child.next; 
+		next = this_parent->d_u.d_child.next;
 		atomic_dec(&this_parent->d_count);
 		this_parent = this_parent->d_parent;
 		goto resume;
@@ -1553,26 +1937,12 @@ ino_t find_inode_number(struct dentry *dir, struct qstr *name)
 	struct dentry * dentry;
 	ino_t ino = 0;
 
-	/*
-	 * Check for a fs-specific hash function. Note that we must
-	 * calculate the standard hash first, as the d_op->d_hash()
-	 * routine may choose to leave the hash value unchanged.
-	 */
-	name->hash = full_name_hash(name->name, name->len);
-	if (dir->d_op && dir->d_op->d_hash)
-	{
-		if (dir->d_op->d_hash(dir, name) != 0)
-			goto out;
-	}
-
-	dentry = d_lookup(dir, name);
-	if (dentry)
-	{
+	dentry = d_hash_and_lookup(dir, name);
+	if (dentry) {
 		if (dentry->d_inode)
 			ino = dentry->d_inode->i_ino;
 		dput(dentry);
 	}
-out:
 	return ino;
 }
 
@@ -1590,14 +1960,21 @@ static void __init dcache_init_early(void)
 {
 	int loop;
 
+	/* If hashes are distributed across NUMA nodes, defer
+	 * hash allocation until vmalloc space is available.
+	 */
+	if (hashdist)
+		return;
+
 	dentry_hashtable =
 		alloc_large_system_hash("Dentry cache",
 					sizeof(struct hlist_head),
 					dhash_entries,
 					13,
-					0,
+					HASH_EARLY,
 					&d_hash_shift,
-					&d_hash_mask);
+					&d_hash_mask,
+					0);
 
 	for (loop = 0; loop < (1 << d_hash_shift); loop++)
 		INIT_HLIST_HEAD(&dentry_hashtable[loop]);
@@ -1622,6 +1999,8 @@ EXPORT_SYMBOL_GPL(flush_dentry_attributes);
 
 static void __init dcache_init(unsigned long mempages)
 {
+	int loop;
+
 	/* 
 	 * A constructor could be added for stable state like the lists,
 	 * but it is probably not worth it because of the cache nature
@@ -1630,17 +2009,35 @@ static void __init dcache_init(unsigned long mempages)
 	dentry_cache = kmem_cache_create("dentry_cache",
 					 sizeof(struct dentry),
 					 0,
-					 SLAB_RECLAIM_ACCOUNT|SLAB_PANIC,
+					 (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|
+					 SLAB_MEM_SPREAD),
 					 NULL, NULL);
 	
 	set_shrinker(DEFAULT_SEEKS, shrink_dcache_memory);
+
+	/* Hash may have been set up in dcache_init_early */
+	if (!hashdist)
+		return;
+
+	dentry_hashtable =
+		alloc_large_system_hash("Dentry cache",
+					sizeof(struct hlist_head),
+					dhash_entries,
+					13,
+					0,
+					&d_hash_shift,
+					&d_hash_mask,
+					0);
+
+	for (loop = 0; loop < (1 << d_hash_shift); loop++)
+		INIT_HLIST_HEAD(&dentry_hashtable[loop]);
 }
 
 /* SLAB cache for __getname() consumers */
-kmem_cache_t *names_cachep;
+kmem_cache_t *names_cachep __read_mostly;
 
 /* SLAB cache for file structures */
-kmem_cache_t *filp_cachep;
+kmem_cache_t *filp_cachep __read_mostly;
 
 EXPORT_SYMBOL(d_genocide);
 
@@ -1667,7 +2064,7 @@ void __init vfs_caches_init(unsigned long mempages)
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL);
 
 	filp_cachep = kmem_cache_create("filp", sizeof(struct file), 0,
-			SLAB_HWCACHE_ALIGN|SLAB_PANIC, filp_ctor, filp_dtor);
+			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL);
 
 	dcache_init(mempages);
 	inode_init(mempages);
@@ -1686,6 +2083,7 @@ EXPORT_SYMBOL(d_instantiate);
 EXPORT_SYMBOL(d_invalidate);
 EXPORT_SYMBOL(d_lookup);
 EXPORT_SYMBOL(d_move);
+EXPORT_SYMBOL_GPL(d_materialise_unique);
 EXPORT_SYMBOL(d_path);
 EXPORT_SYMBOL(d_prune_aliases);
 EXPORT_SYMBOL(d_rehash);