1 files changed, 243 insertions, 113 deletions

diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index d96e5c14a9c..394e90b02c5 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -26,10 +26,6 @@ static inline int autofs4_can_expire(struct dentry *dentry,
 	if (ino == NULL)
 		return 0;
 
-	/* No point expiring a pending mount */
-	if (dentry->d_flags & DCACHE_AUTOFS_PENDING)
-		return 0;
-
 	if (!do_now) {
 		/* Too young to die */
 		if (!timeout || time_after(ino->last_used + timeout, now))
@@ -48,23 +44,27 @@ static inline int autofs4_can_expire(struct dentry *dentry,
 static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry)
 {
 	struct dentry *top = dentry;
+	struct path path = {.mnt = mnt, .dentry = dentry};
 	int status = 1;
 
 	DPRINTK("dentry %p %.*s",
 		dentry, (int)dentry->d_name.len, dentry->d_name.name);
 
-	mntget(mnt);
-	dget(dentry);
+	path_get(&path);
 
-	if (!autofs4_follow_mount(&mnt, &dentry))
+	if (!follow_down_one(&path))
 		goto done;
 
-	/* This is an autofs submount, we can't expire it */
-	if (is_autofs4_dentry(dentry))
-		goto done;
+	if (is_autofs4_dentry(path.dentry)) {
+		struct autofs_sb_info *sbi = autofs4_sbi(path.dentry->d_sb);
+
+		/* This is an autofs submount, we can't expire it */
+		if (autofs_type_indirect(sbi->type))
+			goto done;
+	}
 
 	/* Update the expiry counter if fs is busy */
-	if (!may_umount_tree(mnt)) {
+	if (!may_umount_tree(path.mnt)) {
 		struct autofs_info *ino = autofs4_dentry_ino(top);
 		ino->last_used = jiffies;
 		goto done;
@@ -73,30 +73,118 @@ static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry)
 	status = 0;
 done:
 	DPRINTK("returning = %d", status);
-	mntput(mnt);
-	dput(dentry);
+	path_put(&path);
 	return status;
 }
 
 /*
- * Calculate next entry in top down tree traversal.
- * From next_mnt in namespace.c - elegant.
+ * Calculate and dget next entry in the subdirs list under root.
  */
-static struct dentry *next_dentry(struct dentry *p, struct dentry *root)
+static struct dentry *get_next_positive_subdir(struct dentry *prev,
+						struct dentry *root)
 {
-	struct list_head *next = p->d_subdirs.next;
+	struct autofs_sb_info *sbi = autofs4_sbi(root->d_sb);
+	struct list_head *next;
+	struct dentry *q;
+
+	spin_lock(&sbi->lookup_lock);
+	spin_lock(&root->d_lock);
+
+	if (prev)
+		next = prev->d_u.d_child.next;
+	else {
+		prev = dget_dlock(root);
+		next = prev->d_subdirs.next;
+	}
 
+cont:
+	if (next == &root->d_subdirs) {
+		spin_unlock(&root->d_lock);
+		spin_unlock(&sbi->lookup_lock);
+		dput(prev);
+		return NULL;
+	}
+
+	q = list_entry(next, struct dentry, d_u.d_child);
+
+	spin_lock_nested(&q->d_lock, DENTRY_D_LOCK_NESTED);
+	/* Already gone or negative dentry (under construction) - try next */
+	if (!d_count(q) || !simple_positive(q)) {
+		spin_unlock(&q->d_lock);
+		next = q->d_u.d_child.next;
+		goto cont;
+	}
+	dget_dlock(q);
+	spin_unlock(&q->d_lock);
+	spin_unlock(&root->d_lock);
+	spin_unlock(&sbi->lookup_lock);
+
+	dput(prev);
+
+	return q;
+}
+
+/*
+ * Calculate and dget next entry in top down tree traversal.
+ */
+static struct dentry *get_next_positive_dentry(struct dentry *prev,
+						struct dentry *root)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(root->d_sb);
+	struct list_head *next;
+	struct dentry *p, *ret;
+
+	if (prev == NULL)
+		return dget(root);
+
+	spin_lock(&sbi->lookup_lock);
+relock:
+	p = prev;
+	spin_lock(&p->d_lock);
+again:
+	next = p->d_subdirs.next;
 	if (next == &p->d_subdirs) {
 		while (1) {
-			if (p == root)
+			struct dentry *parent;
+
+			if (p == root) {
+				spin_unlock(&p->d_lock);
+				spin_unlock(&sbi->lookup_lock);
+				dput(prev);
 				return NULL;
+			}
+
+			parent = p->d_parent;
+			if (!spin_trylock(&parent->d_lock)) {
+				spin_unlock(&p->d_lock);
+				cpu_relax();
+				goto relock;
+			}
+			spin_unlock(&p->d_lock);
 			next = p->d_u.d_child.next;
-			if (next != &p->d_parent->d_subdirs)
+			p = parent;
+			if (next != &parent->d_subdirs)
 				break;
-			p = p->d_parent;
 		}
 	}
-	return list_entry(next, struct dentry, d_u.d_child);
+	ret = list_entry(next, struct dentry, d_u.d_child);
+
+	spin_lock_nested(&ret->d_lock, DENTRY_D_LOCK_NESTED);
+	/* Negative dentry - try next */
+	if (!simple_positive(ret)) {
+		spin_unlock(&p->d_lock);
+		lock_set_subclass(&ret->d_lock.dep_map, 0, _RET_IP_);
+		p = ret;
+		goto again;
+	}
+	dget_dlock(ret);
+	spin_unlock(&ret->d_lock);
+	spin_unlock(&p->d_lock);
+	spin_unlock(&sbi->lookup_lock);
+
+	dput(prev);
+
+	return ret;
 }
 
 /*
@@ -146,18 +234,11 @@ static int autofs4_tree_busy(struct vfsmount *mnt,
 	if (!simple_positive(top))
 		return 1;
 
-	spin_lock(&dcache_lock);
-	for (p = top; p; p = next_dentry(p, top)) {
-		/* Negative dentry - give up */
-		if (!simple_positive(p))
-			continue;
-
+	p = NULL;
+	while ((p = get_next_positive_dentry(p, top))) {
 		DPRINTK("dentry %p %.*s",
 			p, (int) p->d_name.len, p->d_name.name);
 
-		p = dget(p);
-		spin_unlock(&dcache_lock);
-
 		/*
 		 * Is someone visiting anywhere in the subtree ?
 		 * If there's no mount we need to check the usage
@@ -186,16 +267,13 @@ static int autofs4_tree_busy(struct vfsmount *mnt,
 			else
 				ino_count++;
 
-			if (atomic_read(&p->d_count) > ino_count) {
+			if (d_count(p) > ino_count) {
 				top_ino->last_used = jiffies;
 				dput(p);
 				return 1;
 			}
 		}
-		dput(p);
-		spin_lock(&dcache_lock);
 	}
-	spin_unlock(&dcache_lock);
 
 	/* Timeout of a tree mount is ultimately determined by its top dentry */
 	if (!autofs4_can_expire(top, timeout, do_now))
@@ -214,44 +292,34 @@ static struct dentry *autofs4_check_leaves(struct vfsmount *mnt,
 	DPRINTK("parent %p %.*s",
 		parent, (int)parent->d_name.len, parent->d_name.name);
 
-	spin_lock(&dcache_lock);
-	for (p = parent; p; p = next_dentry(p, parent)) {
-		/* Negative dentry - give up */
-		if (!simple_positive(p))
-			continue;
-
+	p = NULL;
+	while ((p = get_next_positive_dentry(p, parent))) {
 		DPRINTK("dentry %p %.*s",
 			p, (int) p->d_name.len, p->d_name.name);
 
-		p = dget(p);
-		spin_unlock(&dcache_lock);
-
 		if (d_mountpoint(p)) {
 			/* Can we umount this guy */
 			if (autofs4_mount_busy(mnt, p))
-				goto cont;
+				continue;
 
 			/* Can we expire this guy */
 			if (autofs4_can_expire(p, timeout, do_now))
 				return p;
 		}
-cont:
-		dput(p);
-		spin_lock(&dcache_lock);
 	}
-	spin_unlock(&dcache_lock);
 	return NULL;
 }
 
 /* Check if we can expire a direct mount (possibly a tree) */
-static struct dentry *autofs4_expire_direct(struct super_block *sb,
-					    struct vfsmount *mnt,
-					    struct autofs_sb_info *sbi,
-					    int how)
+struct dentry *autofs4_expire_direct(struct super_block *sb,
+				     struct vfsmount *mnt,
+				     struct autofs_sb_info *sbi,
+				     int how)
 {
 	unsigned long timeout;
 	struct dentry *root = dget(sb->s_root);
 	int do_now = how & AUTOFS_EXP_IMMEDIATE;
+	struct autofs_info *ino;
 
 	if (!root)
 		return NULL;
@@ -259,16 +327,19 @@ static struct dentry *autofs4_expire_direct(struct super_block *sb,
 	now = jiffies;
 	timeout = sbi->exp_timeout;
 
-	/* Lock the tree as we must expire as a whole */
 	spin_lock(&sbi->fs_lock);
+	ino = autofs4_dentry_ino(root);
+	/* No point expiring a pending mount */
+	if (ino->flags & AUTOFS_INF_PENDING)
+		goto out;
 	if (!autofs4_direct_busy(mnt, root, timeout, do_now)) {
 		struct autofs_info *ino = autofs4_dentry_ino(root);
-
-		/* Set this flag early to catch sys_chdir and the like */
 		ino->flags |= AUTOFS_INF_EXPIRING;
+		init_completion(&ino->expire_complete);
 		spin_unlock(&sbi->fs_lock);
 		return root;
 	}
+out:
 	spin_unlock(&sbi->fs_lock);
 	dput(root);
 
@@ -281,17 +352,19 @@ static struct dentry *autofs4_expire_direct(struct super_block *sb,
  *  - it is unused by any user process
  *  - it has been unused for exp_timeout time
  */
-static struct dentry *autofs4_expire_indirect(struct super_block *sb,
-					      struct vfsmount *mnt,
-					      struct autofs_sb_info *sbi,
-					      int how)
+struct dentry *autofs4_expire_indirect(struct super_block *sb,
+				       struct vfsmount *mnt,
+				       struct autofs_sb_info *sbi,
+				       int how)
 {
 	unsigned long timeout;
 	struct dentry *root = sb->s_root;
+	struct dentry *dentry;
 	struct dentry *expired = NULL;
-	struct list_head *next;
 	int do_now = how & AUTOFS_EXP_IMMEDIATE;
 	int exp_leaves = how & AUTOFS_EXP_LEAVES;
+	struct autofs_info *ino;
+	unsigned int ino_count;
 
 	if (!root)
 		return NULL;
@@ -299,22 +372,13 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 	now = jiffies;
 	timeout = sbi->exp_timeout;
 
-	spin_lock(&dcache_lock);
-	next = root->d_subdirs.next;
-
-	/* On exit from the loop expire is set to a dgot dentry
-	 * to expire or it's NULL */
-	while ( next != &root->d_subdirs ) {
-		struct dentry *dentry = list_entry(next, struct dentry, d_u.d_child);
-
-		/* Negative dentry - give up */
-		if (!simple_positive(dentry)) {
-			next = next->next;
-			continue;
-		}
-
-		dentry = dget(dentry);
-		spin_unlock(&dcache_lock);
+	dentry = NULL;
+	while ((dentry = get_next_positive_subdir(dentry, root))) {
+		spin_lock(&sbi->fs_lock);
+		ino = autofs4_dentry_ino(dentry);
+		/* No point expiring a pending mount */
+		if (ino->flags & AUTOFS_INF_PENDING)
+			goto next;
 
 		/*
 		 * Case 1: (i) indirect mount or top level pseudo direct mount
@@ -333,7 +397,21 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 			/* Can we expire this guy */
 			if (autofs4_can_expire(dentry, timeout, do_now)) {
 				expired = dentry;
-				break;
+				goto found;
+			}
+			goto next;
+		}
+
+		if (dentry->d_inode && S_ISLNK(dentry->d_inode->i_mode)) {
+			DPRINTK("checking symlink %p %.*s",
+				dentry, (int)dentry->d_name.len, dentry->d_name.name);
+			/*
+			 * A symlink can't be "busy" in the usual sense so
+			 * just check last used for expire timeout.
+			 */
+			if (autofs4_can_expire(dentry, timeout, do_now)) {
+				expired = dentry;
+				goto found;
 			}
 			goto next;
 		}
@@ -343,46 +421,80 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 
 		/* Case 2: tree mount, expire iff entire tree is not busy */
 		if (!exp_leaves) {
-			/* Lock the tree as we must expire as a whole */
-			spin_lock(&sbi->fs_lock);
-			if (!autofs4_tree_busy(mnt, dentry, timeout, do_now)) {
-				struct autofs_info *inf = autofs4_dentry_ino(dentry);
+			/* Path walk currently on this dentry? */
+			ino_count = atomic_read(&ino->count) + 1;
+			if (d_count(dentry) > ino_count)
+				goto next;
 
-				/* Set this flag early to catch sys_chdir and the like */
-				inf->flags |= AUTOFS_INF_EXPIRING;
-				spin_unlock(&sbi->fs_lock);
+			if (!autofs4_tree_busy(mnt, dentry, timeout, do_now)) {
 				expired = dentry;
-				break;
+				goto found;
 			}
-			spin_unlock(&sbi->fs_lock);
 		/*
 		 * Case 3: pseudo direct mount, expire individual leaves
 		 *	   (autofs-4.1).
 		 */
 		} else {
+			/* Path walk currently on this dentry? */
+			ino_count = atomic_read(&ino->count) + 1;
+			if (d_count(dentry) > ino_count)
+				goto next;
+
 			expired = autofs4_check_leaves(mnt, dentry, timeout, do_now);
 			if (expired) {
 				dput(dentry);
-				break;
+				goto found;
 			}
 		}
 next:
-		dput(dentry);
-		spin_lock(&dcache_lock);
-		next = next->next;
+		spin_unlock(&sbi->fs_lock);
 	}
+	return NULL;
+
+found:
+	DPRINTK("returning %p %.*s",
+		expired, (int)expired->d_name.len, expired->d_name.name);
+	ino = autofs4_dentry_ino(expired);
+	ino->flags |= AUTOFS_INF_EXPIRING;
+	init_completion(&ino->expire_complete);
+	spin_unlock(&sbi->fs_lock);
+	spin_lock(&sbi->lookup_lock);
+	spin_lock(&expired->d_parent->d_lock);
+	spin_lock_nested(&expired->d_lock, DENTRY_D_LOCK_NESTED);
+	list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child);
+	spin_unlock(&expired->d_lock);
+	spin_unlock(&expired->d_parent->d_lock);
+	spin_unlock(&sbi->lookup_lock);
+	return expired;
+}
 
-	if (expired) {
-		DPRINTK("returning %p %.*s",
-			expired, (int)expired->d_name.len, expired->d_name.name);
-		spin_lock(&dcache_lock);
-		list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child);
-		spin_unlock(&dcache_lock);
-		return expired;
+int autofs4_expire_wait(struct dentry *dentry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	int status;
+
+	/* Block on any pending expire */
+	spin_lock(&sbi->fs_lock);
+	if (ino->flags & AUTOFS_INF_EXPIRING) {
+		spin_unlock(&sbi->fs_lock);
+
+		DPRINTK("waiting for expire %p name=%.*s",
+			 dentry, dentry->d_name.len, dentry->d_name.name);
+
+		status = autofs4_wait(sbi, dentry, NFY_NONE);
+		wait_for_completion(&ino->expire_complete);
+
+		DPRINTK("expire done status=%d", status);
+
+		if (d_unhashed(dentry))
+			return -EAGAIN;
+
+		return status;
 	}
-	spin_unlock(&dcache_lock);
+	spin_unlock(&sbi->fs_lock);
 
-	return NULL;
+	return 0;
 }
 
 /* Perform an expiry operation */
@@ -392,7 +504,9 @@ int autofs4_expire_run(struct super_block *sb,
 		      struct autofs_packet_expire __user *pkt_p)
 {
 	struct autofs_packet_expire pkt;
+	struct autofs_info *ino;
 	struct dentry *dentry;
+	int ret = 0;
 
 	memset(&pkt,0,sizeof pkt);
 
@@ -408,39 +522,55 @@ int autofs4_expire_run(struct super_block *sb,
 	dput(dentry);
 
 	if ( copy_to_user(pkt_p, &pkt, sizeof(struct autofs_packet_expire)) )
-		return -EFAULT;
+		ret = -EFAULT;
 
-	return 0;
+	spin_lock(&sbi->fs_lock);
+	ino = autofs4_dentry_ino(dentry);
+	ino->flags &= ~AUTOFS_INF_EXPIRING;
+	complete_all(&ino->expire_complete);
+	spin_unlock(&sbi->fs_lock);
+
+	return ret;
 }
 
-/* Call repeatedly until it returns -EAGAIN, meaning there's nothing
-   more to be done */
-int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt,
-			struct autofs_sb_info *sbi, int __user *arg)
+int autofs4_do_expire_multi(struct super_block *sb, struct vfsmount *mnt,
+			    struct autofs_sb_info *sbi, int when)
 {
 	struct dentry *dentry;
 	int ret = -EAGAIN;
-	int do_now = 0;
 
-	if (arg && get_user(do_now, arg))
-		return -EFAULT;
-
-	if (sbi->type & AUTOFS_TYPE_DIRECT)
-		dentry = autofs4_expire_direct(sb, mnt, sbi, do_now);
+	if (autofs_type_trigger(sbi->type))
+		dentry = autofs4_expire_direct(sb, mnt, sbi, when);
 	else
-		dentry = autofs4_expire_indirect(sb, mnt, sbi, do_now);
+		dentry = autofs4_expire_indirect(sb, mnt, sbi, when);
 
 	if (dentry) {
 		struct autofs_info *ino = autofs4_dentry_ino(dentry);
 
 		/* This is synchronous because it makes the daemon a
                    little easier */
-		ino->flags |= AUTOFS_INF_EXPIRING;
 		ret = autofs4_wait(sbi, dentry, NFY_EXPIRE);
+
+		spin_lock(&sbi->fs_lock);
 		ino->flags &= ~AUTOFS_INF_EXPIRING;
+		complete_all(&ino->expire_complete);
+		spin_unlock(&sbi->fs_lock);
 		dput(dentry);
 	}
 
 	return ret;
 }
 
+/* Call repeatedly until it returns -EAGAIN, meaning there's nothing
+   more to be done */
+int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt,
+			struct autofs_sb_info *sbi, int __user *arg)
+{
+	int do_now = 0;
+
+	if (arg && get_user(do_now, arg))
+		return -EFAULT;
+
+	return autofs4_do_expire_multi(sb, mnt, sbi, do_now);
+}
+