[PATCH 0/2]v2 configfs: symlink() fixes

[PATCH 0/2]v2 configfs: symlink() fixes

[Louis Rilling]

[ applies on top of the previously submitted rename() vs rmdir() deadlock fix ]

Hi,

The following patchset fixes incorrect symlinks to dead items in configfs, which
are forbidden by specification.

The first patch actually prevents such dangling symlinks from being created, but
introduces a weird(?) behavior where a failing symlink() can make a racing
rmdir() fail in the symlink's parent and in the symlink's target as well. This
behavior is fixed with the next patch.

Changelog:
  - fix error code when symlink's target is being removed
  - re-implemented the weird(?) behavior fix in a way that does not temporarily
    instantiate the new symlink in the VFS.

Summary:
  configfs: Fix symlink() to a removing item
  configfs: Fix failing symlink() making rmdir() fail

 fs/configfs/configfs_internal.h |    1 +
 fs/configfs/dir.c               |   24 +++++++++++++++++-------
 fs/configfs/symlink.c           |   14 +++++++++++++-
 3 files changed, 31 insertions(+), 8 deletions(-)

-- 
Dr Louis Rilling			Kerlabs
Skype: louis.rilling			Batiment Germanium
Phone: (+33|0) 6 80 89 08 23		80 avenue des Buttes de Coesmes
http://www.kerlabs.com/			35700 Rennes

[PATCH 1/2] configfs: Fix symlink() to a removing item

[Louis Rilling]

The rule for configfs symlinks is that symlinks always point to valid
config_items, and prevent the target from being removed. However,
configfs_symlink() only checks that it can grab a reference on the target item,
without ensuring that it remains alive until the symlink is correctly attached.

This patch makes configfs_symlink() fail whenever the target is being removed,
using the CONFIGFS_USET_DROPPING flag set by configfs_detach_prep() and
protected by configfs_dirent_lock.

This patch introduces a similar (weird?) behavior as with mkdir failures making
rmdir fail: if symlink() races with rmdir() of the parent directory (or its
youngest user-created ancestor if parent is a default group) or rmdir() of the
target directory, and then fails in configfs_create(), this can make the racing
rmdir() fail despite the concerned directory having no user-created entry (resp.
no symlink pointing to it or one of its default groups) in the end.
This behavior is fixed in later patches.

Signed-off-by: Louis Rilling <louis.rilling@kerlabs.com>
---
 fs/configfs/dir.c     |   14 +++++++-------
 fs/configfs/symlink.c |    6 ++++++
 2 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 614e382..f2a12d0 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -370,6 +370,9 @@ static int configfs_detach_prep(struct dentry *dentry, struct mutex **wait_mutex
 	struct configfs_dirent *sd;
 	int ret;
 
+	/* Mark that we're trying to drop the group */
+	parent_sd->s_type |= CONFIGFS_USET_DROPPING;
+
 	ret = -EBUSY;
 	if (!list_empty(&parent_sd->s_links))
 		goto out;
@@ -385,8 +388,6 @@ static int configfs_detach_prep(struct dentry *dentry, struct mutex **wait_mutex
 					*wait_mutex = &sd->s_dentry->d_inode->i_mutex;
 				return -EAGAIN;
 			}
-			/* Mark that we're trying to drop the group */
-			sd->s_type |= CONFIGFS_USET_DROPPING;
 
 			/*
 			 * Yup, recursive.  If there's a problem, blame
@@ -414,12 +415,11 @@ static void configfs_detach_rollback(struct dentry *dentry)
 	struct configfs_dirent *parent_sd = dentry->d_fsdata;
 	struct configfs_dirent *sd;
 
-	list_for_each_entry(sd, &parent_sd->s_children, s_sibling) {
-		if (sd->s_type & CONFIGFS_USET_DEFAULT) {
+	parent_sd->s_type &= ~CONFIGFS_USET_DROPPING;
+
+	list_for_each_entry(sd, &parent_sd->s_children, s_sibling)
+		if (sd->s_type & CONFIGFS_USET_DEFAULT)
 			configfs_detach_rollback(sd->s_dentry);
-			sd->s_type &= ~CONFIGFS_USET_DROPPING;
-		}
-	}
 }
 
 static void detach_attrs(struct config_item * item)
diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c
index faeb441..b66908d 100644
--- a/fs/configfs/symlink.c
+++ b/fs/configfs/symlink.c
@@ -78,6 +78,12 @@ static int create_link(struct config_item *parent_item,
 	if (sl) {
 		sl->sl_target = config_item_get(item);
 		spin_lock(&configfs_dirent_lock);
+		if (target_sd->s_type & CONFIGFS_USET_DROPPING) {
+			spin_unlock(&configfs_dirent_lock);
+			config_item_put(item);
+			kfree(sl);
+			return -ENOENT;
+		}
 		list_add(&sl->sl_list, &target_sd->s_links);
 		spin_unlock(&configfs_dirent_lock);
 		ret = configfs_create_link(sl, parent_item->ci_dentry,
-- 
1.5.5.3

[PATCH 2/2] configfs: Fix failing symlink() making rmdir() fail

[Louis Rilling]

On a similar pattern as mkdir() vs rmdir(), a failing symlink() may make rmdir()
fail for the symlink's parent and the symlink's target as well.

failing symlink() making target's rmdir() fail:

	process 1:				process 2:
	symlink("A/S" -> "B")
	  allow_link()
	  create_link()
	    attach to "B" links list
						rmdir("B")
						  detach_prep("B")
						    error because of new link
	    configfs_create_link("A", "S")
	      error (eg -ENOMEM)

failing symlink() making parent's rmdir() fail:

	process 1:				process 2:
	symlink("A/D/S" -> "B")
	  allow_link()
	  create_link()
	    attach to "B" links list
	    configfs_create_link("A/D", "S")
	      make_dirent("A/D", "S")
						rmdir("A")
						  detach_prep("A")
						    detach_prep("A/D")
						      error because of "S"
	      create("S")
	        error (eg -ENOMEM)

We cannot use the same solution as for mkdir() vs rmdir(), since rmdir() on the
target cannot wait on the i_mutex of the new symlink's parent without risking a
deadlock (with other symlink() or sys_rename()). Instead we define a global
mutex protecting all configfs symlinks attachment, so that rmdir() can avoid the
races above.

Signed-off-by: Louis Rilling <louis.rilling@kerlabs.com>
---
 fs/configfs/configfs_internal.h |    1 +
 fs/configfs/dir.c               |   10 ++++++++++
 fs/configfs/symlink.c           |    8 +++++++-
 3 files changed, 18 insertions(+), 1 deletions(-)

diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h
index da015c1..5f61b26 100644
--- a/fs/configfs/configfs_internal.h
+++ b/fs/configfs/configfs_internal.h
@@ -51,6 +51,7 @@ struct configfs_dirent {
 #define CONFIGFS_USET_IN_MKDIR	0x0200
 #define CONFIGFS_NOT_PINNED	(CONFIGFS_ITEM_ATTR)
 
+extern struct mutex configfs_symlink_mutex;
 extern spinlock_t configfs_dirent_lock;
 
 extern struct vfsmount * configfs_mount;
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index f2a12d0..2c873fd 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -1202,6 +1202,11 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry)
 		return -EINVAL;
 	}
 
+	/*
+	 * Ensure that no racing symlink() will make detach_prep() fail while
+	 * the new link is temporarily attached
+	 */
+	mutex_lock(&configfs_symlink_mutex);
 	spin_lock(&configfs_dirent_lock);
 	do {
 		struct mutex *wait_mutex;
@@ -1210,6 +1215,7 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry)
 		if (ret) {
 			configfs_detach_rollback(dentry);
 			spin_unlock(&configfs_dirent_lock);
+			mutex_unlock(&configfs_symlink_mutex);
 			if (ret != -EAGAIN) {
 				config_item_put(parent_item);
 				return ret;
@@ -1219,10 +1225,12 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry)
 			mutex_lock(wait_mutex);
 			mutex_unlock(wait_mutex);
 
+			mutex_lock(&configfs_symlink_mutex);
 			spin_lock(&configfs_dirent_lock);
 		}
 	} while (ret == -EAGAIN);
 	spin_unlock(&configfs_dirent_lock);
+	mutex_unlock(&configfs_symlink_mutex);
 
 	/* Get a working ref for the duration of this function */
 	item = configfs_get_config_item(dentry);
@@ -1512,11 +1520,13 @@ void configfs_unregister_subsystem(struct configfs_subsystem *subsys)
 	mutex_lock_nested(&configfs_sb->s_root->d_inode->i_mutex,
 			  I_MUTEX_PARENT);
 	mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD);
+	mutex_lock(&configfs_symlink_mutex);
 	spin_lock(&configfs_dirent_lock);
 	if (configfs_detach_prep(dentry, NULL)) {
 		printk(KERN_ERR "configfs: Tried to unregister non-empty subsystem!\n");
 	}
 	spin_unlock(&configfs_dirent_lock);
+	mutex_unlock(&configfs_symlink_mutex);
 	configfs_detach_group(&group->cg_item);
 	dentry->d_inode->i_flags |= S_DEAD;
 	mutex_unlock(&dentry->d_inode->i_mutex);
diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c
index b66908d..21bd751 100644
--- a/fs/configfs/symlink.c
+++ b/fs/configfs/symlink.c
@@ -31,6 +31,9 @@
 #include <linux/configfs.h>
 #include "configfs_internal.h"
 
+/* Protects attachments of new symlinks */
+DEFINE_MUTEX(configfs_symlink_mutex);
+
 static int item_depth(struct config_item * item)
 {
 	struct config_item * p = item;
@@ -146,8 +149,11 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna
 		goto out_put;
 
 	ret = type->ct_item_ops->allow_link(parent_item, target_item);
-	if (!ret)
+	if (!ret) {
+		mutex_lock(&configfs_symlink_mutex);
 		ret = create_link(parent_item, target_item, dentry);
+		mutex_unlock(&configfs_symlink_mutex);
+	}
 
 	config_item_put(target_item);
 	path_put(&nd.path);
-- 
1.5.5.3

Re: [PATCH 0/2]v2 configfs: symlink() fixes

[Joel Becker]

On Mon, Jun 23, 2008 at 02:16:16PM +0200, Louis Rilling wrote:
> The following patchset fixes incorrect symlinks to dead items in configfs, which
> are forbidden by specification.
> 
> The first patch actually prevents such dangling symlinks from being created, but
> introduces a weird(?) behavior where a failing symlink() can make a racing
> rmdir() fail in the symlink's parent and in the symlink's target as well. This
> behavior is fixed with the next patch.

	Silly question: you've tested this, right?

Joel

-- 

"I almost ran over an angel
 He had a nice big fat cigar.
 'In a sense,' he said, 'You're alone here
 So if you jump, you'd best jump far.'"

Joel Becker
Principal Software Developer
Oracle
E-mail: joel.becker@oracle.com
Phone: (650) 506-8127

Re: [PATCH 0/2]v2 configfs: symlink() fixes

[Louis Rilling]

[-- Attachment #1.1: Type: text/plain, Size: 2051 bytes --]

On Mon, Jun 23, 2008 at 03:20:44PM -0700, Joel Becker wrote:
> On Mon, Jun 23, 2008 at 02:16:16PM +0200, Louis Rilling wrote:
> > The following patchset fixes incorrect symlinks to dead items in configfs, which
> > are forbidden by specification.
> > 
> > The first patch actually prevents such dangling symlinks from being created, but
> > introduces a weird(?) behavior where a failing symlink() can make a racing
> > rmdir() fail in the symlink's parent and in the symlink's target as well. This
> > behavior is fixed with the next patch.
> 
> 	Silly question: you've tested this, right?

Yup. I tested this on Linux 2.6.26-rc5 with a modified example_configfs.c
(attached), and on a backport for 2.6.20 with my own subsystem (based on
Kerrighed http://www.kerrighed.org/wiki/index.php/SchedConfig ).

Do you have some regression tests?

sample tests for symlinks:

mkdir -p /config/03-group-children/c1/c1.1
cd /config/03-group-children/c1/c1.1

# repeatedly create and remove a symlink to self
ln -s . link
rm link
ln -s . link
rm link

mkdir /config/02-simple-children/c2

# repeatedly create and remove a symlink to another item
ln -s /config/02-simple-children/c2 link
rm link
ln -s /config/02-simple-children/c2 link
rm link

# create two links with inverted source and target
ln -s /config/02-simple-children/c2 link
ln -s . /config/02-simple-children/c2/link
rm link
rm /config/02-simple-children/c2/link

# try hard to create a link to a removed item
# last loop is to be run in a separate shell
cd /config/03-group-children/c1
while true; do mkdir c1.1 && echo mkdir ok; rmdir c1.1 && echo rmdir ok; done
# in a separate shell
cd /config/02-simple-children/c2
while true; do ln -s /config/03-group-children/c1/c1.1 link && echo ln ok && \
(ls link/ || (echo failed!!!; read foo)) && rm link; done

Louis

-- 
Dr Louis Rilling			Kerlabs
Skype: louis.rilling			Batiment Germanium
Phone: (+33|0) 6 80 89 08 23		80 avenue des Buttes de Coesmes
http://www.kerlabs.com/			35700 Rennes

[-- Attachment #1.2: configfs_example.c --]
[-- Type: text/x-csrc, Size: 12773 bytes --]

/*
 * vim: noexpandtab ts=8 sts=0 sw=8:
 *
 * configfs_example.c - This file is a demonstration module containing
 *      a number of configfs subsystems.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License as published by the Free Software Foundation; either
 * version 2 of the License, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 *
 * Based on sysfs:
 * 	sysfs is Copyright (C) 2001, 2002, 2003 Patrick Mochel
 *
 * configfs Copyright (C) 2005 Oracle.  All rights reserved.
 */

#include <linux/init.h>
#include <linux/module.h>
#include <linux/slab.h>

#include <linux/configfs.h>



/*
 * 01-childless
 *
 * This first example is a childless subsystem.  It cannot create
 * any config_items.  It just has attributes.
 *
 * Note that we are enclosing the configfs_subsystem inside a container.
 * This is not necessary if a subsystem has no attributes directly
 * on the subsystem.  See the next example, 02-simple-children, for
 * such a subsystem.
 */

struct childless {
	struct configfs_subsystem subsys;
	int showme;
	int storeme;
};

struct childless_attribute {
	struct configfs_attribute attr;
	ssize_t (*show)(struct childless *, char *);
	ssize_t (*store)(struct childless *, const char *, size_t);
};

static inline struct childless *to_childless(struct config_item *item)
{
	return item ? container_of(to_configfs_subsystem(to_config_group(item)), struct childless, subsys) : NULL;
}

static ssize_t childless_showme_read(struct childless *childless,
				     char *page)
{
	ssize_t pos;

	pos = sprintf(page, "%d\n", childless->showme);
	childless->showme++;

	return pos;
}

static ssize_t childless_storeme_read(struct childless *childless,
				      char *page)
{
	return sprintf(page, "%d\n", childless->storeme);
}

static ssize_t childless_storeme_write(struct childless *childless,
				       const char *page,
				       size_t count)
{
	unsigned long tmp;
	char *p = (char *) page;

	tmp = simple_strtoul(p, &p, 10);
	if (!p || (*p && (*p != '\n')))
		return -EINVAL;

	if (tmp > INT_MAX)
		return -ERANGE;

	childless->storeme = tmp;

	return count;
}

static ssize_t childless_description_read(struct childless *childless,
					  char *page)
{
	return sprintf(page,
"[01-childless]\n"
"\n"
"The childless subsystem is the simplest possible subsystem in\n"
"configfs.  It does not support the creation of child config_items.\n"
"It only has a few attributes.  In fact, it isn't much different\n"
"than a directory in /proc.\n");
}

static struct childless_attribute childless_attr_showme = {
	.attr	= { .ca_owner = THIS_MODULE, .ca_name = "showme", .ca_mode = S_IRUGO },
	.show	= childless_showme_read,
};
static struct childless_attribute childless_attr_storeme = {
	.attr	= { .ca_owner = THIS_MODULE, .ca_name = "storeme", .ca_mode = S_IRUGO | S_IWUSR },
	.show	= childless_storeme_read,
	.store	= childless_storeme_write,
};
static struct childless_attribute childless_attr_description = {
	.attr = { .ca_owner = THIS_MODULE, .ca_name = "description", .ca_mode = S_IRUGO },
	.show = childless_description_read,
};

static struct configfs_attribute *childless_attrs[] = {
	&childless_attr_showme.attr,
	&childless_attr_storeme.attr,
	&childless_attr_description.attr,
	NULL,
};

static ssize_t childless_attr_show(struct config_item *item,
				   struct configfs_attribute *attr,
				   char *page)
{
	struct childless *childless = to_childless(item);
	struct childless_attribute *childless_attr =
		container_of(attr, struct childless_attribute, attr);
	ssize_t ret = 0;

	if (childless_attr->show)
		ret = childless_attr->show(childless, page);
	return ret;
}

static ssize_t childless_attr_store(struct config_item *item,
				    struct configfs_attribute *attr,
				    const char *page, size_t count)
{
	struct childless *childless = to_childless(item);
	struct childless_attribute *childless_attr =
		container_of(attr, struct childless_attribute, attr);
	ssize_t ret = -EINVAL;

	if (childless_attr->store)
		ret = childless_attr->store(childless, page, count);
	return ret;
}

static struct configfs_item_operations childless_item_ops = {
	.show_attribute		= childless_attr_show,
	.store_attribute	= childless_attr_store,
};

static struct config_item_type childless_type = {
	.ct_item_ops	= &childless_item_ops,
	.ct_attrs	= childless_attrs,
	.ct_owner	= THIS_MODULE,
};

static struct childless childless_subsys = {
	.subsys = {
		.su_group = {
			.cg_item = {
				.ci_namebuf = "01-childless",
				.ci_type = &childless_type,
			},
		},
	},
};


/* ----------------------------------------------------------------- */

/*
 * 02-simple-children
 *
 * This example merely has a simple one-attribute child.  Note that
 * there is no extra attribute structure, as the child's attribute is
 * known from the get-go.  Also, there is no container for the
 * subsystem, as it has no attributes of its own.
 */

struct simple_child {
	struct config_item item;
	int storeme;
};

static inline struct simple_child *to_simple_child(struct config_item *item)
{
	return item ? container_of(item, struct simple_child, item) : NULL;
}

static struct configfs_attribute simple_child_attr_storeme = {
	.ca_owner = THIS_MODULE,
	.ca_name = "storeme",
	.ca_mode = S_IRUGO | S_IWUSR,
};

static struct configfs_attribute *simple_child_attrs[] = {
	&simple_child_attr_storeme,
	NULL,
};

static ssize_t simple_child_attr_show(struct config_item *item,
				      struct configfs_attribute *attr,
				      char *page)
{
	ssize_t count;
	struct simple_child *simple_child = to_simple_child(item);

	count = sprintf(page, "%d\n", simple_child->storeme);

	return count;
}

static ssize_t simple_child_attr_store(struct config_item *item,
				       struct configfs_attribute *attr,
				       const char *page, size_t count)
{
	struct simple_child *simple_child = to_simple_child(item);
	unsigned long tmp;
	char *p = (char *) page;

	tmp = simple_strtoul(p, &p, 10);
	if (!p || (*p && (*p != '\n')))
		return -EINVAL;

	if (tmp > INT_MAX)
		return -ERANGE;

	simple_child->storeme = tmp;

	return count;
}

static void simple_child_release(struct config_item *item)
{
	kfree(to_simple_child(item));
}

/* Forward declaration */
static struct config_item_type simple_child_type;

static int simple_child_allow_link(struct config_item *src,
				   struct config_item *target)
{
	return (target->ci_type == &simple_child_type) ? 0 : -EPERM;
}

static struct configfs_item_operations simple_child_item_ops = {
	.release		= simple_child_release,
	.show_attribute		= simple_child_attr_show,
	.store_attribute	= simple_child_attr_store,
	.allow_link		= simple_child_allow_link,
};

static struct config_item_type simple_child_type = {
	.ct_item_ops	= &simple_child_item_ops,
	.ct_attrs	= simple_child_attrs,
	.ct_owner	= THIS_MODULE,
};


struct simple_children {
	struct config_group group;
};

static inline struct simple_children *to_simple_children(struct config_item *item)
{
	return item ? container_of(to_config_group(item), struct simple_children, group) : NULL;
}

static struct config_item *simple_children_make_item(struct config_group *group, const char *name)
{
	struct simple_child *simple_child;

	simple_child = kzalloc(sizeof(struct simple_child), GFP_KERNEL);
	if (!simple_child)
		return NULL;


	config_item_init_type_name(&simple_child->item, name,
				   &simple_child_type);

	simple_child->storeme = 0;

	return &simple_child->item;
}

static struct configfs_attribute simple_children_attr_description = {
	.ca_owner = THIS_MODULE,
	.ca_name = "description",
	.ca_mode = S_IRUGO,
};

static struct configfs_attribute *simple_children_attrs[] = {
	&simple_children_attr_description,
	NULL,
};

static ssize_t simple_children_attr_show(struct config_item *item,
			   		 struct configfs_attribute *attr,
			   		 char *page)
{
	return sprintf(page,
"[02-simple-children]\n"
"\n"
"This subsystem allows the creation of child config_items.  These\n"
"items have only one attribute that is readable and writeable.\n");
}

static void simple_children_release(struct config_item *item)
{
	kfree(to_simple_children(item));
}

static struct configfs_item_operations simple_children_item_ops = {
	.release 	= simple_children_release,
	.show_attribute	= simple_children_attr_show,
};

/*
 * Note that, since no extra work is required on ->drop_item(),
 * no ->drop_item() is provided.
 */
static struct configfs_group_operations simple_children_group_ops = {
	.make_item	= simple_children_make_item,
};

static struct config_item_type simple_children_type = {
	.ct_item_ops	= &simple_children_item_ops,
	.ct_group_ops	= &simple_children_group_ops,
	.ct_attrs	= simple_children_attrs,
	.ct_owner	= THIS_MODULE,
};

static struct configfs_subsystem simple_children_subsys = {
	.su_group = {
		.cg_item = {
			.ci_namebuf = "02-simple-children",
			.ci_type = &simple_children_type,
		},
	},
};


/* ----------------------------------------------------------------- */

/*
 * 03-group-children
 *
 * This example reuses the simple_children group from above.  However,
 * the simple_children group is not the subsystem itself, it is a
 * child of the subsystem.  Creation of a group in the subsystem creates
 * a new simple_children group.  That group can then have simple_child
 * children of its own.
 */

static struct config_group *group_children_make_group(struct config_group *group, const char *name)
{
	struct simple_children *simple_children;

	simple_children = kzalloc(sizeof(struct simple_children),
				  GFP_KERNEL);
	if (!simple_children)
		return NULL;


	config_group_init_type_name(&simple_children->group, name,
				    &simple_children_type);

	return &simple_children->group;
}

static struct configfs_attribute group_children_attr_description = {
	.ca_owner = THIS_MODULE,
	.ca_name = "description",
	.ca_mode = S_IRUGO,
};

static struct configfs_attribute *group_children_attrs[] = {
	&group_children_attr_description,
	NULL,
};

static ssize_t group_children_attr_show(struct config_item *item,
			   		struct configfs_attribute *attr,
			   		char *page)
{
	return sprintf(page,
"[03-group-children]\n"
"\n"
"This subsystem allows the creation of child config_groups.  These\n"
"groups are like the subsystem simple-children.\n");
}

static struct configfs_item_operations group_children_item_ops = {
	.show_attribute	= group_children_attr_show,
};

/*
 * Note that, since no extra work is required on ->drop_item(),
 * no ->drop_item() is provided.
 */
static struct configfs_group_operations group_children_group_ops = {
	.make_group	= group_children_make_group,
};

static struct config_item_type group_children_type = {
	.ct_item_ops	= &group_children_item_ops,
	.ct_group_ops	= &group_children_group_ops,
	.ct_attrs	= group_children_attrs,
	.ct_owner	= THIS_MODULE,
};

static struct configfs_subsystem group_children_subsys = {
	.su_group = {
		.cg_item = {
			.ci_namebuf = "03-group-children",
			.ci_type = &group_children_type,
		},
	},
};

/* ----------------------------------------------------------------- */

/*
 * We're now done with our subsystem definitions.
 * For convenience in this module, here's a list of them all.  It
 * allows the init function to easily register them.  Most modules
 * will only have one subsystem, and will only call register_subsystem
 * on it directly.
 */
static struct configfs_subsystem *example_subsys[] = {
	&childless_subsys.subsys,
	&simple_children_subsys,
	&group_children_subsys,
	NULL,
};

static int __init configfs_example_init(void)
{
	int ret;
	int i;
	struct configfs_subsystem *subsys;

	for (i = 0; example_subsys[i]; i++) {
		subsys = example_subsys[i];

		config_group_init(&subsys->su_group);
		mutex_init(&subsys->su_mutex);
		ret = configfs_register_subsystem(subsys);
		if (ret) {
			printk(KERN_ERR "Error %d while registering subsystem %s\n",
			       ret,
			       subsys->su_group.cg_item.ci_namebuf);
			goto out_unregister;
		}
	}

	return 0;

out_unregister:
	for (; i >= 0; i--) {
		configfs_unregister_subsystem(example_subsys[i]);
	}

	return ret;
}

static void __exit configfs_example_exit(void)
{
	int i;

	for (i = 0; example_subsys[i]; i++) {
		configfs_unregister_subsystem(example_subsys[i]);
	}
}

module_init(configfs_example_init);
module_exit(configfs_example_exit);
MODULE_LICENSE("GPL");


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH 0/2]v2 configfs: symlink() fixes

[Joel Becker]

On Tue, Jun 24, 2008 at 03:20:04PM +0200, Louis Rilling wrote:
> On Mon, Jun 23, 2008 at 03:20:44PM -0700, Joel Becker wrote:
> > 	Silly question: you've tested this, right?
> 
> Yup. I tested this on Linux 2.6.26-rc5 with a modified example_configfs.c
> (attached), and on a backport for 2.6.20 with my own subsystem (based on
> Kerrighed http://www.kerrighed.org/wiki/index.php/SchedConfig ).

	Excellent!

> Do you have some regression tests?

	No.  I've been meaning to create a complex configfs_example with
lots of different scenarios so I could write some.  Feel free to do so!
	Thank you for sending this along.

Joel

-- 

"There is a country in Europe where multiple-choice tests are
 illegal."
        - Sigfried Hulzer

Joel Becker
Principal Software Developer
Oracle
E-mail: joel.becker@oracle.com
Phone: (650) 506-8127