sysfs: Implement sysfs tagged directory support.

The problem.  When implementing a network namespace I need to be able
to have multiple network devices with the same name.  Currently this
is a problem for /sys/class/net/*, /sys/devices/virtual/net/*, and
potentially a few other directories of the form /sys/ ... /net/*.

What this patch does is to add an additional tag field to the
sysfs dirent structure.  For directories that should show different
contents depending on the context such as /sys/class/net/, and
/sys/devices/virtual/net/ this tag field is used to specify the
context in which those directories should be visible.  Effectively
this is the same as creating multiple distinct directories with
the same name but internally to sysfs the result is nicer.

I am calling the concept of a single directory that looks like multiple
directories all at the same path in the filesystem tagged directories.

For the networking namespace the set of directories whose contents I need
to filter with tags can depend on the presence or absence of hotplug
hardware or which modules are currently loaded.  Which means I need
a simple race free way to setup those directories as tagged.

To achieve a reace free design all tagged directories are created
and managed by sysfs itself.

Users of this interface:
- define a type in the sysfs_tag_type enumeration.
- call sysfs_register_ns_types with the type and it's operations
- sysfs_exit_ns when an individual tag is no longer valid

- Implement mount_ns() which returns the ns of the calling process
  so we can attach it to a sysfs superblock.
- Implement ktype.namespace() which returns the ns of a syfs kobject.

Everything else is left up to sysfs and the driver layer.

For the network namespace mount_ns and namespace() are essentially
one line functions, and look to remain that.

Tags are currently represented a const void * pointers as that is
both generic, prevides enough information for equality comparisons,
and is trivial to create for current users, as it is just the
existing namespace pointer.

The work needed in sysfs is more extensive.  At each directory
or symlink creating I need to check if the directory it is being
created in is a tagged directory and if so generate the appropriate
tag to place on the sysfs_dirent.  Likewise at each symlink or
directory removal I need to check if the sysfs directory it is
being removed from is a tagged directory and if so figure out
which tag goes along with the name I am deleting.

Currently only directories which hold kobjects, and
symlinks are supported.  There is not enough information
in the current file attribute interfaces to give us anything
to discriminate on which makes it useless, and there are
no potential users which makes it an uninteresting problem
to solve.

Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Benjamin Thery <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

drivers/gpio/gpiolib.c

@@ -399,7 +399,7 @@ static int gpio_setup_irq(struct gpio_desc *desc, struct device *dev,
 			goto free_id;
 		}
 
-		pdesc->value_sd = sysfs_get_dirent(dev->kobj.sd, "value");
+		pdesc->value_sd = sysfs_get_dirent(dev->kobj.sd, NULL, "value");
 		if (!pdesc->value_sd) {
 			ret = -ENODEV;
 			goto free_id;

drivers/md/bitmap.c

@@ -1678,9 +1678,9 @@ int bitmap_create(mddev_t *mddev)
 
 	bitmap->mddev = mddev;
 
-	bm = sysfs_get_dirent(mddev->kobj.sd, "bitmap");
+	bm = sysfs_get_dirent(mddev->kobj.sd, NULL, "bitmap");
 	if (bm) {
-		bitmap->sysfs_can_clear = sysfs_get_dirent(bm, "can_clear");
+		bitmap->sysfs_can_clear = sysfs_get_dirent(bm, NULL, "can_clear");
 		sysfs_put(bm);
 	} else
 		bitmap->sysfs_can_clear = NULL;

drivers/md/md.c

@@ -1766,7 +1766,7 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 		kobject_del(&rdev->kobj);
 		goto fail;
 	}
-	rdev->sysfs_state = sysfs_get_dirent(rdev->kobj.sd, "state");
+	rdev->sysfs_state = sysfs_get_dirent(rdev->kobj.sd, NULL, "state");
 
 	list_add_rcu(&rdev->same_set, &mddev->disks);
 	bd_claim_by_disk(rdev->bdev, rdev->bdev->bd_holder, mddev->gendisk);
@@ -4189,7 +4189,7 @@ static int md_alloc(dev_t dev, char *name)
 	mutex_unlock(&disks_mutex);
 	if (!error) {
 		kobject_uevent(&mddev->kobj, KOBJ_ADD);
-		mddev->sysfs_state = sysfs_get_dirent(mddev->kobj.sd, "array_state");
+		mddev->sysfs_state = sysfs_get_dirent(mddev->kobj.sd, NULL, "array_state");
 	}
 	mddev_put(mddev);
 	return error;
@@ -4398,7 +4398,7 @@ static int do_md_run(mddev_t * mddev)
 			printk(KERN_WARNING
 			       "md: cannot register extra attributes for %s\n",
 			       mdname(mddev));
-		mddev->sysfs_action = sysfs_get_dirent(mddev->kobj.sd, "sync_action");
+		mddev->sysfs_action = sysfs_get_dirent(mddev->kobj.sd, NULL, "sync_action");
 	} else if (mddev->ro == 2) /* auto-readonly not meaningful */
 		mddev->ro = 0;
 

fs/sysfs/bin.c

@@ -501,7 +501,7 @@ int sysfs_create_bin_file(struct kobject *kobj,
 void sysfs_remove_bin_file(struct kobject *kobj,
 			   const struct bin_attribute *attr)
 {
-	sysfs_hash_and_remove(kobj->sd, attr->attr.name);
+	sysfs_hash_and_remove(kobj->sd, NULL, attr->attr.name);
 }
 
 EXPORT_SYMBOL_GPL(sysfs_create_bin_file);

fs/sysfs/dir.c

@@ -380,9 +380,15 @@ int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
 {
 	struct sysfs_inode_attrs *ps_iattr;
 
-	if (sysfs_find_dirent(acxt->parent_sd, sd->s_name))
+	if (sysfs_find_dirent(acxt->parent_sd, sd->s_ns, sd->s_name))
 		return -EEXIST;
 
+	if (sysfs_ns_type(acxt->parent_sd) && !sd->s_ns) {
+		WARN(1, KERN_WARNING "sysfs: ns required in '%s' for '%s'\n",
+			acxt->parent_sd->s_name, sd->s_name);
+		return -EINVAL;
+	}
+
 	sd->s_parent = sysfs_get(acxt->parent_sd);
 
 	sysfs_link_sibling(sd);
@@ -533,13 +539,17 @@ void sysfs_addrm_finish(struct sysfs_addrm_cxt *acxt)
  *	Pointer to sysfs_dirent if found, NULL if not.
  */
 struct sysfs_dirent *sysfs_find_dirent(struct sysfs_dirent *parent_sd,
+				       const void *ns,
 				       const unsigned char *name)
 {
 	struct sysfs_dirent *sd;
 
-	for (sd = parent_sd->s_dir.children; sd; sd = sd->s_sibling)
+	for (sd = parent_sd->s_dir.children; sd; sd = sd->s_sibling) {
+		if (sd->s_ns != ns)
+			continue;
 		if (!strcmp(sd->s_name, name))
 			return sd;
+	}
 	return NULL;
 }
 
@@ -558,12 +568,13 @@ struct sysfs_dirent *sysfs_find_dirent(struct sysfs_dirent *parent_sd,
  *	Pointer to sysfs_dirent if found, NULL if not.
  */
 struct sysfs_dirent *sysfs_get_dirent(struct sysfs_dirent *parent_sd,
+				      const void *ns,
 				      const unsigned char *name)
 {
 	struct sysfs_dirent *sd;
 
 	mutex_lock(&sysfs_mutex);
-	sd = sysfs_find_dirent(parent_sd, name);
+	sd = sysfs_find_dirent(parent_sd, ns, name);
 	sysfs_get(sd);
 	mutex_unlock(&sysfs_mutex);
 
@@ -572,7 +583,8 @@ struct sysfs_dirent *sysfs_get_dirent(struct sysfs_dirent *parent_sd,
 EXPORT_SYMBOL_GPL(sysfs_get_dirent);
 
 static int create_dir(struct kobject *kobj, struct sysfs_dirent *parent_sd,
-		      const char *name, struct sysfs_dirent **p_sd)
+	enum kobj_ns_type type, const void *ns, const char *name,
+	struct sysfs_dirent **p_sd)
 {
 	umode_t mode = S_IFDIR| S_IRWXU | S_IRUGO | S_IXUGO;
 	struct sysfs_addrm_cxt acxt;
@@ -583,6 +595,9 @@ static int create_dir(struct kobject *kobj, struct sysfs_dirent *parent_sd,
 	sd = sysfs_new_dirent(name, mode, SYSFS_DIR);
 	if (!sd)
 		return -ENOMEM;
+
+	sd->s_flags |= (type << SYSFS_NS_TYPE_SHIFT);
+	sd->s_ns = ns;
 	sd->s_dir.kobj = kobj;
 
 	/* link in */
@@ -601,7 +616,25 @@ static int create_dir(struct kobject *kobj, struct sysfs_dirent *parent_sd,
 int sysfs_create_subdir(struct kobject *kobj, const char *name,
 			struct sysfs_dirent **p_sd)
 {
-	return create_dir(kobj, kobj->sd, name, p_sd);
+	return create_dir(kobj, kobj->sd,
+			  KOBJ_NS_TYPE_NONE, NULL, name, p_sd);
+}
+
+static enum kobj_ns_type sysfs_read_ns_type(struct kobject *kobj)
+{
+	const struct kobj_ns_type_operations *ops;
+	enum kobj_ns_type type;
+
+	ops = kobj_child_ns_ops(kobj);
+	if (!ops)
+		return KOBJ_NS_TYPE_NONE;
+
+	type = ops->type;
+	BUG_ON(type <= KOBJ_NS_TYPE_NONE);
+	BUG_ON(type >= KOBJ_NS_TYPES);
+	BUG_ON(!kobj_ns_type_registered(type));
+
+	return type;
 }
 
 /**
@@ -610,7 +643,9 @@ int sysfs_create_subdir(struct kobject *kobj, const char *name,
  */
 int sysfs_create_dir(struct kobject * kobj)
 {
+	enum kobj_ns_type type;
 	struct sysfs_dirent *parent_sd, *sd;
+	const void *ns = NULL;
 	int error = 0;
 
 	BUG_ON(!kobj);
@@ -620,7 +655,11 @@ int sysfs_create_dir(struct kobject * kobj)
 	else
 		parent_sd = &sysfs_root;
 
-	error = create_dir(kobj, parent_sd, kobject_name(kobj), &sd);
+	if (sysfs_ns_type(parent_sd))
+		ns = kobj->ktype->namespace(kobj);
+	type = sysfs_read_ns_type(kobj);
+
+	error = create_dir(kobj, parent_sd, type, ns, kobject_name(kobj), &sd);
 	if (!error)
 		kobj->sd = sd;
 	return error;
@@ -630,13 +669,19 @@ static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry,
 				struct nameidata *nd)
 {
 	struct dentry *ret = NULL;
-	struct sysfs_dirent *parent_sd = dentry->d_parent->d_fsdata;
+	struct dentry *parent = dentry->d_parent;
+	struct sysfs_dirent *parent_sd = parent->d_fsdata;
 	struct sysfs_dirent *sd;
 	struct inode *inode;
+	enum kobj_ns_type type;
+	const void *ns;
 
 	mutex_lock(&sysfs_mutex);
 
-	sd = sysfs_find_dirent(parent_sd, dentry->d_name.name);
+	type = sysfs_ns_type(parent_sd);
+	ns = sysfs_info(dir->i_sb)->ns[type];
+
+	sd = sysfs_find_dirent(parent_sd, ns, dentry->d_name.name);
 
 	/* no such entry */
 	if (!sd) {
@@ -735,20 +780,21 @@ void sysfs_remove_dir(struct kobject * kobj)
 }
 
 int sysfs_rename(struct sysfs_dirent *sd,
-	struct sysfs_dirent *new_parent_sd, const char *new_name)
+	struct sysfs_dirent *new_parent_sd, const void *new_ns,
+	const char *new_name)
 {
 	const char *dup_name = NULL;
 	int error;
 
 	mutex_lock(&sysfs_mutex);
 
 	error = 0;
-	if ((sd->s_parent == new_parent_sd) &&
+	if ((sd->s_parent == new_parent_sd) && (sd->s_ns == new_ns) &&
 	    (strcmp(sd->s_name, new_name) == 0))
 		goto out;	/* nothing to rename */
 
 	error = -EEXIST;
-	if (sysfs_find_dirent(new_parent_sd, new_name))
+	if (sysfs_find_dirent(new_parent_sd, new_ns, new_name))
 		goto out;
 
 	/* rename sysfs_dirent */
@@ -770,6 +816,7 @@ int sysfs_rename(struct sysfs_dirent *sd,
 		sd->s_parent = new_parent_sd;
 		sysfs_link_sibling(sd);
 	}
+	sd->s_ns = new_ns;
 
 	error = 0;
  out:
@@ -780,19 +827,28 @@ int sysfs_rename(struct sysfs_dirent *sd,
 
 int sysfs_rename_dir(struct kobject *kobj, const char *new_name)
 {
-	return sysfs_rename(kobj->sd, kobj->sd->s_parent, new_name);
+	struct sysfs_dirent *parent_sd = kobj->sd->s_parent;
+	const void *new_ns = NULL;
+
+	if (sysfs_ns_type(parent_sd))
+		new_ns = kobj->ktype->namespace(kobj);
+
+	return sysfs_rename(kobj->sd, parent_sd, new_ns, new_name);
 }
 
 int sysfs_move_dir(struct kobject *kobj, struct kobject *new_parent_kobj)
 {
 	struct sysfs_dirent *sd = kobj->sd;
 	struct sysfs_dirent *new_parent_sd;
+	const void *new_ns = NULL;
 
 	BUG_ON(!sd->s_parent);
+	if (sysfs_ns_type(sd->s_parent))
+		new_ns = kobj->ktype->namespace(kobj);
 	new_parent_sd = new_parent_kobj && new_parent_kobj->sd ?
 		new_parent_kobj->sd : &sysfs_root;
 
-	return sysfs_rename(sd, new_parent_sd, sd->s_name);
+	return sysfs_rename(sd, new_parent_sd, new_ns, sd->s_name);
 }
 
 /* Relationship between s_mode and the DT_xxx types */
@@ -807,32 +863,35 @@ static int sysfs_dir_release(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-static struct sysfs_dirent *sysfs_dir_pos(struct sysfs_dirent *parent_sd,
-	ino_t ino, struct sysfs_dirent *pos)
+static struct sysfs_dirent *sysfs_dir_pos(const void *ns,
+	struct sysfs_dirent *parent_sd,	ino_t ino, struct sysfs_dirent *pos)
 {
 	if (pos) {
 		int valid = !(pos->s_flags & SYSFS_FLAG_REMOVED) &&
 			pos->s_parent == parent_sd &&
 			ino == pos->s_ino;
 		sysfs_put(pos);
-		if (valid)
-			return pos;
+		if (!valid)
+			pos = NULL;
 	}
-	pos = NULL;
-	if ((ino > 1) && (ino < INT_MAX)) {
+	if (!pos && (ino > 1) && (ino < INT_MAX)) {
 		pos = parent_sd->s_dir.children;
 		while (pos && (ino > pos->s_ino))
 			pos = pos->s_sibling;
 	}
+	while (pos && pos->s_ns != ns)
+		pos = pos->s_sibling;
 	return pos;
 }
 
-static struct sysfs_dirent *sysfs_dir_next_pos(struct sysfs_dirent *parent_sd,
-	ino_t ino, struct sysfs_dirent *pos)
+static struct sysfs_dirent *sysfs_dir_next_pos(const void *ns,
+	struct sysfs_dirent *parent_sd,	ino_t ino, struct sysfs_dirent *pos)
 {
-	pos = sysfs_dir_pos(parent_sd, ino, pos);
+	pos = sysfs_dir_pos(ns, parent_sd, ino, pos);
 	if (pos)
 		pos = pos->s_sibling;
+	while (pos && pos->s_ns != ns)
+		pos = pos->s_sibling;
 	return pos;
 }
 
@@ -841,8 +900,13 @@ static int sysfs_readdir(struct file * filp, void * dirent, filldir_t filldir)
 	struct dentry *dentry = filp->f_path.dentry;
 	struct sysfs_dirent * parent_sd = dentry->d_fsdata;
 	struct sysfs_dirent *pos = filp->private_data;
+	enum kobj_ns_type type;
+	const void *ns;
 	ino_t ino;
 
+	type = sysfs_ns_type(parent_sd);
+	ns = sysfs_info(dentry->d_sb)->ns[type];
+
 	if (filp->f_pos == 0) {
 		ino = parent_sd->s_ino;
 		if (filldir(dirent, ".", 1, filp->f_pos, ino, DT_DIR) == 0)
@@ -857,9 +921,9 @@ static int sysfs_readdir(struct file * filp, void * dirent, filldir_t filldir)
 			filp->f_pos++;
 	}
 	mutex_lock(&sysfs_mutex);
-	for (pos = sysfs_dir_pos(parent_sd, filp->f_pos, pos);
+	for (pos = sysfs_dir_pos(ns, parent_sd, filp->f_pos, pos);
 	     pos;
-	     pos = sysfs_dir_next_pos(parent_sd, filp->f_pos, pos)) {
+	     pos = sysfs_dir_next_pos(ns, parent_sd, filp->f_pos, pos)) {
 		const char * name;
 		unsigned int type;
 		int len, ret;

fs/sysfs/file.c

@@ -478,9 +478,12 @@ void sysfs_notify(struct kobject *k, const char *dir, const char *attr)
 	mutex_lock(&sysfs_mutex);
 
 	if (sd && dir)
-		sd = sysfs_find_dirent(sd, dir);
+		/* Only directories are tagged, so no need to pass
+		 * a tag explicitly.
+		 */
+		sd = sysfs_find_dirent(sd, NULL, dir);
 	if (sd && attr)
-		sd = sysfs_find_dirent(sd, attr);
+		sd = sysfs_find_dirent(sd, NULL, attr);
 	if (sd)
 		sysfs_notify_dirent(sd);
 
@@ -569,7 +572,7 @@ int sysfs_add_file_to_group(struct kobject *kobj,
 	int error;
 
 	if (group)
-		dir_sd = sysfs_get_dirent(kobj->sd, group);
+		dir_sd = sysfs_get_dirent(kobj->sd, NULL, group);
 	else
 		dir_sd = sysfs_get(kobj->sd);
 
@@ -599,7 +602,7 @@ int sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode)
 	mutex_lock(&sysfs_mutex);
 
 	rc = -ENOENT;
-	sd = sysfs_find_dirent(kobj->sd, attr->name);
+	sd = sysfs_find_dirent(kobj->sd, NULL, attr->name);
 	if (!sd)
 		goto out;
 
@@ -624,7 +627,7 @@ EXPORT_SYMBOL_GPL(sysfs_chmod_file);
 
 void sysfs_remove_file(struct kobject * kobj, const struct attribute * attr)
 {
-	sysfs_hash_and_remove(kobj->sd, attr->name);
+	sysfs_hash_and_remove(kobj->sd, NULL, attr->name);
 }
 
 void sysfs_remove_files(struct kobject * kobj, const struct attribute **ptr)
@@ -646,11 +649,11 @@ void sysfs_remove_file_from_group(struct kobject *kobj,
 	struct sysfs_dirent *dir_sd;
 
 	if (group)
-		dir_sd = sysfs_get_dirent(kobj->sd, group);
+		dir_sd = sysfs_get_dirent(kobj->sd, NULL, group);
 	else
 		dir_sd = sysfs_get(kobj->sd);
 	if (dir_sd) {
-		sysfs_hash_and_remove(dir_sd, attr->name);
+		sysfs_hash_and_remove(dir_sd, NULL, attr->name);
 		sysfs_put(dir_sd);
 	}
 }