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 I want is for each network namespace to have it's own separate
set of directories.  /sys/class/net/, /sys/devices/virtual/net,
and /sys/ ... /net/, and in each set I want to name them
/sys/class/net/, sys/devices/virtual/net/ and /sys/ ... /net/ respectively.

I looked and the VFS actually allows that.  All that is needed is
for /sys/class/net to implement a follow link method to redirect
lookups to the real directory you want.

I am calling the concept of multiple directories all at the same path
in the filesystem shadow directories, the directory entry that
implements the follow_link method the shadow master, and the directories
that are the target of the follow link method shadow directories.

It turns out that just implementing a follow_link method is not
quite enough.  The existince of directories of the form /sys/ ... /net/
can depend on the presence or absence of hotplug hardware, which
means I need a simple race free way to create and destroy these
directories.

To achieve a race free design all shadow directories are created
and managed by sysfs itself.  The upper level code that knows what
shadow directories we need provides just two methods that enable
this:
  current_tag() - that returns a "void *" tag that identifies the context of
	the current process.
  kobject_tag(kobj) - that returns a "void *" tag that identifies the context
	a kobject should be in.
Everything else is left up to sysfs.

For the network namespace current_tag and kobject_tag are essentially
one line functions, and look to remain that.

The work needed in sysfs is more extensive.  At each directory or
symlink creation I need to check if the shadow directory it belongs
in exists and if it does not create it.  Likewise at each symlink
or directory removal I need to check if sysfs directory it is being
removed from is a shadow directory and if this is the last object
in the shadow directory and if so to remove the shadow directory
as well.

I also need a bunch of boiler plate that properly finds, creates, and
removes/frees the shadow directories.

Doing all of that in sysfs isn't bad it is just a bit tedious.  Being race
free is just a manner of ensure we have the directory inode mutex when
we add or remove a shadow directory.  Trying to do this race free
anywhere besides in sysfs is very nasty, and requires unhealthy
amounts of information about how sysfs is implemented.

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
not potential users which makes it an uniteresting problem
to solve.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 fs/sysfs/bin.c        |    2 +-
 fs/sysfs/dir.c        |  347 ++++++++++++++++++++++++++++++++++++++++++++-----
 fs/sysfs/file.c       |    4 +-
 fs/sysfs/group.c      |   12 +-
 fs/sysfs/inode.c      |   15 ++-
 fs/sysfs/symlink.c    |   26 +++-
 fs/sysfs/sysfs.h      |    6 +-
 include/linux/sysfs.h |   14 ++
 8 files changed, 374 insertions(+), 52 deletions(-)

diff --git a/fs/sysfs/bin.c b/fs/sysfs/bin.c
index 3c5574a..b96a893 100644
--- a/fs/sysfs/bin.c
+++ b/fs/sysfs/bin.c
@@ -248,7 +248,7 @@ int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr)
 
 void sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr)
 {
-	if (sysfs_hash_and_remove(kobj->dentry, attr->attr.name) < 0) {
+	if (sysfs_hash_and_remove(kobj, kobj->dentry, attr->attr.name) < 0) {
 		printk(KERN_ERR "%s: "
 			"bad dentry or inode or no such file: \"%s\"\n",
 			__FUNCTION__, attr->attr.name);
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index b1da4fc..043464e 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -302,7 +302,8 @@ static void sysfs_attach_dentry(struct sysfs_dirent *sd, struct dentry *dentry)
 	sd->s_dentry = dentry;
 	spin_unlock(&sysfs_lock);
 
-	d_rehash(dentry);
+	if (dentry->d_flags & DCACHE_UNHASHED)
+		d_rehash(dentry);
 }
 
 void sysfs_attach_dirent(struct sysfs_dirent *sd,
@@ -348,12 +349,17 @@ static int create_dir(struct kobject *kobj, struct dentry *parent,
 	umode_t mode = S_IFDIR| S_IRWXU | S_IRUGO | S_IXUGO;
 	struct dentry *dentry;
 	struct inode *inode;
-	struct sysfs_dirent *sd;
+	struct sysfs_dirent *sd, *parent_sd;
 
 	mutex_lock(&parent->d_inode->i_mutex);
 
 	/* allocate */
-	dentry = lookup_one_len(name, parent, strlen(name));
+	parent_sd = sysfs_resolve_for_create(kobj, parent);
+	if (IS_ERR(parent_sd)) {
+		error = PTR_ERR(parent_sd);
+		goto out_unlock;
+	}
+	dentry = lookup_one_len(name, parent_sd->s_dentry, strlen(name));
 	if (IS_ERR(dentry)) {
 		error = PTR_ERR(dentry);
 		goto out_unlock;
@@ -382,12 +388,12 @@ static int create_dir(struct kobject *kobj, struct dentry *parent,
 
 	/* link in */
 	error = -EEXIST;
-	if (sysfs_dirent_exist(parent->d_fsdata, name))
+	if (sysfs_dirent_exist(parent_sd, name))
 		goto out_iput;
 
 	sysfs_instantiate(dentry, inode);
-	inc_nlink(parent->d_inode);
-	sysfs_attach_dirent(sd, parent->d_fsdata, dentry);
+	inc_nlink(parent_sd->s_dentry->d_inode);
+	sysfs_attach_dirent(sd, parent_sd, dentry);
 
 	*p_dentry = dentry;
 	error = 0;
@@ -502,9 +508,11 @@ static void remove_dir(struct dentry * d)
 
 	mutex_unlock(&parent->d_inode->i_mutex);
 
+	sysfs_prune_shadow_sd(sd->s_parent);
 	sysfs_drop_dentry(sd);
 	sysfs_deactivate(sd);
 	sysfs_put(sd);
+
 }
 
 void sysfs_remove_subdir(struct dentry * d)
@@ -512,6 +520,64 @@ void sysfs_remove_subdir(struct dentry * d)
 	remove_dir(d);
 }
 
+static void sysfs_empty_dir(struct dentry *dentry,
+				struct sysfs_dirent **removed)
+{
+	struct sysfs_dirent *parent_sd;
+	struct sysfs_dirent **pos;
+
+	parent_sd = dentry->d_fsdata;
+	pos = &parent_sd->s_children;
+	while (*pos) {
+		struct sysfs_dirent *sd = *pos;
+
+		if (sd->s_type && (sd->s_type & SYSFS_NOT_PINNED)) {
+			*pos = sd->s_sibling;
+			sd->s_sibling = *removed;
+			*removed = sd;
+		} else
+			pos = &(*pos)->s_sibling;
+	}
+}
+
+static void __sysfs_remove_empty_shadow(struct dentry *shadow)
+{
+	struct sysfs_dirent *sd;
+
+	if (d_unhashed(shadow))
+		return;
+
+	sd = shadow->d_fsdata;
+	sysfs_unlink_sibling(sd);
+	simple_rmdir(shadow->d_inode, shadow);
+	d_delete(shadow);
+	sysfs_put(sd);
+}
+
+static void sysfs_remove_empty_shadow(struct dentry *shadow)
+{
+	mutex_lock(&shadow->d_inode->i_mutex);
+	__sysfs_remove_empty_shadow(shadow);
+	mutex_unlock(&shadow->d_inode->i_mutex);
+}
+
+static void sysfs_remove_shadows(struct dentry *dentry,
+				struct sysfs_dirent **removed)
+{
+	struct sysfs_dirent *parent_sd, **pos;
+
+	parent_sd = dentry->d_fsdata;
+	pos = &parent_sd->s_children;
+	while (*pos) {
+		struct sysfs_dirent *sd = *pos;
+		struct dentry *shadow;
+
+		shadow = dget(sd->s_dentry);
+		sysfs_empty_dir(shadow, removed);
+		__sysfs_remove_empty_shadow(shadow);
+		dput(shadow);
+	}
+}
 
 /**
  *	sysfs_remove_dir - remove an object's directory.
@@ -524,10 +590,8 @@ void sysfs_remove_subdir(struct dentry * d)
 
 void sysfs_remove_dir(struct kobject * kobj)
 {
-	struct dentry *dentry = kobj->dentry;
+	struct dentry *dentry = dget(kobj->dentry);
 	struct sysfs_dirent *removed = NULL;
-	struct sysfs_dirent *parent_sd;
-	struct sysfs_dirent **pos;
 
 	spin_lock(&kobj_sysfs_assoc_lock);
 	kobj->dentry = NULL;
@@ -538,18 +602,10 @@ void sysfs_remove_dir(struct kobject * kobj)
 
 	pr_debug("sysfs %s: removing dir\n",dentry->d_name.name);
 	mutex_lock(&dentry->d_inode->i_mutex);
-	parent_sd = dentry->d_fsdata;
-	pos = &parent_sd->s_children;
-	while (*pos) {
-		struct sysfs_dirent *sd = *pos;
-
-		if (sd->s_type && (sd->s_type & SYSFS_NOT_PINNED)) {
-			*pos = sd->s_sibling;
-			sd->s_sibling = removed;
-			removed = sd;
-		} else
-			pos = &(*pos)->s_sibling;
-	}
+	if (dentry->d_inode->i_op == &sysfs_dir_inode_operations)
+		sysfs_empty_dir(dentry, &removed);
+	else
+		sysfs_remove_shadows(dentry, &removed);
 	mutex_unlock(&dentry->d_inode->i_mutex);
 
 	while (removed) {
@@ -586,7 +642,7 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name)
 	down_write(&sysfs_rename_sem);
 	mutex_lock(&inode->i_mutex);
 
-	parent_sd = parent->d_fsdata;
+	parent_sd = sysfs_resolve_for_create(kobj, parent);
 	new_dentry = lookup_one_len(new_name, parent_sd->s_dentry, strlen(new_name));
 	if (IS_ERR(new_dentry)) {
 		error = PTR_ERR(new_dentry);
@@ -637,6 +693,7 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name)
 	mutex_unlock(&inode->i_mutex);
 	up_write(&sysfs_rename_sem);
 
+	sysfs_prune_shadow_sd(parent->d_fsdata);
 	dput(parent);
 
 	return error;
@@ -644,25 +701,36 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name)
 
 int sysfs_move_dir(struct kobject *kobj, struct kobject *new_parent)
 {
+	struct inode *old_parent_inode, *new_parent_inode;
 	struct dentry *old_parent_dentry, *new_parent_dentry, *new_dentry;
 	struct sysfs_dirent *new_parent_sd, *sd;
 	int error;
 
-	old_parent_dentry = kobj->parent ?
-		kobj->parent->dentry : sysfs_mount->mnt_sb->s_root;
+	old_parent_dentry = kobj->dentry->d_parent;
 	new_parent_dentry = new_parent ?
 		new_parent->dentry : sysfs_mount->mnt_sb->s_root;
 
-	if (old_parent_dentry->d_inode == new_parent_dentry->d_inode)
+	old_parent_inode = old_parent_dentry->d_inode;
+	new_parent_inode = new_parent_dentry