[PATCH v4 0/3] cgroup: add xattr support

[PATCH v4 0/3] cgroup: add xattr support

[Aristeu Rozanski]

This series are a refreshed version of a patchset submitted by Li Zefan back
in march:
	https://lkml.org/lkml/2012/3/1/13

With Li's permission, I refreshed the patches to apply over the latest upstream
and added the modifications suggested by others in the thread:
- using a mount option instead of config option to enable the xattr support
- reinitialize the list in kmem_xattrs_free()
- renamed functions to simple_xattr_*()

v4:
- implemented requested changes by Tejun Heo in patch #2

Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Aristeu Rozanski <aris@redhat.com>

[PATCH v4 1/3] xattr: extract simple_xattr code from tmpfs

[Aristeu Rozanski]

[-- Attachment #1: cgroup1.patch --]
[-- Type: text/plain, Size: 14732 bytes --]

From: Li Zefan <lizefan@huawei.com>

Extract in-memory xattr APIs from tmpfs. Will be used by cgroup.

$ size vmlinux.o
   text    data     bss     dec     hex filename
4658782  880729 5195032 10734543         a3cbcf vmlinux.o
$ size vmlinux.o
   text    data     bss     dec     hex filename
4658957  880729 5195032 10734718         a3cc7e vmlinux.o

v5:
- no changes
v4:
- move simple_xattrs_free() to fs/xattr.c
v3:
- in kmem_xattrs_free(), reinitialize the list
- use simple_xattr_* prefix
- introduce simple_xattr_add() to prevent direct list usage

Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Aristeu Rozanski <aris@redhat.com>

---
 fs/xattr.c               |  200 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/shmem_fs.h |    3 
 include/linux/xattr.h    |   25 +++++
 mm/shmem.c               |  171 +++-------------------------------------
 4 files changed, 240 insertions(+), 159 deletions(-)

Index: xattr/fs/xattr.c
===================================================================
--- xattr.orig/fs/xattr.c	2012-07-18 14:14:28.673977819 -0400
+++ xattr/fs/xattr.c	2012-07-18 14:14:29.746007233 -0400
@@ -783,3 +783,203 @@
 EXPORT_SYMBOL(generic_listxattr);
 EXPORT_SYMBOL(generic_setxattr);
 EXPORT_SYMBOL(generic_removexattr);
+
+/*
+ * initialize the simple_xattrs structure
+ */
+void simple_xattrs_init(struct simple_xattrs *xattrs)
+{
+	INIT_LIST_HEAD(&xattrs->head);
+	spin_lock_init(&xattrs->lock);
+}
+
+/*
+ * Allocate new xattr and copy in the value; but leave the name to callers.
+ */
+struct simple_xattr *simple_xattr_alloc(const void *value, size_t size)
+{
+	struct simple_xattr *new_xattr;
+	size_t len;
+
+	/* wrap around? */
+	len = sizeof(*new_xattr) + size;
+	if (len <= sizeof(*new_xattr))
+		return NULL;
+
+	new_xattr = kmalloc(len, GFP_KERNEL);
+	if (!new_xattr)
+		return NULL;
+
+	new_xattr->size = size;
+	memcpy(new_xattr->value, value, size);
+	return new_xattr;
+}
+
+/*
+ * free all the xattrs
+ */
+void simple_xattrs_free(struct simple_xattrs *xattrs)
+{
+	struct simple_xattr *xattr, *node;
+
+	spin_lock(&xattrs->lock);
+	list_for_each_entry_safe(xattr, node, &xattrs->head, list) {
+		kfree(xattr->name);
+		kfree(xattr);
+	}
+	INIT_LIST_HEAD(&xattrs->head);
+	spin_unlock(&xattrs->lock);
+}
+
+/*
+ * xattr GET operation for in-memory/pseudo filesystems
+ */
+int simple_xattr_get(struct simple_xattrs *xattrs, const char *name,
+		     void *buffer, size_t size)
+{
+	struct simple_xattr *xattr;
+	int ret = -ENODATA;
+
+	spin_lock(&xattrs->lock);
+	list_for_each_entry(xattr, &xattrs->head, list) {
+		if (strcmp(name, xattr->name))
+			continue;
+
+		ret = xattr->size;
+		if (buffer) {
+			if (size < xattr->size)
+				ret = -ERANGE;
+			else
+				memcpy(buffer, xattr->value, xattr->size);
+		}
+		break;
+	}
+	spin_unlock(&xattrs->lock);
+	return ret;
+}
+
+static int __simple_xattr_set(struct simple_xattrs *xattrs, const char *name,
+			      const void *value, size_t size, int flags)
+{
+	struct simple_xattr *xattr;
+	struct simple_xattr *new_xattr = NULL;
+	size_t len;
+	int err = 0;
+
+	/* value == NULL means remove */
+	if (value) {
+		/* wrap around? */
+		len = sizeof(*new_xattr) + size;
+		if (len <= sizeof(*new_xattr))
+			return -ENOMEM;
+
+		new_xattr = kmalloc(len, GFP_KERNEL);
+		if (!new_xattr)
+			return -ENOMEM;
+
+		new_xattr->name = kstrdup(name, GFP_KERNEL);
+		if (!new_xattr->name) {
+			kfree(new_xattr);
+			return -ENOMEM;
+		}
+
+		new_xattr->size = size;
+		memcpy(new_xattr->value, value, size);
+	}
+
+	spin_lock(&xattrs->lock);
+	list_for_each_entry(xattr, &xattrs->head, list) {
+		if (!strcmp(name, xattr->name)) {
+			if (flags & XATTR_CREATE) {
+				xattr = new_xattr;
+				err = -EEXIST;
+			} else if (new_xattr) {
+				list_replace(&xattr->list, &new_xattr->list);
+			} else {
+				list_del(&xattr->list);
+			}
+			goto out;
+		}
+	}
+	if (flags & XATTR_REPLACE) {
+		xattr = new_xattr;
+		err = -ENODATA;
+	} else {
+		list_add(&new_xattr->list, &xattrs->head);
+		xattr = NULL;
+	}
+out:
+	spin_unlock(&xattrs->lock);
+	if (xattr) {
+		kfree(xattr->name);
+		kfree(xattr);
+	}
+	return err;
+
+}
+
+/*
+ * xattr SET operation for in-memory/pseudo filesystems
+ */
+int simple_xattr_set(struct simple_xattrs *xattrs, const char *name,
+		     const void *value, size_t size, int flags)
+{
+	if (size == 0)
+		value = ""; /* empty EA, do not remove */
+	return __simple_xattr_set(xattrs, name, value, size, flags);
+}
+
+/*
+ * xattr REMOVE operation for in-memory/pseudo filesystems
+ */
+int simple_xattr_remove(struct simple_xattrs *xattrs, const char *name)
+{
+        return __simple_xattr_set(xattrs, name, NULL, 0, XATTR_REPLACE);
+}
+
+static bool xattr_is_trusted(const char *name)
+{
+	return !strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN);
+}
+
+/*
+ * xattr LIST operation for in-memory/pseudo filesystems
+ */
+ssize_t simple_xattr_list(struct simple_xattrs *xattrs, char *buffer, size_t size)
+{
+	bool trusted = capable(CAP_SYS_ADMIN);
+	struct simple_xattr *xattr;
+	size_t used = 0;
+
+	spin_lock(&xattrs->lock);
+	list_for_each_entry(xattr, &xattrs->head, list) {
+		size_t len;
+
+		/* skip "trusted." attributes for unprivileged callers */
+		if (!trusted && xattr_is_trusted(xattr->name))
+			continue;
+
+		len = strlen(xattr->name) + 1;
+		used += len;
+		if (buffer) {
+			if (size < used) {
+				used = -ERANGE;
+				break;
+			}
+			memcpy(buffer, xattr->name, len);
+			buffer += len;
+		}
+	}
+	spin_unlock(&xattrs->lock);
+
+	return used;
+}
+
+void simple_xattr_list_add(struct simple_xattrs *xattrs,
+			   struct simple_xattr *new_xattr)
+{
+	spin_lock(&xattrs->lock);
+	list_add(&new_xattr->list, &xattrs->head);
+	spin_unlock(&xattrs->lock);
+}
+
Index: xattr/include/linux/shmem_fs.h
===================================================================
--- xattr.orig/include/linux/shmem_fs.h	2012-07-18 14:14:28.673977819 -0400
+++ xattr/include/linux/shmem_fs.h	2012-07-18 14:14:29.750007343 -0400
@@ -5,6 +5,7 @@
 #include <linux/mempolicy.h>
 #include <linux/pagemap.h>
 #include <linux/percpu_counter.h>
+#include <linux/xattr.h>
 
 /* inode in-kernel data */
 
@@ -18,7 +19,7 @@
 	};
 	struct shared_policy	policy;		/* NUMA memory alloc policy */
 	struct list_head	swaplist;	/* chain of maybes on swap */
-	struct list_head	xattr_list;	/* list of shmem_xattr */
+	struct simple_xattrs	xattrs;		/* list of xattrs */
 	struct inode		vfs_inode;
 };
 
Index: xattr/include/linux/xattr.h
===================================================================
--- xattr.orig/include/linux/xattr.h	2012-07-18 14:14:28.673977819 -0400
+++ xattr/include/linux/xattr.h	2012-07-18 14:14:29.750007343 -0400
@@ -60,6 +60,7 @@
 #ifdef  __KERNEL__
 
 #include <linux/types.h>
+#include <linux/spinlock.h>
 
 struct inode;
 struct dentry;
@@ -96,6 +97,30 @@
 			   char **xattr_value, size_t size, gfp_t flags);
 int vfs_xattr_cmp(struct dentry *dentry, const char *xattr_name,
 		  const char *value, size_t size, gfp_t flags);
+
+struct simple_xattrs {
+	struct list_head head;
+	spinlock_t lock;
+};
+
+struct simple_xattr {
+	struct list_head list;
+	char *name;
+	size_t size;
+	char value[0];
+};
+
+void simple_xattrs_init(struct simple_xattrs *xattrs);
+struct simple_xattr *simple_xattr_alloc(const void *value, size_t size);
+void simple_xattrs_free(struct simple_xattrs *xattrs);
+int simple_xattr_get(struct simple_xattrs *xattrs, const char *name,
+		     void *buffer, size_t size);
+int simple_xattr_set(struct simple_xattrs *xattrs, const char *name,
+		     const void *value, size_t size, int flags);
+int simple_xattr_remove(struct simple_xattrs *xattrs, const char *name);
+ssize_t simple_xattr_list(struct simple_xattrs *xattrs, char *buffer, size_t size);
+void simple_xattr_list_add(struct simple_xattrs *xattrs, struct simple_xattr *new_xattr);
+
 #endif  /*  __KERNEL__  */
 
 #endif	/* _LINUX_XATTR_H */
Index: xattr/mm/shmem.c
===================================================================
--- xattr.orig/mm/shmem.c	2012-07-18 14:14:28.673977819 -0400
+++ xattr/mm/shmem.c	2012-07-18 14:14:29.750007343 -0400
@@ -77,13 +77,6 @@
 /* Symlink up to this size is kmalloc'ed instead of using a swappable page */
 #define SHORT_SYMLINK_LEN 128
 
-struct shmem_xattr {
-	struct list_head list;	/* anchored by shmem_inode_info->xattr_list */
-	char *name;		/* xattr name */
-	size_t size;
-	char value[0];
-};
-
 /*
  * shmem_fallocate and shmem_writepage communicate via inode->i_private
  * (with i_mutex making sure that it has only one user at a time):
@@ -627,7 +620,6 @@
 static void shmem_evict_inode(struct inode *inode)
 {
 	struct shmem_inode_info *info = SHMEM_I(inode);
-	struct shmem_xattr *xattr, *nxattr;
 
 	if (inode->i_mapping->a_ops == &shmem_aops) {
 		shmem_unacct_size(info->flags, inode->i_size);
@@ -641,10 +633,7 @@
 	} else
 		kfree(info->symlink);
 
-	list_for_each_entry_safe(xattr, nxattr, &info->xattr_list, list) {
-		kfree(xattr->name);
-		kfree(xattr);
-	}
+	simple_xattrs_free(&info->xattrs);
 	BUG_ON(inode->i_blocks);
 	shmem_free_inode(inode->i_sb);
 	clear_inode(inode);
@@ -1360,7 +1349,7 @@
 		spin_lock_init(&info->lock);
 		info->flags = flags & VM_NORESERVE;
 		INIT_LIST_HEAD(&info->swaplist);
-		INIT_LIST_HEAD(&info->xattr_list);
+		simple_xattrs_init(&info->xattrs);
 		cache_no_acl(inode);
 
 		switch (mode & S_IFMT) {
@@ -2134,28 +2123,6 @@
  */
 
 /*
- * Allocate new xattr and copy in the value; but leave the name to callers.
- */
-static struct shmem_xattr *shmem_xattr_alloc(const void *value, size_t size)
-{
-	struct shmem_xattr *new_xattr;
-	size_t len;
-
-	/* wrap around? */
-	len = sizeof(*new_xattr) + size;
-	if (len <= sizeof(*new_xattr))
-		return NULL;
-
-	new_xattr = kmalloc(len, GFP_KERNEL);
-	if (!new_xattr)
-		return NULL;
-
-	new_xattr->size = size;
-	memcpy(new_xattr->value, value, size);
-	return new_xattr;
-}
-
-/*
  * Callback for security_inode_init_security() for acquiring xattrs.
  */
 static int shmem_initxattrs(struct inode *inode,
@@ -2164,11 +2131,11 @@
 {
 	struct shmem_inode_info *info = SHMEM_I(inode);
 	const struct xattr *xattr;
-	struct shmem_xattr *new_xattr;
+	struct simple_xattr *new_xattr;
 	size_t len;
 
 	for (xattr = xattr_array; xattr->name != NULL; xattr++) {
-		new_xattr = shmem_xattr_alloc(xattr->value, xattr->value_len);
+		new_xattr = simple_xattr_alloc(xattr->value, xattr->value_len);
 		if (!new_xattr)
 			return -ENOMEM;
 
@@ -2185,91 +2152,12 @@
 		memcpy(new_xattr->name + XATTR_SECURITY_PREFIX_LEN,
 		       xattr->name, len);
 
-		spin_lock(&info->lock);
-		list_add(&new_xattr->list, &info->xattr_list);
-		spin_unlock(&info->lock);
+		simple_xattr_list_add(&info->xattrs, new_xattr);
 	}
 
 	return 0;
 }
 
-static int shmem_xattr_get(struct dentry *dentry, const char *name,
-			   void *buffer, size_t size)
-{
-	struct shmem_inode_info *info;
-	struct shmem_xattr *xattr;
-	int ret = -ENODATA;
-
-	info = SHMEM_I(dentry->d_inode);
-
-	spin_lock(&info->lock);
-	list_for_each_entry(xattr, &info->xattr_list, list) {
-		if (strcmp(name, xattr->name))
-			continue;
-
-		ret = xattr->size;
-		if (buffer) {
-			if (size < xattr->size)
-				ret = -ERANGE;
-			else
-				memcpy(buffer, xattr->value, xattr->size);
-		}
-		break;
-	}
-	spin_unlock(&info->lock);
-	return ret;
-}
-
-static int shmem_xattr_set(struct inode *inode, const char *name,
-			   const void *value, size_t size, int flags)
-{
-	struct shmem_inode_info *info = SHMEM_I(inode);
-	struct shmem_xattr *xattr;
-	struct shmem_xattr *new_xattr = NULL;
-	int err = 0;
-
-	/* value == NULL means remove */
-	if (value) {
-		new_xattr = shmem_xattr_alloc(value, size);
-		if (!new_xattr)
-			return -ENOMEM;
-
-		new_xattr->name = kstrdup(name, GFP_KERNEL);
-		if (!new_xattr->name) {
-			kfree(new_xattr);
-			return -ENOMEM;
-		}
-	}
-
-	spin_lock(&info->lock);
-	list_for_each_entry(xattr, &info->xattr_list, list) {
-		if (!strcmp(name, xattr->name)) {
-			if (flags & XATTR_CREATE) {
-				xattr = new_xattr;
-				err = -EEXIST;
-			} else if (new_xattr) {
-				list_replace(&xattr->list, &new_xattr->list);
-			} else {
-				list_del(&xattr->list);
-			}
-			goto out;
-		}
-	}
-	if (flags & XATTR_REPLACE) {
-		xattr = new_xattr;
-		err = -ENODATA;
-	} else {
-		list_add(&new_xattr->list, &info->xattr_list);
-		xattr = NULL;
-	}
-out:
-	spin_unlock(&info->lock);
-	if (xattr)
-		kfree(xattr->name);
-	kfree(xattr);
-	return err;
-}
-
 static const struct xattr_handler *shmem_xattr_handlers[] = {
 #ifdef CONFIG_TMPFS_POSIX_ACL
 	&generic_acl_access_handler,
@@ -2300,6 +2188,7 @@
 static ssize_t shmem_getxattr(struct dentry *dentry, const char *name,
 			      void *buffer, size_t size)
 {
+	struct shmem_inode_info *info = SHMEM_I(dentry->d_inode);
 	int err;
 
 	/*
@@ -2314,12 +2203,13 @@
 	if (err)
 		return err;
 
-	return shmem_xattr_get(dentry, name, buffer, size);
+	return simple_xattr_get(&info->xattrs, name, buffer, size);
 }
 
 static int shmem_setxattr(struct dentry *dentry, const char *name,
 			  const void *value, size_t size, int flags)
 {
+	struct shmem_inode_info *info = SHMEM_I(dentry->d_inode);
 	int err;
 
 	/*
@@ -2334,15 +2224,12 @@
 	if (err)
 		return err;
 
-	if (size == 0)
-		value = "";  /* empty EA, do not remove */
-
-	return shmem_xattr_set(dentry->d_inode, name, value, size, flags);
-
+	return simple_xattr_set(&info->xattrs, name, value, size, flags);
 }
 
 static int shmem_removexattr(struct dentry *dentry, const char *name)
 {
+	struct shmem_inode_info *info = SHMEM_I(dentry->d_inode);
 	int err;
 
 	/*
@@ -2357,45 +2244,13 @@
 	if (err)
 		return err;
 
-	return shmem_xattr_set(dentry->d_inode, name, NULL, 0, XATTR_REPLACE);
-}
-
-static bool xattr_is_trusted(const char *name)
-{
-	return !strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN);
+	return simple_xattr_remove(&info->xattrs, name);
 }
 
 static ssize_t shmem_listxattr(struct dentry *dentry, char *buffer, size_t size)
 {
-	bool trusted = capable(CAP_SYS_ADMIN);
-	struct shmem_xattr *xattr;
-	struct shmem_inode_info *info;
-	size_t used = 0;
-
-	info = SHMEM_I(dentry->d_inode);
-
-	spin_lock(&info->lock);
-	list_for_each_entry(xattr, &info->xattr_list, list) {
-		size_t len;
-
-		/* skip "trusted." attributes for unprivileged callers */
-		if (!trusted && xattr_is_trusted(xattr->name))
-			continue;
-
-		len = strlen(xattr->name) + 1;
-		used += len;
-		if (buffer) {
-			if (size < used) {
-				used = -ERANGE;
-				break;
-			}
-			memcpy(buffer, xattr->name, len);
-			buffer += len;
-		}
-	}
-	spin_unlock(&info->lock);
-
-	return used;
+	struct shmem_inode_info *info = SHMEM_I(dentry->d_inode);
+	return simple_xattr_list(&info->xattrs, buffer, size);
 }
 #endif /* CONFIG_TMPFS_XATTR */
 

[PATCH v4 2/3] cgroup: revise how we re-populate root directory

[Aristeu Rozanski]

[-- Attachment #1: cgroup2.patch --]
[-- Type: text/plain, Size: 5631 bytes --]

From: Li Zefan <lizefan@huawei.com>

When remounting cgroupfs with some subsystems added to it and some
removed, cgroup will remove all the files in root directory and then
re-popluate it.

What I'm doing here is, only remove files which belong to subsystems that
are to be unbinded, and only create files for newly-added subsystems.
The purpose is to have all other files untouched.

This is a preparation for cgroup xattr support.

v4:
- refactored cgroup_clear_directory() to not use cgroup_rm_file()
- instead of going thru the list of files, get the file list using the
  subsystems
- use 'subsys_mask' instead of {added,removed}_bits and made
  cgroup_populate_dir() to match the parameters with cgroup_clear_directory()
v3:
- refresh patches after recent refactoring

Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Aristeu Rozanski <aris@redhat.com>

---
 kernel/cgroup.c |   61 ++++++++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 48 insertions(+), 13 deletions(-)

Index: xattr/kernel/cgroup.c
===================================================================
--- xattr.orig/kernel/cgroup.c	2012-07-18 14:14:47.858504180 -0400
+++ xattr/kernel/cgroup.c	2012-07-18 14:15:04.826969685 -0400
@@ -824,7 +824,8 @@
 static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode);
 static struct dentry *cgroup_lookup(struct inode *, struct dentry *, struct nameidata *);
 static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry);
-static int cgroup_populate_dir(struct cgroup *cgrp);
+static int cgroup_populate_dir(struct cgroup *cgrp, bool base_files,
+			       unsigned long subsys_mask);
 static const struct inode_operations cgroup_dir_inode_operations;
 static const struct file_operations proc_cgroupstats_operations;
 
@@ -973,12 +974,29 @@
 	return -ENOENT;
 }
 
-static void cgroup_clear_directory(struct dentry *dir)
+/**
+ * cgroup_clear_directory - selective removal of base and subsystem files
+ * @dir: directory containing the files
+ * @base_files: true if the base files should be removed
+ * @subsys_mask: mask of the subsystem ids whose files should be removed
+ */
+static void cgroup_clear_directory(struct dentry *dir, bool base_files,
+				   unsigned long subsys_mask)
 {
 	struct cgroup *cgrp = __d_cgrp(dir);
+	struct cgroup_subsys *ss;
 
-	while (!list_empty(&cgrp->files))
-		cgroup_rm_file(cgrp, NULL);
+	for_each_subsys(cgrp->root, ss) {
+		struct cftype_set *set;
+		if (!test_bit(ss->subsys_id, &subsys_mask))
+			continue;
+		list_for_each_entry(set, &ss->cftsets, node)
+			cgroup_rm_file(cgrp, set->cfts);
+	}
+	if (base_files) {
+		while (!list_empty(&cgrp->files))
+			cgroup_rm_file(cgrp, NULL);
+	}
 }
 
 /*
@@ -987,8 +1005,9 @@
 static void cgroup_d_remove_dir(struct dentry *dentry)
 {
 	struct dentry *parent;
+	struct cgroupfs_root *root = dentry->d_sb->s_fs_info;
 
-	cgroup_clear_directory(dentry);
+	cgroup_clear_directory(dentry, true, root->subsys_bits);
 
 	parent = dentry->d_parent;
 	spin_lock(&parent->d_lock);
@@ -1353,6 +1372,7 @@
 	struct cgroupfs_root *root = sb->s_fs_info;
 	struct cgroup *cgrp = &root->top_cgroup;
 	struct cgroup_sb_opts opts;
+	unsigned long added_bits, removed_bits;
 
 	mutex_lock(&cgrp->dentry->d_inode->i_mutex);
 	mutex_lock(&cgroup_mutex);
@@ -1368,6 +1388,9 @@
 		pr_warning("cgroup: option changes via remount are deprecated (pid=%d comm=%s)\n",
 			   task_tgid_nr(current), current->comm);
 
+	added_bits = opts.subsys_bits & ~root->subsys_bits;
+	removed_bits = root->subsys_bits & ~opts.subsys_bits;
+
 	/* Don't allow flags or name to change at remount */
 	if (opts.flags != root->flags ||
 	    (opts.name && strcmp(opts.name, root->name))) {
@@ -1383,8 +1406,9 @@
 	}
 
 	/* clear out any existing files and repopulate subsystem files */
-	cgroup_clear_directory(cgrp->dentry);
-	cgroup_populate_dir(cgrp);
+	cgroup_clear_directory(cgrp->dentry, false, removed_bits);
+	/* re-populate subsystem files */
+	cgroup_populate_dir(cgrp, false, added_bits);
 
 	if (opts.release_agent)
 		strcpy(root->release_agent_path, opts.release_agent);
@@ -1684,7 +1708,7 @@
 		BUG_ON(root->number_of_cgroups != 1);
 
 		cred = override_creds(&init_cred);
-		cgroup_populate_dir(root_cgrp);
+		cgroup_populate_dir(root_cgrp, true, root->subsys_bits);
 		revert_creds(cred);
 		mutex_unlock(&cgroup_root_mutex);
 		mutex_unlock(&cgroup_mutex);
@@ -3858,18 +3882,29 @@
 	{ }	/* terminate */
 };
 
-static int cgroup_populate_dir(struct cgroup *cgrp)
+/**
+ * cgroup_populate_dir - selectively creation of files in a directory
+ * @cgrp: target cgroup
+ * @base_files: true if the base files should be added
+ * @subsys_mask: mask of the subsystem ids whose files should be added
+ */
+static int cgroup_populate_dir(struct cgroup *cgrp, bool base_files,
+			       unsigned long subsys_mask)
 {
 	int err;
 	struct cgroup_subsys *ss;
 
-	err = cgroup_addrm_files(cgrp, NULL, files, true);
-	if (err < 0)
-		return err;
+        if (base_files) {
+		err = cgroup_addrm_files(cgrp, NULL, files, true);
+		if (err < 0)
+			return err;
+	}
 
 	/* process cftsets of each subsystem */
 	for_each_subsys(cgrp->root, ss) {
 		struct cftype_set *set;
+		if (!test_bit(ss->subsys_id, &subsys_mask))
+			continue;
 
 		list_for_each_entry(set, &ss->cftsets, node)
 			cgroup_addrm_files(cgrp, ss, set->cfts, true);
@@ -4032,7 +4067,7 @@
 
 	list_add_tail(&cgrp->allcg_node, &root->allcg_list);
 
-	err = cgroup_populate_dir(cgrp);
+	err = cgroup_populate_dir(cgrp, true, root->subsys_bits);
 	/* If err < 0, we have a half-filled directory - oh well ;) */
 
 	mutex_unlock(&cgroup_mutex);

[PATCH v4 3/3] cgroup: add xattr support

[Aristeu Rozanski]

[-- Attachment #1: cgroup3.patch --]
[-- Type: text/plain, Size: 10145 bytes --]

From: Li Zefan <lizefan@huawei.com>

This is one of the items in the plumber's wish list.

For use cases:

>> What would the use case be for this?
>
> Attaching meta information to services, in an easily discoverable
> way. For example, in systemd we create one cgroup for each service, and
> could then store data like the main pid of the specific service as an
> xattr on the cgroup itself. That way we'd have almost all service state
> in the cgroupfs, which would make it possible to terminate systemd and
> later restart it without losing any state information. But there's more:
> for example, some very peculiar services cannot be terminated on
> shutdown (i.e. fakeraid DM stuff) and it would be really nice if the
> services in question could just mark that on their cgroup, by setting an
> xattr. On the more desktopy side of things there are other
> possibilities: for example there are plans defining what an application
> is along the lines of a cgroup (i.e. an app being a collection of
> processes). With xattrs one could then attach an icon or human readable
> program name on the cgroup.
>
> The key idea is that this would allow attaching runtime meta information
> to cgroups and everything they model (services, apps, vms), that doesn't
> need any complex userspace infrastructure, has good access control
> (i.e. because the file system enforces that anyway, and there's the
> "trusted." xattr namespace), notifications (inotify), and can easily be
> shared among applications.
>
> Lennart

v4:
- no changes
v3:
- instead of config option, use mount option to enable xattr support

Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Aristeu Rozanski <aris@redhat.com>

Index: xattr/include/linux/cgroup.h
===================================================================
--- xattr.orig/include/linux/cgroup.h	2012-07-18 09:31:04.388106982 -0400
+++ xattr/include/linux/cgroup.h	2012-07-18 10:00:08.187595868 -0400
@@ -17,6 +17,7 @@
 #include <linux/rwsem.h>
 #include <linux/idr.h>
 #include <linux/workqueue.h>
+#include <linux/xattr.h>
 
 #ifdef CONFIG_CGROUPS
 
@@ -216,6 +217,9 @@
 	/* List of events which userspace want to receive */
 	struct list_head event_list;
 	spinlock_t event_list_lock;
+
+	/* directory xattrs */
+	struct simple_xattrs xattrs;
 };
 
 /*
@@ -309,6 +313,9 @@
 	/* CFTYPE_* flags */
 	unsigned int flags;
 
+	/* file xattrs */
+	struct simple_xattrs xattrs;
+
 	int (*open)(struct inode *inode, struct file *file);
 	ssize_t (*read)(struct cgroup *cgrp, struct cftype *cft,
 			struct file *file,
@@ -394,7 +401,7 @@
  */
 struct cftype_set {
 	struct list_head		node;	/* chained at subsys->cftsets */
-	const struct cftype		*cfts;
+	struct cftype			*cfts;
 };
 
 struct cgroup_scanner {
@@ -406,8 +413,8 @@
 	void *data;
 };
 
-int cgroup_add_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts);
-int cgroup_rm_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts);
+int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts);
+int cgroup_rm_cftypes(struct cgroup_subsys *ss, struct cftype *cfts);
 
 int cgroup_is_removed(const struct cgroup *cgrp);
 
Index: xattr/kernel/cgroup.c
===================================================================
--- xattr.orig/kernel/cgroup.c	2012-07-18 09:48:19.914320313 -0400
+++ xattr/kernel/cgroup.c	2012-07-18 10:00:08.187595869 -0400
@@ -276,7 +276,8 @@
 
 /* bits in struct cgroupfs_root flags field */
 enum {
-	ROOT_NOPREFIX, /* mounted subsystems have no named prefix */
+	ROOT_NOPREFIX,	/* mounted subsystems have no named prefix */
+	ROOT_XATTR,	/* supports extended attributes */
 };
 
 static int cgroup_is_releasable(const struct cgroup *cgrp)
@@ -916,15 +917,19 @@
 		 */
 		BUG_ON(!list_empty(&cgrp->pidlists));
 
+		simple_xattrs_free(&cgrp->xattrs);
+
 		kfree_rcu(cgrp, rcu_head);
 	} else {
 		struct cfent *cfe = __d_cfe(dentry);
 		struct cgroup *cgrp = dentry->d_parent->d_fsdata;
+		struct cftype *cft = cfe->type;
 
 		WARN_ONCE(!list_empty(&cfe->node) &&
 			  cgrp != &cgrp->root->top_cgroup,
 			  "cfe still linked for %s\n", cfe->type->name);
 		kfree(cfe);
+		simple_xattrs_free(&cft->xattrs);
 	}
 	iput(inode);
 }
@@ -1154,6 +1159,8 @@
 		seq_printf(seq, ",%s", ss->name);
 	if (test_bit(ROOT_NOPREFIX, &root->flags))
 		seq_puts(seq, ",noprefix");
+	if (test_bit(ROOT_XATTR, &root->flags))
+		seq_puts(seq, ",xattr");
 	if (strlen(root->release_agent_path))
 		seq_printf(seq, ",release_agent=%s", root->release_agent_path);
 	if (clone_children(&root->top_cgroup))
@@ -1222,6 +1229,10 @@
 			opts->clone_children = true;
 			continue;
 		}
+		if (!strcmp(token, "xattr")) {
+			set_bit(ROOT_XATTR, &opts->flags);
+			continue;
+		}
 		if (!strncmp(token, "release_agent=", 14)) {
 			/* Specifying two release agents is forbidden */
 			if (opts->release_agent)
@@ -1439,6 +1450,7 @@
 	mutex_init(&cgrp->pidlist_mutex);
 	INIT_LIST_HEAD(&cgrp->event_list);
 	spin_lock_init(&cgrp->event_list_lock);
+	simple_xattrs_init(&cgrp->xattrs);
 }
 
 static void init_cgroup_root(struct cgroupfs_root *root)
@@ -1784,6 +1796,8 @@
 	mutex_unlock(&cgroup_root_mutex);
 	mutex_unlock(&cgroup_mutex);
 
+	simple_xattrs_free(&cgrp->xattrs);
+
 	kill_litter_super(sb);
 	cgroup_drop_root(root);
 }
@@ -2590,19 +2604,89 @@
 	return simple_rename(old_dir, old_dentry, new_dir, new_dentry);
 }
 
+static struct simple_xattrs *__d_xattrs(struct dentry *dentry)
+{
+	if (S_ISDIR(dentry->d_inode->i_mode))
+		return &__d_cgrp(dentry)->xattrs;
+	else
+		return &__d_cft(dentry)->xattrs;
+}
+
+static inline int xattr_enabled(struct dentry *dentry)
+{
+	struct cgroupfs_root *root = dentry->d_sb->s_fs_info;
+	return test_bit(ROOT_XATTR, &root->flags);
+}
+
+static bool is_valid_xattr(const char *name)
+{
+	if (!strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN) ||
+	    !strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) ||
+	    !strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN))
+		return true;
+	return false;
+}
+
+static int cgroup_setxattr(struct dentry *dentry, const char *name,
+			   const void *val, size_t size, int flags)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	if (!is_valid_xattr(name))
+		return -EINVAL;
+	return simple_xattr_set(__d_xattrs(dentry), name, val, size, flags);
+}
+
+static int cgroup_removexattr(struct dentry *dentry, const char *name)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	if (!is_valid_xattr(name))
+		return -EINVAL;
+	return simple_xattr_remove(__d_xattrs(dentry), name);
+}
+
+static ssize_t cgroup_getxattr(struct dentry *dentry, const char *name,
+			       void *buf, size_t size)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	if (!is_valid_xattr(name))
+		return -EINVAL;
+	return simple_xattr_get(__d_xattrs(dentry), name, buf, size);
+}
+
+static ssize_t cgroup_listxattr(struct dentry *dentry, char *buf, size_t size)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	return simple_xattr_list(__d_xattrs(dentry), buf, size);
+}
+
 static const struct file_operations cgroup_file_operations = {
-	.read = cgroup_file_read,
-	.write = cgroup_file_write,
-	.llseek = generic_file_llseek,
-	.open = cgroup_file_open,
-	.release = cgroup_file_release,
+	.read		= cgroup_file_read,
+	.write		= cgroup_file_write,
+	.llseek		= generic_file_llseek,
+	.open		= cgroup_file_open,
+	.release	= cgroup_file_release,
+};
+
+static const struct inode_operations cgroup_file_inode_operations = {
+	.setxattr	= cgroup_setxattr,
+	.getxattr	= cgroup_getxattr,
+	.listxattr	= cgroup_listxattr,
+	.removexattr	= cgroup_removexattr,
 };
 
 static const struct inode_operations cgroup_dir_inode_operations = {
-	.lookup = cgroup_lookup,
-	.mkdir = cgroup_mkdir,
-	.rmdir = cgroup_rmdir,
-	.rename = cgroup_rename,
+	.lookup		= cgroup_lookup,
+	.mkdir		= cgroup_mkdir,
+	.rmdir		= cgroup_rmdir,
+	.rename		= cgroup_rename,
+	.setxattr	= cgroup_setxattr,
+	.getxattr	= cgroup_getxattr,
+	.listxattr	= cgroup_listxattr,
+	.removexattr	= cgroup_removexattr,
 };
 
 static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
@@ -2650,6 +2734,7 @@
 	} else if (S_ISREG(mode)) {
 		inode->i_size = 0;
 		inode->i_fop = &cgroup_file_operations;
+		inode->i_op = &cgroup_file_inode_operations;
 	}
 	d_instantiate(dentry, inode);
 	dget(dentry);	/* Extra count - pin the dentry in core */
@@ -2710,7 +2795,7 @@
 }
 
 static int cgroup_add_file(struct cgroup *cgrp, struct cgroup_subsys *subsys,
-			   const struct cftype *cft)
+			   struct cftype *cft)
 {
 	struct dentry *dir = cgrp->dentry;
 	struct cgroup *parent = __d_cgrp(dir);
@@ -2720,6 +2805,8 @@
 	umode_t mode;
 	char name[MAX_CGROUP_TYPE_NAMELEN + MAX_CFTYPE_NAME + 2] = { 0 };
 
+	simple_xattrs_init(&cft->xattrs);
+
 	/* does @cft->flags tell us to skip creation on @cgrp? */
 	if ((cft->flags & CFTYPE_NOT_ON_ROOT) && !cgrp->parent)
 		return 0;
@@ -2760,9 +2847,9 @@
 }
 
 static int cgroup_addrm_files(struct cgroup *cgrp, struct cgroup_subsys *subsys,
-			      const struct cftype cfts[], bool is_add)
+			      struct cftype cfts[], bool is_add)
 {
-	const struct cftype *cft;
+	struct cftype *cft;
 	int err, ret = 0;
 
 	for (cft = cfts; cft->name[0] != '\0'; cft++) {
@@ -2796,7 +2883,7 @@
 }
 
 static void cgroup_cfts_commit(struct cgroup_subsys *ss,
-			       const struct cftype *cfts, bool is_add)
+			       struct cftype *cfts, bool is_add)
 	__releases(&cgroup_mutex) __releases(&cgroup_cft_mutex)
 {
 	LIST_HEAD(pending);
@@ -2847,7 +2934,7 @@
  * function currently returns 0 as long as @cfts registration is successful
  * even if some file creation attempts on existing cgroups fail.
  */
-int cgroup_add_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts)
+int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts)
 {
 	struct cftype_set *set;
 
@@ -2877,7 +2964,7 @@
  * Returns 0 on successful unregistration, -ENOENT if @cfts is not
  * registered with @ss.
  */
-int cgroup_rm_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts)
+int cgroup_rm_cftypes(struct cgroup_subsys *ss, struct cftype *cfts)
 {
 	struct cftype_set *set;
 

[PATCH v5 3/3] cgroup: add xattr support

[Aristeu Rozanski]

From: Li Zefan <lizefan@huawei.com>

This is one of the items in the plumber's wish list.

For use cases:

>> What would the use case be for this?
>
> Attaching meta information to services, in an easily discoverable
> way. For example, in systemd we create one cgroup for each service, and
> could then store data like the main pid of the specific service as an
> xattr on the cgroup itself. That way we'd have almost all service state
> in the cgroupfs, which would make it possible to terminate systemd and
> later restart it without losing any state information. But there's more:
> for example, some very peculiar services cannot be terminated on
> shutdown (i.e. fakeraid DM stuff) and it would be really nice if the
> services in question could just mark that on their cgroup, by setting an
> xattr. On the more desktopy side of things there are other
> possibilities: for example there are plans defining what an application
> is along the lines of a cgroup (i.e. an app being a collection of
> processes). With xattrs one could then attach an icon or human readable
> program name on the cgroup.
>
> The key idea is that this would allow attaching runtime meta information
> to cgroups and everything they model (services, apps, vms), that doesn't
> need any complex userspace infrastructure, has good access control
> (i.e. because the file system enforces that anyway, and there's the
> "trusted." xattr namespace), notifications (inotify), and can easily be
> shared among applications.
>
> Lennart

v5:
- check for capabilities before setting/removing xattrs
v4:
- no changes
v3:
- instead of config option, use mount option to enable xattr support

Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Aristeu Rozanski <aris@redhat.com>

---
 include/linux/cgroup.h |   13 +++-
 kernel/cgroup.c        |  130 ++++++++++++++++++++++++++++++++++++++++++-------
 2 files changed, 124 insertions(+), 19 deletions(-)

Index: xattr/include/linux/cgroup.h
===================================================================
--- xattr.orig/include/linux/cgroup.h	2012-07-18 15:13:53.493351282 -0400
+++ xattr/include/linux/cgroup.h	2012-07-18 15:14:01.829571940 -0400
@@ -17,6 +17,7 @@
 #include <linux/rwsem.h>
 #include <linux/idr.h>
 #include <linux/workqueue.h>
+#include <linux/xattr.h>
 
 #ifdef CONFIG_CGROUPS
 
@@ -216,6 +217,9 @@
 	/* List of events which userspace want to receive */
 	struct list_head event_list;
 	spinlock_t event_list_lock;
+
+	/* directory xattrs */
+	struct simple_xattrs xattrs;
 };
 
 /*
@@ -309,6 +313,9 @@
 	/* CFTYPE_* flags */
 	unsigned int flags;
 
+	/* file xattrs */
+	struct simple_xattrs xattrs;
+
 	int (*open)(struct inode *inode, struct file *file);
 	ssize_t (*read)(struct cgroup *cgrp, struct cftype *cft,
 			struct file *file,
@@ -394,7 +401,7 @@
  */
 struct cftype_set {
 	struct list_head		node;	/* chained at subsys->cftsets */
-	const struct cftype		*cfts;
+	struct cftype			*cfts;
 };
 
 struct cgroup_scanner {
@@ -406,8 +413,8 @@
 	void *data;
 };
 
-int cgroup_add_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts);
-int cgroup_rm_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts);
+int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts);
+int cgroup_rm_cftypes(struct cgroup_subsys *ss, struct cftype *cfts);
 
 int cgroup_is_removed(const struct cgroup *cgrp);
 
Index: xattr/kernel/cgroup.c
===================================================================
--- xattr.orig/kernel/cgroup.c	2012-07-18 15:14:01.817571622 -0400
+++ xattr/kernel/cgroup.c	2012-07-25 12:09:10.957226644 -0400
@@ -276,7 +276,8 @@
 
 /* bits in struct cgroupfs_root flags field */
 enum {
-	ROOT_NOPREFIX, /* mounted subsystems have no named prefix */
+	ROOT_NOPREFIX,	/* mounted subsystems have no named prefix */
+	ROOT_XATTR,	/* supports extended attributes */
 };
 
 static int cgroup_is_releasable(const struct cgroup *cgrp)
@@ -913,15 +914,19 @@
 		 */
 		BUG_ON(!list_empty(&cgrp->pidlists));
 
+		simple_xattrs_free(&cgrp->xattrs);
+
 		kfree_rcu(cgrp, rcu_head);
 	} else {
 		struct cfent *cfe = __d_cfe(dentry);
 		struct cgroup *cgrp = dentry->d_parent->d_fsdata;
+		struct cftype *cft = cfe->type;
 
 		WARN_ONCE(!list_empty(&cfe->node) &&
 			  cgrp != &cgrp->root->top_cgroup,
 			  "cfe still linked for %s\n", cfe->type->name);
 		kfree(cfe);
+		simple_xattrs_free(&cft->xattrs);
 	}
 	iput(inode);
 }
@@ -1144,6 +1149,8 @@
 		seq_printf(seq, ",%s", ss->name);
 	if (test_bit(ROOT_NOPREFIX, &root->flags))
 		seq_puts(seq, ",noprefix");
+	if (test_bit(ROOT_XATTR, &root->flags))
+		seq_puts(seq, ",xattr");
 	if (strlen(root->release_agent_path))
 		seq_printf(seq, ",release_agent=%s", root->release_agent_path);
 	if (clone_children(&root->top_cgroup))
@@ -1212,6 +1219,10 @@
 			opts->clone_children = true;
 			continue;
 		}
+		if (!strcmp(token, "xattr")) {
+			set_bit(ROOT_XATTR, &opts->flags);
+			continue;
+		}
 		if (!strncmp(token, "release_agent=", 14)) {
 			/* Specifying two release agents is forbidden */
 			if (opts->release_agent)
@@ -1429,6 +1440,7 @@
 	mutex_init(&cgrp->pidlist_mutex);
 	INIT_LIST_HEAD(&cgrp->event_list);
 	spin_lock_init(&cgrp->event_list_lock);
+	simple_xattrs_init(&cgrp->xattrs);
 }
 
 static void init_cgroup_root(struct cgroupfs_root *root)
@@ -1773,6 +1785,8 @@
 	mutex_unlock(&cgroup_root_mutex);
 	mutex_unlock(&cgroup_mutex);
 
+	simple_xattrs_free(&cgrp->xattrs);
+
 	kill_litter_super(sb);
 	cgroup_drop_root(root);
 }
@@ -2579,19 +2593,100 @@
 	return simple_rename(old_dir, old_dentry, new_dir, new_dentry);
 }
 
+static struct simple_xattrs *__d_xattrs(struct dentry *dentry)
+{
+	if (S_ISDIR(dentry->d_inode->i_mode))
+		return &__d_cgrp(dentry)->xattrs;
+	else
+		return &__d_cft(dentry)->xattrs;
+}
+
+static inline int xattr_enabled(struct dentry *dentry)
+{
+	struct cgroupfs_root *root = dentry->d_sb->s_fs_info;
+	return test_bit(ROOT_XATTR, &root->flags);
+}
+
+static bool is_valid_xattr(const char *name)
+{
+	if (!strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN) ||
+	    !strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) ||
+	    !strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN))
+		return true;
+	return false;
+}
+
+static bool check_permission(const char *name)
+{
+	if (!strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN))
+		return capable(CAP_IPC_LOCK);
+	return true;
+}
+
+static int cgroup_setxattr(struct dentry *dentry, const char *name,
+			   const void *val, size_t size, int flags)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	if (!check_permission(name))
+		return -EPERM;
+	if (!is_valid_xattr(name))
+		return -EINVAL;
+	return simple_xattr_set(__d_xattrs(dentry), name, val, size, flags);
+}
+
+static int cgroup_removexattr(struct dentry *dentry, const char *name)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	if (!check_permission(name))
+		return -EPERM;
+	if (!is_valid_xattr(name))
+		return -EINVAL;
+	return simple_xattr_remove(__d_xattrs(dentry), name);
+}
+
+static ssize_t cgroup_getxattr(struct dentry *dentry, const char *name,
+			       void *buf, size_t size)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	if (!is_valid_xattr(name))
+		return -EINVAL;
+	return simple_xattr_get(__d_xattrs(dentry), name, buf, size);
+}
+
+static ssize_t cgroup_listxattr(struct dentry *dentry, char *buf, size_t size)
+{
+	if (!xattr_enabled(dentry))
+		return -EOPNOTSUPP;
+	return simple_xattr_list(__d_xattrs(dentry), buf, size);
+}
+
 static const struct file_operations cgroup_file_operations = {
-	.read = cgroup_file_read,
-	.write = cgroup_file_write,
-	.llseek = generic_file_llseek,
-	.open = cgroup_file_open,
-	.release = cgroup_file_release,
+	.read		= cgroup_file_read,
+	.write		= cgroup_file_write,
+	.llseek		= generic_file_llseek,
+	.open		= cgroup_file_open,
+	.release	= cgroup_file_release,
+};
+
+static const struct inode_operations cgroup_file_inode_operations = {
+	.setxattr	= cgroup_setxattr,
+	.getxattr	= cgroup_getxattr,
+	.listxattr	= cgroup_listxattr,
+	.removexattr	= cgroup_removexattr,
 };
 
 static const struct inode_operations cgroup_dir_inode_operations = {
-	.lookup = cgroup_lookup,
-	.mkdir = cgroup_mkdir,
-	.rmdir = cgroup_rmdir,
-	.rename = cgroup_rename,
+	.lookup		= cgroup_lookup,
+	.mkdir		= cgroup_mkdir,
+	.rmdir		= cgroup_rmdir,
+	.rename		= cgroup_rename,
+	.setxattr	= cgroup_setxattr,
+	.getxattr	= cgroup_getxattr,
+	.listxattr	= cgroup_listxattr,
+	.removexattr	= cgroup_removexattr,
 };
 
 static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
@@ -2639,6 +2734,7 @@
 	} else if (S_ISREG(mode)) {
 		inode->i_size = 0;
 		inode->i_fop = &cgroup_file_operations;
+		inode->i_op = &cgroup_file_inode_operations;
 	}
 	d_instantiate(dentry, inode);
 	dget(dentry);	/* Extra count - pin the dentry in core */
@@ -2699,7 +2795,7 @@
 }
 
 static int cgroup_add_file(struct cgroup *cgrp, struct cgroup_subsys *subsys,
-			   const struct cftype *cft)
+			   struct cftype *cft)
 {
 	struct dentry *dir = cgrp->dentry;
 	struct cgroup *parent = __d_cgrp(dir);
@@ -2709,6 +2805,8 @@
 	umode_t mode;
 	char name[MAX_CGROUP_TYPE_NAMELEN + MAX_CFTYPE_NAME + 2] = { 0 };
 
+	simple_xattrs_init(&cft->xattrs);
+
 	/* does @cft->flags tell us to skip creation on @cgrp? */
 	if ((cft->flags & CFTYPE_NOT_ON_ROOT) && !cgrp->parent)
 		return 0;
@@ -2749,9 +2847,9 @@
 }
 
 static int cgroup_addrm_files(struct cgroup *cgrp, struct cgroup_subsys *subsys,
-			      const struct cftype cfts[], bool is_add)
+			      struct cftype cfts[], bool is_add)
 {
-	const struct cftype *cft;
+	struct cftype *cft;
 	int err, ret = 0;
 
 	for (cft = cfts; cft->name[0] != '\0'; cft++) {
@@ -2785,7 +2883,7 @@
 }
 
 static void cgroup_cfts_commit(struct cgroup_subsys *ss,
-			       const struct cftype *cfts, bool is_add)
+			       struct cftype *cfts, bool is_add)
 	__releases(&cgroup_mutex) __releases(&cgroup_cft_mutex)
 {
 	LIST_HEAD(pending);
@@ -2836,7 +2934,7 @@
  * function currently returns 0 as long as @cfts registration is successful
  * even if some file creation attempts on existing cgroups fail.
  */
-int cgroup_add_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts)
+int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts)
 {
 	struct cftype_set *set;
 
@@ -2866,7 +2964,7 @@
  * Returns 0 on successful unregistration, -ENOENT if @cfts is not
  * registered with @ss.
  */
-int cgroup_rm_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts)
+int cgroup_rm_cftypes(struct cgroup_subsys *ss, struct cftype *cfts)
 {
 	struct cftype_set *set;