[linux-block.git] / fs / sysfs / inode.c

/*
 * fs/sysfs/inode.c - basic sysfs inode and dentry operations
 *
 * Copyright (c) 2001-3 Patrick Mochel
 * Copyright (c) 2007 SUSE Linux Products GmbH
 * Copyright (c) 2007 Tejun Heo <teheo@suse.de>
 *
 * This file is released under the GPLv2.
 *
 * Please see Documentation/filesystems/sysfs.txt for more information.
 */

#undef DEBUG 

#include <linux/pagemap.h>
#include <linux/namei.h>
#include <linux/backing-dev.h>
#include <linux/capability.h>
#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/xattr.h>
#include <linux/security.h>
#include "sysfs.h"

extern struct super_block * sysfs_sb;

static const struct address_space_operations sysfs_aops = {
	.readpage	= simple_readpage,
	.write_begin	= simple_write_begin,
	.write_end	= simple_write_end,
};

static struct backing_dev_info sysfs_backing_dev_info = {
	.name		= "sysfs",
	.ra_pages	= 0,	/* No readahead */
	.capabilities	= BDI_CAP_NO_ACCT_AND_WRITEBACK,
};

static const struct inode_operations sysfs_inode_operations ={
	.setattr	= sysfs_setattr,
	.setxattr	= sysfs_setxattr,
};

int __init sysfs_inode_init(void)
{
	return bdi_init(&sysfs_backing_dev_info);
}

struct sysfs_inode_attrs *sysfs_init_inode_attrs(struct sysfs_dirent *sd)
{
	struct sysfs_inode_attrs *attrs;
	struct iattr *iattrs;

	attrs = kzalloc(sizeof(struct sysfs_inode_attrs), GFP_KERNEL);
	if (!attrs)
		return NULL;
	iattrs = &attrs->ia_iattr;

	/* assign default attributes */
	iattrs->ia_mode = sd->s_mode;
	iattrs->ia_uid = 0;
	iattrs->ia_gid = 0;
	iattrs->ia_atime = iattrs->ia_mtime = iattrs->ia_ctime = CURRENT_TIME;

	return attrs;
}
int sysfs_setattr(struct dentry * dentry, struct iattr * iattr)
{
	struct inode * inode = dentry->d_inode;
	struct sysfs_dirent * sd = dentry->d_fsdata;
	struct sysfs_inode_attrs *sd_attrs;
	struct iattr *iattrs;
	unsigned int ia_valid = iattr->ia_valid;
	int error;

	if (!sd)
		return -EINVAL;

	sd_attrs = sd->s_iattr;

	error = inode_change_ok(inode, iattr);
	if (error)
		return error;

	iattr->ia_valid &= ~ATTR_SIZE; /* ignore size changes */

	error = inode_setattr(inode, iattr);
	if (error)
		return error;

	if (!sd_attrs) {
		/* setting attributes for the first time, allocate now */
		sd_attrs = sysfs_init_inode_attrs(sd);
		if (!sd_attrs)
			return -ENOMEM;
		sd->s_iattr = sd_attrs;
	} else {
		/* attributes were changed at least once in past */
		iattrs = &sd_attrs->ia_iattr;

		if (ia_valid & ATTR_UID)
			iattrs->ia_uid = iattr->ia_uid;
		if (ia_valid & ATTR_GID)
			iattrs->ia_gid = iattr->ia_gid;
		if (ia_valid & ATTR_ATIME)
			iattrs->ia_atime = timespec_trunc(iattr->ia_atime,
					inode->i_sb->s_time_gran);
		if (ia_valid & ATTR_MTIME)
			iattrs->ia_mtime = timespec_trunc(iattr->ia_mtime,
					inode->i_sb->s_time_gran);
		if (ia_valid & ATTR_CTIME)
			iattrs->ia_ctime = timespec_trunc(iattr->ia_ctime,
					inode->i_sb->s_time_gran);
		if (ia_valid & ATTR_MODE) {
			umode_t mode = iattr->ia_mode;

			if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
				mode &= ~S_ISGID;
			iattrs->ia_mode = sd->s_mode = mode;
		}
	}
	return error;
}

int sysfs_setxattr(struct dentry *dentry, const char *name, const void *value,
		size_t size, int flags)
{
	struct sysfs_dirent *sd = dentry->d_fsdata;
	struct sysfs_inode_attrs *iattrs;
	void *secdata;
	int error;
	u32 secdata_len = 0;

	if (!sd)
		return -EINVAL;
	if (!sd->s_iattr)
		sd->s_iattr = sysfs_init_inode_attrs(sd);
	if (!sd->s_iattr)
		return -ENOMEM;

	iattrs = sd->s_iattr;

	if (!strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN)) {
		const char *suffix = name + XATTR_SECURITY_PREFIX_LEN;
		error = security_inode_setsecurity(dentry->d_inode, suffix,
						value, size, flags);
		if (error)
			goto out;
		error = security_inode_getsecctx(dentry->d_inode,
						&secdata, &secdata_len);
		if (error)
			goto out;
		if (iattrs->ia_secdata)
			security_release_secctx(iattrs->ia_secdata,
						iattrs->ia_secdata_len);
		iattrs->ia_secdata = secdata;
		iattrs->ia_secdata_len = secdata_len;

	} else
		return -EINVAL;
out:
	return error;
}

static inline void set_default_inode_attr(struct inode * inode, mode_t mode)
{
	inode->i_mode = mode;
	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
}

static inline void set_inode_attr(struct inode * inode, struct iattr * iattr)
{
	inode->i_mode = iattr->ia_mode;
	inode->i_uid = iattr->ia_uid;
	inode->i_gid = iattr->ia_gid;
	inode->i_atime = iattr->ia_atime;
	inode->i_mtime = iattr->ia_mtime;
	inode->i_ctime = iattr->ia_ctime;
}


/*
 * sysfs has a different i_mutex lock order behavior for i_mutex than other
 * filesystems; sysfs i_mutex is called in many places with subsystem locks
 * held. At the same time, many of the VFS locking rules do not apply to
 * sysfs at all (cross directory rename for example). To untangle this mess
 * (which gives false positives in lockdep), we're giving sysfs inodes their
 * own class for i_mutex.
 */
static struct lock_class_key sysfs_inode_imutex_key;

static int sysfs_count_nlink(struct sysfs_dirent *sd)
{
	struct sysfs_dirent *child;
	int nr = 0;

	for (child = sd->s_dir.children; child; child = child->s_sibling)
		if (sysfs_type(child) == SYSFS_DIR)
			nr++;

	return nr + 2;
}

static void sysfs_init_inode(struct sysfs_dirent *sd, struct inode *inode)
{
	struct bin_attribute *bin_attr;
	struct sysfs_inode_attrs *iattrs;

	inode->i_private = sysfs_get(sd);
	inode->i_mapping->a_ops = &sysfs_aops;
	inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info;
	inode->i_op = &sysfs_inode_operations;
	inode->i_ino = sd->s_ino;
	lockdep_set_class(&inode->i_mutex, &sysfs_inode_imutex_key);

	iattrs = sd->s_iattr;
	if (iattrs) {
		/* sysfs_dirent has non-default attributes
		 * get them for the new inode from persistent copy
		 * in sysfs_dirent
		 */
		set_inode_attr(inode, &iattrs->ia_iattr);
		if (iattrs->ia_secdata)
			security_inode_notifysecctx(inode,
						iattrs->ia_secdata,
						iattrs->ia_secdata_len);
	} else
		set_default_inode_attr(inode, sd->s_mode);

	/* initialize inode according to type */
	switch (sysfs_type(sd)) {
	case SYSFS_DIR:
		inode->i_op = &sysfs_dir_inode_operations;
		inode->i_fop = &sysfs_dir_operations;
		inode->i_nlink = sysfs_count_nlink(sd);
		break;
	case SYSFS_KOBJ_ATTR:
		inode->i_size = PAGE_SIZE;
		inode->i_fop = &sysfs_file_operations;
		break;
	case SYSFS_KOBJ_BIN_ATTR:
		bin_attr = sd->s_bin_attr.bin_attr;
		inode->i_size = bin_attr->size;
		inode->i_fop = &bin_fops;
		break;
	case SYSFS_KOBJ_LINK:
		inode->i_op = &sysfs_symlink_inode_operations;
		break;
	default:
		BUG();
	}

	unlock_new_inode(inode);
}

/**
 *	sysfs_get_inode - get inode for sysfs_dirent
 *	@sd: sysfs_dirent to allocate inode for
 *
 *	Get inode for @sd.  If such inode doesn't exist, a new inode
 *	is allocated and basics are initialized.  New inode is
 *	returned locked.
 *
 *	LOCKING:
 *	Kernel thread context (may sleep).
 *
 *	RETURNS:
 *	Pointer to allocated inode on success, NULL on failure.
 */
struct inode * sysfs_get_inode(struct sysfs_dirent *sd)
{
	struct inode *inode;

	inode = iget_locked(sysfs_sb, sd->s_ino);
	if (inode && (inode->i_state & I_NEW))
		sysfs_init_inode(sd, inode);

	return inode;
}

/*
 * The sysfs_dirent serves as both an inode and a directory entry for sysfs.
 * To prevent the sysfs inode numbers from being freed prematurely we take a
 * reference to sysfs_dirent from the sysfs inode.  A
 * super_operations.delete_inode() implementation is needed to drop that
 * reference upon inode destruction.
 */
void sysfs_delete_inode(struct inode *inode)
{
	struct sysfs_dirent *sd  = inode->i_private;

	truncate_inode_pages(&inode->i_data, 0);
	clear_inode(inode);
	sysfs_put(sd);
}

int sysfs_hash_and_remove(struct sysfs_dirent *dir_sd, const char *name)
{
	struct sysfs_addrm_cxt acxt;
	struct sysfs_dirent *sd;

	if (!dir_sd)
		return -ENOENT;

	sysfs_addrm_start(&acxt, dir_sd);

	sd = sysfs_find_dirent(dir_sd, name);
	if (sd)
		sysfs_remove_one(&acxt, sd);

	sysfs_addrm_finish(&acxt);

	if (sd)
		return 0;
	else
		return -ENOENT;
}
Commit	Line	Data
1da177e4	1	/*
6d66f5cd	2	* fs/sysfs/inode.c - basic sysfs inode and dentry operations
1da177e4	3	*
6d66f5cd TH	4	* Copyright (c) 2001-3 Patrick Mochel
	5	* Copyright (c) 2007 SUSE Linux Products GmbH
	6	* Copyright (c) 2007 Tejun Heo <teheo@suse.de>
	7	*
	8	* This file is released under the GPLv2.
1da177e4 LT	9	*
	10	* Please see Documentation/filesystems/sysfs.txt for more information.
	11	*/
	12
	13	#undef DEBUG
	14
	15	#include <linux/pagemap.h>
	16	#include <linux/namei.h>
	17	#include <linux/backing-dev.h>
16f7e0fe	18	#include <linux/capability.h>
995982ca	19	#include <linux/errno.h>
e8edc6e0	20	#include <linux/sched.h>
ddd29ec6 DQ	21	#include <linux/xattr.h>
ddd29ec6 DQ	22	#include <linux/security.h>
1da177e4 LT	23	#include "sysfs.h"
	24
	25	extern struct super_block * sysfs_sb;
	26
f5e54d6e	27	static const struct address_space_operations sysfs_aops = {
1da177e4	28	.readpage = simple_readpage,
800d15a5 NP	29	.write_begin = simple_write_begin,
800d15a5 NP	30	.write_end = simple_write_end,
1da177e4 LT	31	};
	32
	33	static struct backing_dev_info sysfs_backing_dev_info = {
d993831f	34	.name = "sysfs",
1da177e4	35	.ra_pages = 0, /* No readahead */
e4ad08fe	36	.capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK,
1da177e4 LT	37	};
1da177e4 LT	38
c5ef1c42	39	static const struct inode_operations sysfs_inode_operations ={
988d186d	40	.setattr = sysfs_setattr,
ddd29ec6	41	.setxattr = sysfs_setxattr,
988d186d MS	42	};
988d186d MS	43
e0bf68dd PZ	44	int __init sysfs_inode_init(void)
	45	{
	46	return bdi_init(&sysfs_backing_dev_info);
	47	}
	48
ddd29ec6 DQ	49	struct sysfs_inode_attrs sysfs_init_inode_attrs(struct sysfs_dirent sd)
	50	{
	51	struct sysfs_inode_attrs *attrs;
	52	struct iattr *iattrs;
	53
	54	attrs = kzalloc(sizeof(struct sysfs_inode_attrs), GFP_KERNEL);
	55	if (!attrs)
	56	return NULL;
	57	iattrs = &attrs->ia_iattr;
	58
	59	/* assign default attributes */
	60	iattrs->ia_mode = sd->s_mode;
	61	iattrs->ia_uid = 0;
	62	iattrs->ia_gid = 0;
	63	iattrs->ia_atime = iattrs->ia_mtime = iattrs->ia_ctime = CURRENT_TIME;
	64
	65	return attrs;
	66	}
988d186d MS	67	int sysfs_setattr(struct dentry * dentry, struct iattr * iattr)
	68	{
	69	struct inode * inode = dentry->d_inode;
	70	struct sysfs_dirent * sd = dentry->d_fsdata;
ddd29ec6 DQ	71	struct sysfs_inode_attrs *sd_attrs;
ddd29ec6 DQ	72	struct iattr *iattrs;
988d186d MS	73	unsigned int ia_valid = iattr->ia_valid;
	74	int error;
	75
	76	if (!sd)
	77	return -EINVAL;
	78
ddd29ec6	79	sd_attrs = sd->s_iattr;
988d186d MS	80
	81	error = inode_change_ok(inode, iattr);
	82	if (error)
	83	return error;
	84
40a2159a BH	85	iattr->ia_valid &= ~ATTR_SIZE; /* ignore size changes */
40a2159a BH	86
988d186d MS	87	error = inode_setattr(inode, iattr);
	88	if (error)
	89	return error;
	90
ddd29ec6	91	if (!sd_attrs) {
988d186d	92	/* setting attributes for the first time, allocate now */
ddd29ec6 DQ	93	sd_attrs = sysfs_init_inode_attrs(sd);
ddd29ec6 DQ	94	if (!sd_attrs)
988d186d	95	return -ENOMEM;
ddd29ec6 DQ	96	sd->s_iattr = sd_attrs;
	97	} else {
	98	/* attributes were changed at least once in past */
	99	iattrs = &sd_attrs->ia_iattr;
	100
	101	if (ia_valid & ATTR_UID)
	102	iattrs->ia_uid = iattr->ia_uid;
	103	if (ia_valid & ATTR_GID)
	104	iattrs->ia_gid = iattr->ia_gid;
	105	if (ia_valid & ATTR_ATIME)
	106	iattrs->ia_atime = timespec_trunc(iattr->ia_atime,
	107	inode->i_sb->s_time_gran);
	108	if (ia_valid & ATTR_MTIME)
	109	iattrs->ia_mtime = timespec_trunc(iattr->ia_mtime,
	110	inode->i_sb->s_time_gran);
	111	if (ia_valid & ATTR_CTIME)
	112	iattrs->ia_ctime = timespec_trunc(iattr->ia_ctime,
	113	inode->i_sb->s_time_gran);
	114	if (ia_valid & ATTR_MODE) {
	115	umode_t mode = iattr->ia_mode;
	116
	117	if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
	118	mode &= ~S_ISGID;
	119	iattrs->ia_mode = sd->s_mode = mode;
	120	}
988d186d	121	}
ddd29ec6 DQ	122	return error;
ddd29ec6 DQ	123	}
988d186d	124
ddd29ec6 DQ	125	int sysfs_setxattr(struct dentry dentry, const char name, const void *value,
	126	size_t size, int flags)
	127	{
	128	struct sysfs_dirent *sd = dentry->d_fsdata;
	129	struct sysfs_inode_attrs *iattrs;
	130	void *secdata;
	131	int error;
	132	u32 secdata_len = 0;
	133
	134	if (!sd)
	135	return -EINVAL;
	136	if (!sd->s_iattr)
	137	sd->s_iattr = sysfs_init_inode_attrs(sd);
	138	if (!sd->s_iattr)
	139	return -ENOMEM;
	140
	141	iattrs = sd->s_iattr;
	142
	143	if (!strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN)) {
	144	const char *suffix = name + XATTR_SECURITY_PREFIX_LEN;
	145	error = security_inode_setsecurity(dentry->d_inode, suffix,
	146	value, size, flags);
	147	if (error)
	148	goto out;
	149	error = security_inode_getsecctx(dentry->d_inode,
	150	&secdata, &secdata_len);
	151	if (error)
	152	goto out;
	153	if (iattrs->ia_secdata)
	154	security_release_secctx(iattrs->ia_secdata,
	155	iattrs->ia_secdata_len);
	156	iattrs->ia_secdata = secdata;
	157	iattrs->ia_secdata_len = secdata_len;
988d186d	158
ddd29ec6 DQ	159	} else
	160	return -EINVAL;
	161	out:
988d186d MS	162	return error;
	163	}
	164
8215534c MS	165	static inline void set_default_inode_attr(struct inode * inode, mode_t mode)
	166	{
	167	inode->i_mode = mode;
8215534c MS	168	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	169	}
	170
	171	static inline void set_inode_attr(struct inode * inode, struct iattr * iattr)
	172	{
	173	inode->i_mode = iattr->ia_mode;
	174	inode->i_uid = iattr->ia_uid;
	175	inode->i_gid = iattr->ia_gid;
	176	inode->i_atime = iattr->ia_atime;
	177	inode->i_mtime = iattr->ia_mtime;
	178	inode->i_ctime = iattr->ia_ctime;
	179	}
	180
232ba9db AV	181
	182	/*
	183	* sysfs has a different i_mutex lock order behavior for i_mutex than other
	184	* filesystems; sysfs i_mutex is called in many places with subsystem locks
	185	* held. At the same time, many of the VFS locking rules do not apply to
	186	* sysfs at all (cross directory rename for example). To untangle this mess
	187	* (which gives false positives in lockdep), we're giving sysfs inodes their
	188	* own class for i_mutex.
	189	*/
	190	static struct lock_class_key sysfs_inode_imutex_key;
	191
372e88bd EB	192	static int sysfs_count_nlink(struct sysfs_dirent *sd)
	193	{
	194	struct sysfs_dirent *child;
	195	int nr = 0;
	196
bc747f37	197	for (child = sd->s_dir.children; child; child = child->s_sibling)
372e88bd EB	198	if (sysfs_type(child) == SYSFS_DIR)
	199	nr++;
	200
	201	return nr + 2;
	202	}
	203
bc37e283	204	static void sysfs_init_inode(struct sysfs_dirent sd, struct inode inode)
1da177e4	205	{
372e88bd	206	struct bin_attribute *bin_attr;
ddd29ec6	207	struct sysfs_inode_attrs *iattrs;
372e88bd	208
04256b4a	209	inode->i_private = sysfs_get(sd);
fc9f54b9 TH	210	inode->i_mapping->a_ops = &sysfs_aops;
	211	inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info;
	212	inode->i_op = &sysfs_inode_operations;
	213	inode->i_ino = sd->s_ino;
	214	lockdep_set_class(&inode->i_mutex, &sysfs_inode_imutex_key);
	215
ddd29ec6 DQ	216	iattrs = sd->s_iattr;
ddd29ec6 DQ	217	if (iattrs) {
fc9f54b9 TH	218	/* sysfs_dirent has non-default attributes
	219	* get them for the new inode from persistent copy
	220	* in sysfs_dirent
	221	*/
ddd29ec6 DQ	222	set_inode_attr(inode, &iattrs->ia_iattr);
	223	if (iattrs->ia_secdata)
	224	security_inode_notifysecctx(inode,
	225	iattrs->ia_secdata,
	226	iattrs->ia_secdata_len);
fc9f54b9 TH	227	} else
fc9f54b9 TH	228	set_default_inode_attr(inode, sd->s_mode);
372e88bd	229
372e88bd EB	230	/* initialize inode according to type */
372e88bd EB	231	switch (sysfs_type(sd)) {
372e88bd EB	232	case SYSFS_DIR:
	233	inode->i_op = &sysfs_dir_inode_operations;
	234	inode->i_fop = &sysfs_dir_operations;
	235	inode->i_nlink = sysfs_count_nlink(sd);
	236	break;
	237	case SYSFS_KOBJ_ATTR:
	238	inode->i_size = PAGE_SIZE;
	239	inode->i_fop = &sysfs_file_operations;
	240	break;
	241	case SYSFS_KOBJ_BIN_ATTR:
b1fc3d61	242	bin_attr = sd->s_bin_attr.bin_attr;
372e88bd EB	243	inode->i_size = bin_attr->size;
	244	inode->i_fop = &bin_fops;
	245	break;
	246	case SYSFS_KOBJ_LINK:
	247	inode->i_op = &sysfs_symlink_inode_operations;
	248	break;
	249	default:
	250	BUG();
	251	}
	252
	253	unlock_new_inode(inode);
fc9f54b9 TH	254	}
	255
	256	/**
8312a8d7	257	* sysfs_get_inode - get inode for sysfs_dirent
fc9f54b9 TH	258	* @sd: sysfs_dirent to allocate inode for
fc9f54b9 TH	259	*
8312a8d7 TH	260	* Get inode for @sd. If such inode doesn't exist, a new inode
	261	* is allocated and basics are initialized. New inode is
	262	* returned locked.
fc9f54b9 TH	263	*
	264	* LOCKING:
	265	* Kernel thread context (may sleep).
	266	*
	267	* RETURNS:
	268	* Pointer to allocated inode on success, NULL on failure.
	269	*/
8312a8d7	270	struct inode * sysfs_get_inode(struct sysfs_dirent *sd)
fc9f54b9 TH	271	{
	272	struct inode *inode;
	273
8312a8d7 TH	274	inode = iget_locked(sysfs_sb, sd->s_ino);
8312a8d7 TH	275	if (inode && (inode->i_state & I_NEW))
fc9f54b9 TH	276	sysfs_init_inode(sd, inode);
fc9f54b9 TH	277
1da177e4 LT	278	return inode;
	279	}
	280
04256b4a EB	281	/*
	282	* The sysfs_dirent serves as both an inode and a directory entry for sysfs.
	283	* To prevent the sysfs inode numbers from being freed prematurely we take a
	284	* reference to sysfs_dirent from the sysfs inode. A
	285	* super_operations.delete_inode() implementation is needed to drop that
	286	* reference upon inode destruction.
	287	*/
	288	void sysfs_delete_inode(struct inode *inode)
	289	{
	290	struct sysfs_dirent *sd = inode->i_private;
	291
	292	truncate_inode_pages(&inode->i_data, 0);
	293	clear_inode(inode);
	294	sysfs_put(sd);
	295	}
	296
608e266a	297	int sysfs_hash_and_remove(struct sysfs_dirent dir_sd, const char name)
1da177e4	298	{
fb6896da	299	struct sysfs_addrm_cxt acxt;
41fc1c27	300	struct sysfs_dirent *sd;
641e6f30	301
608e266a	302	if (!dir_sd)
995982ca	303	return -ENOENT;
1da177e4	304
fb6896da	305	sysfs_addrm_start(&acxt, dir_sd);
608e266a	306
41fc1c27 TH	307	sd = sysfs_find_dirent(dir_sd, name);
	308	if (sd)
	309	sysfs_remove_one(&acxt, sd);
3007e997	310
990e53f8 TH	311	sysfs_addrm_finish(&acxt);
	312
	313	if (sd)
fb6896da	314	return 0;
990e53f8 TH	315	else
990e53f8 TH	316	return -ENOENT;
1da177e4	317	}