inode.c 8.2 KB
Newer Older
L
Linus Torvalds 已提交
1
/*
T
Tejun Heo 已提交
2
 * fs/sysfs/inode.c - basic sysfs inode and dentry operations
L
Linus Torvalds 已提交
3
 *
T
Tejun Heo 已提交
4 5 6 7 8
 * 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.
L
Linus Torvalds 已提交
9 10 11 12 13 14 15 16 17
 *
 * Please see Documentation/filesystems/sysfs.txt for more information.
 */

#undef DEBUG 

#include <linux/pagemap.h>
#include <linux/namei.h>
#include <linux/backing-dev.h>
18
#include <linux/capability.h>
19
#include <linux/errno.h>
A
Alexey Dobriyan 已提交
20
#include <linux/sched.h>
21 22
#include <linux/xattr.h>
#include <linux/security.h>
L
Linus Torvalds 已提交
23 24 25 26
#include "sysfs.h"

extern struct super_block * sysfs_sb;

27
static const struct address_space_operations sysfs_aops = {
L
Linus Torvalds 已提交
28
	.readpage	= simple_readpage,
N
Nick Piggin 已提交
29 30
	.write_begin	= simple_write_begin,
	.write_end	= simple_write_end,
L
Linus Torvalds 已提交
31 32 33
};

static struct backing_dev_info sysfs_backing_dev_info = {
34
	.name		= "sysfs",
L
Linus Torvalds 已提交
35
	.ra_pages	= 0,	/* No readahead */
36
	.capabilities	= BDI_CAP_NO_ACCT_AND_WRITEBACK,
L
Linus Torvalds 已提交
37 38
};

39
static const struct inode_operations sysfs_inode_operations ={
40
	.setattr	= sysfs_setattr,
41
	.setxattr	= sysfs_setxattr,
42 43
};

P
Peter Zijlstra 已提交
44 45 46 47 48
int __init sysfs_inode_init(void)
{
	return bdi_init(&sysfs_backing_dev_info);
}

49
static struct sysfs_inode_attrs *sysfs_init_inode_attrs(struct sysfs_dirent *sd)
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
{
	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;
}
67

68 69 70 71
int sysfs_setattr(struct dentry * dentry, struct iattr * iattr)
{
	struct inode * inode = dentry->d_inode;
	struct sysfs_dirent * sd = dentry->d_fsdata;
72 73
	struct sysfs_inode_attrs *sd_attrs;
	struct iattr *iattrs;
74 75 76 77 78 79
	unsigned int ia_valid = iattr->ia_valid;
	int error;

	if (!sd)
		return -EINVAL;

80
	sd_attrs = sd->s_iattr;
81 82 83 84 85

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

86 87
	iattr->ia_valid &= ~ATTR_SIZE; /* ignore size changes */

88 89 90 91
	error = inode_setattr(inode, iattr);
	if (error)
		return error;

92
	if (!sd_attrs) {
93
		/* setting attributes for the first time, allocate now */
94 95
		sd_attrs = sysfs_init_inode_attrs(sd);
		if (!sd_attrs)
96
			return -ENOMEM;
97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121
		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;
		}
122
	}
123 124
	return error;
}
125

126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148
static int sysfs_sd_setsecdata(struct sysfs_dirent *sd, void **secdata, u32 *secdata_len)
{
	struct sysfs_inode_attrs *iattrs;
	void *old_secdata;
	size_t old_secdata_len;

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

	old_secdata = iattrs->ia_secdata;
	old_secdata_len = iattrs->ia_secdata_len;

	iattrs->ia_secdata = *secdata;
	iattrs->ia_secdata_len = *secdata_len;

	*secdata = old_secdata;
	*secdata_len = old_secdata_len;
	return 0;
}

149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
int sysfs_setxattr(struct dentry *dentry, const char *name, const void *value,
		size_t size, int flags)
{
	struct sysfs_dirent *sd = dentry->d_fsdata;
	void *secdata;
	int error;
	u32 secdata_len = 0;

	if (!sd)
		return -EINVAL;

	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;
170

171 172 173 174 175 176
		mutex_lock(&sysfs_mutex);
		error = sysfs_sd_setsecdata(sd, &secdata, &secdata_len);
		mutex_unlock(&sysfs_mutex);

		if (secdata)
			security_release_secctx(secdata, secdata_len);
177 178 179
	} else
		return -EINVAL;
out:
180 181 182
	return error;
}

183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
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;
}

199 200 201 202 203 204 205 206 207 208 209

/*
 * 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;

210 211 212 213 214
static int sysfs_count_nlink(struct sysfs_dirent *sd)
{
	struct sysfs_dirent *child;
	int nr = 0;

215
	for (child = sd->s_dir.children; child; child = child->s_sibling)
216 217 218 219 220 221
		if (sysfs_type(child) == SYSFS_DIR)
			nr++;

	return nr + 2;
}

222
static void sysfs_init_inode(struct sysfs_dirent *sd, struct inode *inode)
L
Linus Torvalds 已提交
223
{
224
	struct bin_attribute *bin_attr;
225
	struct sysfs_inode_attrs *iattrs;
226

227
	inode->i_private = sysfs_get(sd);
228 229 230 231 232 233
	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);

234 235
	iattrs = sd->s_iattr;
	if (iattrs) {
236 237 238 239
		/* sysfs_dirent has non-default attributes
		 * get them for the new inode from persistent copy
		 * in sysfs_dirent
		 */
240 241 242 243 244
		set_inode_attr(inode, &iattrs->ia_iattr);
		if (iattrs->ia_secdata)
			security_inode_notifysecctx(inode,
						iattrs->ia_secdata,
						iattrs->ia_secdata_len);
245 246
	} else
		set_default_inode_attr(inode, sd->s_mode);
247 248 249 250 251 252 253 254 255 256 257 258 259

	/* 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:
T
Tejun Heo 已提交
260
		bin_attr = sd->s_bin_attr.bin_attr;
261 262 263 264 265 266 267 268 269 270 271
		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);
272 273 274
}

/**
275
 *	sysfs_get_inode - get inode for sysfs_dirent
276 277
 *	@sd: sysfs_dirent to allocate inode for
 *
278 279 280
 *	Get inode for @sd.  If such inode doesn't exist, a new inode
 *	is allocated and basics are initialized.  New inode is
 *	returned locked.
281 282 283 284 285 286 287
 *
 *	LOCKING:
 *	Kernel thread context (may sleep).
 *
 *	RETURNS:
 *	Pointer to allocated inode on success, NULL on failure.
 */
288
struct inode * sysfs_get_inode(struct sysfs_dirent *sd)
289 290 291
{
	struct inode *inode;

292 293
	inode = iget_locked(sysfs_sb, sd->s_ino);
	if (inode && (inode->i_state & I_NEW))
294 295
		sysfs_init_inode(sd, inode);

L
Linus Torvalds 已提交
296 297 298
	return inode;
}

299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
/*
 * 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);
}

315
int sysfs_hash_and_remove(struct sysfs_dirent *dir_sd, const char *name)
L
Linus Torvalds 已提交
316
{
317
	struct sysfs_addrm_cxt acxt;
318
	struct sysfs_dirent *sd;
319

320
	if (!dir_sd)
321
		return -ENOENT;
L
Linus Torvalds 已提交
322

323
	sysfs_addrm_start(&acxt, dir_sd);
324

325 326 327
	sd = sysfs_find_dirent(dir_sd, name);
	if (sd)
		sysfs_remove_one(&acxt, sd);
328

329 330 331
	sysfs_addrm_finish(&acxt);

	if (sd)
332
		return 0;
333 334
	else
		return -ENOENT;
L
Linus Torvalds 已提交
335
}