proc_sysctl.c 10.9 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11
/*
 * /proc/sys support
 */

#include <linux/sysctl.h>
#include <linux/proc_fs.h>
#include <linux/security.h>
#include "internal.h"

static struct dentry_operations proc_sys_dentry_operations;
static const struct file_operations proc_sys_file_operations;
12
static const struct inode_operations proc_sys_inode_operations;
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

static void proc_sys_refresh_inode(struct inode *inode, struct ctl_table *table)
{
	/* Refresh the cached information bits in the inode */
	if (table) {
		inode->i_uid = 0;
		inode->i_gid = 0;
		inode->i_mode = table->mode;
		if (table->proc_handler) {
			inode->i_mode |= S_IFREG;
			inode->i_nlink = 1;
		} else {
			inode->i_mode |= S_IFDIR;
			inode->i_nlink = 0;	/* It is too hard to figure out */
		}
	}
}

static struct inode *proc_sys_make_inode(struct inode *dir, struct ctl_table *table)
{
	struct inode *inode;
	struct proc_inode *dir_ei, *ei;
	int depth;

	inode = new_inode(dir->i_sb);
	if (!inode)
		goto out;

	/* A directory is always one deeper than it's parent */
	dir_ei = PROC_I(dir);
	depth = dir_ei->fd + 1;

	ei = PROC_I(inode);
	ei->fd = depth;
	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
	inode->i_op = &proc_sys_inode_operations;
	inode->i_fop = &proc_sys_file_operations;
50
	inode->i_flags |= S_PRIVATE; /* tell selinux to ignore this inode */
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173
	proc_sys_refresh_inode(inode, table);
out:
	return inode;
}

static struct dentry *proc_sys_ancestor(struct dentry *dentry, int depth)
{
	for (;;) {
		struct proc_inode *ei;

		ei = PROC_I(dentry->d_inode);
		if (ei->fd == depth)
			break; /* found */

		dentry = dentry->d_parent;
	}
	return dentry;
}

static struct ctl_table *proc_sys_lookup_table_one(struct ctl_table *table,
							struct qstr *name)
{
	int len;
	for ( ; table->ctl_name || table->procname; table++) {

		if (!table->procname)
			continue;

		len = strlen(table->procname);
		if (len != name->len)
			continue;

		if (memcmp(table->procname, name->name, len) != 0)
			continue;

		/* I have a match */
		return table;
	}
	return NULL;
}

static struct ctl_table *proc_sys_lookup_table(struct dentry *dentry,
						struct ctl_table *table)
{
	struct dentry *ancestor;
	struct proc_inode *ei;
	int depth, i;

	ei = PROC_I(dentry->d_inode);
	depth = ei->fd;

	if (depth == 0)
		return table;

	for (i = 1; table && (i <= depth); i++) {
		ancestor = proc_sys_ancestor(dentry, i);
		table = proc_sys_lookup_table_one(table, &ancestor->d_name);
		if (table)
			table = table->child;
	}
	return table;

}
static struct ctl_table *proc_sys_lookup_entry(struct dentry *dparent,
						struct qstr *name,
						struct ctl_table *table)
{
	table = proc_sys_lookup_table(dparent, table);
	if (table)
		table = proc_sys_lookup_table_one(table, name);
	return table;
}

static struct ctl_table *do_proc_sys_lookup(struct dentry *parent,
						struct qstr *name,
						struct ctl_table_header **ptr)
{
	struct ctl_table_header *head;
	struct ctl_table *table = NULL;

	for (head = sysctl_head_next(NULL); head;
			head = sysctl_head_next(head)) {
		table = proc_sys_lookup_entry(parent, name, head->ctl_table);
		if (table)
			break;
	}
	*ptr = head;
	return table;
}

static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry,
					struct nameidata *nd)
{
	struct ctl_table_header *head;
	struct inode *inode;
	struct dentry *err;
	struct ctl_table *table;

	err = ERR_PTR(-ENOENT);
	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
	if (!table)
		goto out;

	err = ERR_PTR(-ENOMEM);
	inode = proc_sys_make_inode(dir, table);
	if (!inode)
		goto out;

	err = NULL;
	dentry->d_op = &proc_sys_dentry_operations;
	d_add(dentry, inode);

out:
	sysctl_head_finish(head);
	return err;
}

static ssize_t proc_sys_read(struct file *filp, char __user *buf,
				size_t count, loff_t *ppos)
{
	struct dentry *dentry = filp->f_dentry;
	struct ctl_table_header *head;
	struct ctl_table *table;
174 175
	ssize_t error;
	size_t res;
176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212

	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
	/* Has the sysctl entry disappeared on us? */
	error = -ENOENT;
	if (!table)
		goto out;

	/* Has the sysctl entry been replaced by a directory? */
	error = -EISDIR;
	if (!table->proc_handler)
		goto out;

	/*
	 * At this point we know that the sysctl was not unregistered
	 * and won't be until we finish.
	 */
	error = -EPERM;
	if (sysctl_perm(table, MAY_READ))
		goto out;

	/* careful: calling conventions are nasty here */
	res = count;
	error = table->proc_handler(table, 0, filp, buf, &res, ppos);
	if (!error)
		error = res;
out:
	sysctl_head_finish(head);

	return error;
}

static ssize_t proc_sys_write(struct file *filp, const char __user *buf,
				size_t count, loff_t *ppos)
{
	struct dentry *dentry = filp->f_dentry;
	struct ctl_table_header *head;
	struct ctl_table *table;
213 214
	ssize_t error;
	size_t res;
215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433

	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
	/* Has the sysctl entry disappeared on us? */
	error = -ENOENT;
	if (!table)
		goto out;

	/* Has the sysctl entry been replaced by a directory? */
	error = -EISDIR;
	if (!table->proc_handler)
		goto out;

	/*
	 * At this point we know that the sysctl was not unregistered
	 * and won't be until we finish.
	 */
	error = -EPERM;
	if (sysctl_perm(table, MAY_WRITE))
		goto out;

	/* careful: calling conventions are nasty here */
	res = count;
	error = table->proc_handler(table, 1, filp, (char __user *)buf,
				    &res, ppos);
	if (!error)
		error = res;
out:
	sysctl_head_finish(head);

	return error;
}


static int proc_sys_fill_cache(struct file *filp, void *dirent,
				filldir_t filldir, struct ctl_table *table)
{
	struct ctl_table_header *head;
	struct ctl_table *child_table = NULL;
	struct dentry *child, *dir = filp->f_path.dentry;
	struct inode *inode;
	struct qstr qname;
	ino_t ino = 0;
	unsigned type = DT_UNKNOWN;
	int ret;

	qname.name = table->procname;
	qname.len  = strlen(table->procname);
	qname.hash = full_name_hash(qname.name, qname.len);

	/* Suppress duplicates.
	 * Only fill a directory entry if it is the value that
	 * an ordinary lookup of that name returns.  Hide all
	 * others.
	 *
	 * If we ever cache this translation in the dcache
	 * I should do a dcache lookup first.  But for now
	 * it is just simpler not to.
	 */
	ret = 0;
	child_table = do_proc_sys_lookup(dir, &qname, &head);
	sysctl_head_finish(head);
	if (child_table != table)
		return 0;

	child = d_lookup(dir, &qname);
	if (!child) {
		struct dentry *new;
		new = d_alloc(dir, &qname);
		if (new) {
			inode = proc_sys_make_inode(dir->d_inode, table);
			if (!inode)
				child = ERR_PTR(-ENOMEM);
			else {
				new->d_op = &proc_sys_dentry_operations;
				d_add(new, inode);
			}
			if (child)
				dput(new);
			else
				child = new;
		}
	}
	if (!child || IS_ERR(child) || !child->d_inode)
		goto end_instantiate;
	inode = child->d_inode;
	if (inode) {
		ino  = inode->i_ino;
		type = inode->i_mode >> 12;
	}
	dput(child);
end_instantiate:
	if (!ino)
		ino= find_inode_number(dir, &qname);
	if (!ino)
		ino = 1;
	return filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type);
}

static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir)
{
	struct dentry *dentry = filp->f_dentry;
	struct inode *inode = dentry->d_inode;
	struct ctl_table_header *head = NULL;
	struct ctl_table *table;
	unsigned long pos;
	int ret;

	ret = -ENOTDIR;
	if (!S_ISDIR(inode->i_mode))
		goto out;

	ret = 0;
	/* Avoid a switch here: arm builds fail with missing __cmpdi2 */
	if (filp->f_pos == 0) {
		if (filldir(dirent, ".", 1, filp->f_pos,
				inode->i_ino, DT_DIR) < 0)
			goto out;
		filp->f_pos++;
	}
	if (filp->f_pos == 1) {
		if (filldir(dirent, "..", 2, filp->f_pos,
				parent_ino(dentry), DT_DIR) < 0)
			goto out;
		filp->f_pos++;
	}
	pos = 2;

	/* - Find each instance of the directory
	 * - Read all entries in each instance
	 * - Before returning an entry to user space lookup the entry
	 *   by name and if I find a different entry don't return
	 *   this one because it means it is a buried dup.
	 * For sysctl this should only happen for directory entries.
	 */
	for (head = sysctl_head_next(NULL); head; head = sysctl_head_next(head)) {
		table = proc_sys_lookup_table(dentry, head->ctl_table);

		if (!table)
			continue;

		for (; table->ctl_name || table->procname; table++, pos++) {
			/* Can't do anything without a proc name */
			if (!table->procname)
				continue;

			if (pos < filp->f_pos)
				continue;

			if (proc_sys_fill_cache(filp, dirent, filldir, table) < 0)
				goto out;
			filp->f_pos = pos + 1;
		}
	}
	ret = 1;
out:
	sysctl_head_finish(head);
	return ret;
}

static int proc_sys_permission(struct inode *inode, int mask, struct nameidata *nd)
{
	/*
	 * sysctl entries that are not writeable,
	 * are _NOT_ writeable, capabilities or not.
	 */
	struct ctl_table_header *head;
	struct ctl_table *table;
	struct dentry *dentry;
	int mode;
	int depth;
	int error;

	head = NULL;
	depth = PROC_I(inode)->fd;

	/* First check the cached permissions, in case we don't have
	 * enough information to lookup the sysctl table entry.
	 */
	error = -EACCES;
	mode = inode->i_mode;

	if (current->euid == 0)
		mode >>= 6;
	else if (in_group_p(0))
		mode >>= 3;

	if ((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask)
		error = 0;

	/* If we can't get a sysctl table entry the permission
	 * checks on the cached mode will have to be enough.
	 */
	if (!nd || !depth)
		goto out;

	dentry = nd->dentry;
	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);

	/* If the entry does not exist deny permission */
	error = -EACCES;
	if (!table)
		goto out;

	/* Use the permissions on the sysctl table entry */
	error = sysctl_perm(table, mask);
out:
	sysctl_head_finish(head);
	return error;
}

static int proc_sys_setattr(struct dentry *dentry, struct iattr *attr)
{
	struct inode *inode = dentry->d_inode;
	int error;

	if (attr->ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID))
		return -EPERM;

	error = inode_change_ok(inode, attr);
434 435
	if (!error)
		error = inode_setattr(inode, attr);
436 437 438 439 440 441 442 443 444 445 446 447 448

	return error;
}

/* I'm lazy and don't distinguish between files and directories,
 * until access time.
 */
static const struct file_operations proc_sys_file_operations = {
	.read		= proc_sys_read,
	.write		= proc_sys_write,
	.readdir	= proc_sys_readdir,
};

449
static const struct inode_operations proc_sys_inode_operations = {
450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478
	.lookup		= proc_sys_lookup,
	.permission	= proc_sys_permission,
	.setattr	= proc_sys_setattr,
};

static int proc_sys_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	struct ctl_table_header *head;
	struct ctl_table *table;
	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
	proc_sys_refresh_inode(dentry->d_inode, table);
	sysctl_head_finish(head);
	return !!table;
}

static struct dentry_operations proc_sys_dentry_operations = {
	.d_revalidate	= proc_sys_revalidate,
};

static struct proc_dir_entry *proc_sys_root;

int proc_sys_init(void)
{
	proc_sys_root = proc_mkdir("sys", NULL);
	proc_sys_root->proc_iops = &proc_sys_inode_operations;
	proc_sys_root->proc_fops = &proc_sys_file_operations;
	proc_sys_root->nlink = 0;
	return 0;
}