device_cgroup.c 12 KB
Newer Older
1
/*
2
 * device_cgroup.c - device cgroup subsystem
3 4 5 6 7 8 9 10 11
 *
 * Copyright 2007 IBM Corp
 */

#include <linux/device_cgroup.h>
#include <linux/cgroup.h>
#include <linux/ctype.h>
#include <linux/list.h>
#include <linux/uaccess.h>
12
#include <linux/seq_file.h>
13
#include <linux/slab.h>
14
#include <linux/rcupdate.h>
15
#include <linux/mutex.h>
16 17 18 19 20 21 22 23 24 25

#define ACC_MKNOD 1
#define ACC_READ  2
#define ACC_WRITE 4
#define ACC_MASK (ACC_MKNOD | ACC_READ | ACC_WRITE)

#define DEV_BLOCK 1
#define DEV_CHAR  2
#define DEV_ALL   4  /* this represents all devices */

26 27
static DEFINE_MUTEX(devcgroup_mutex);

28 29
/*
 * whitelist locking rules:
30
 * hold devcgroup_mutex for update/read.
31
 * hold rcu_read_lock() for read.
32 33 34 35 36 37 38
 */

struct dev_whitelist_item {
	u32 major, minor;
	short type;
	short access;
	struct list_head list;
39
	struct rcu_head rcu;
40 41 42 43 44
};

struct dev_cgroup {
	struct cgroup_subsys_state css;
	struct list_head whitelist;
45
	bool deny_all;
46 47
};

48 49 50 51 52
static inline struct dev_cgroup *css_to_devcgroup(struct cgroup_subsys_state *s)
{
	return container_of(s, struct dev_cgroup, css);
}

53 54
static inline struct dev_cgroup *cgroup_to_devcgroup(struct cgroup *cgroup)
{
55
	return css_to_devcgroup(cgroup_subsys_state(cgroup, devices_subsys_id));
56 57
}

58 59 60 61 62
static inline struct dev_cgroup *task_devcgroup(struct task_struct *task)
{
	return css_to_devcgroup(task_subsys_state(task, devices_subsys_id));
}

63 64
struct cgroup_subsys devices_subsys;

65 66
static int devcgroup_can_attach(struct cgroup *new_cgrp,
				struct cgroup_taskset *set)
67
{
68
	struct task_struct *task = cgroup_taskset_first(set);
69

70 71
	if (current != task && !capable(CAP_SYS_ADMIN))
		return -EPERM;
72 73 74 75
	return 0;
}

/*
76
 * called under devcgroup_mutex
77 78 79 80 81 82
 */
static int dev_whitelist_copy(struct list_head *dest, struct list_head *orig)
{
	struct dev_whitelist_item *wh, *tmp, *new;

	list_for_each_entry(wh, orig, list) {
Li Zefan's avatar
Li Zefan committed
83
		new = kmemdup(wh, sizeof(*wh), GFP_KERNEL);
84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
		if (!new)
			goto free_and_exit;
		list_add_tail(&new->list, dest);
	}

	return 0;

free_and_exit:
	list_for_each_entry_safe(wh, tmp, dest, list) {
		list_del(&wh->list);
		kfree(wh);
	}
	return -ENOMEM;
}

/* Stupid prototype - don't bother combining existing entries */
/*
101
 * called under devcgroup_mutex
102 103 104 105
 */
static int dev_whitelist_add(struct dev_cgroup *dev_cgroup,
			struct dev_whitelist_item *wh)
{
106
	struct dev_whitelist_item *whcopy, *walk;
107

Li Zefan's avatar
Li Zefan committed
108
	whcopy = kmemdup(wh, sizeof(*wh), GFP_KERNEL);
109 110 111
	if (!whcopy)
		return -ENOMEM;

112 113 114 115 116 117 118 119 120 121 122 123 124 125
	list_for_each_entry(walk, &dev_cgroup->whitelist, list) {
		if (walk->type != wh->type)
			continue;
		if (walk->major != wh->major)
			continue;
		if (walk->minor != wh->minor)
			continue;

		walk->access |= wh->access;
		kfree(whcopy);
		whcopy = NULL;
	}

	if (whcopy != NULL)
126
		list_add_tail_rcu(&whcopy->list, &dev_cgroup->whitelist);
127 128 129 130
	return 0;
}

/*
131
 * called under devcgroup_mutex
132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150
 */
static void dev_whitelist_rm(struct dev_cgroup *dev_cgroup,
			struct dev_whitelist_item *wh)
{
	struct dev_whitelist_item *walk, *tmp;

	list_for_each_entry_safe(walk, tmp, &dev_cgroup->whitelist, list) {
		if (walk->type == DEV_ALL)
			goto remove;
		if (walk->type != wh->type)
			continue;
		if (walk->major != ~0 && walk->major != wh->major)
			continue;
		if (walk->minor != ~0 && walk->minor != wh->minor)
			continue;

remove:
		walk->access &= ~wh->access;
		if (!walk->access) {
151
			list_del_rcu(&walk->list);
152
			kfree_rcu(walk, rcu);
153 154 155 156
		}
	}
}

157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
/**
 * dev_whitelist_clean - frees all entries of the whitelist
 * @dev_cgroup: dev_cgroup with the whitelist to be cleaned
 *
 * called under devcgroup_mutex
 */
static void dev_whitelist_clean(struct dev_cgroup *dev_cgroup)
{
	struct dev_whitelist_item *wh, *tmp;

	list_for_each_entry_safe(wh, tmp, &dev_cgroup->whitelist, list) {
		list_del(&wh->list);
		kfree(wh);
	}
}

173 174 175
/*
 * called from kernel/cgroup.c with cgroup_lock() held.
 */
176
static struct cgroup_subsys_state *devcgroup_create(struct cgroup *cgroup)
177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196
{
	struct dev_cgroup *dev_cgroup, *parent_dev_cgroup;
	struct cgroup *parent_cgroup;
	int ret;

	dev_cgroup = kzalloc(sizeof(*dev_cgroup), GFP_KERNEL);
	if (!dev_cgroup)
		return ERR_PTR(-ENOMEM);
	INIT_LIST_HEAD(&dev_cgroup->whitelist);
	parent_cgroup = cgroup->parent;

	if (parent_cgroup == NULL) {
		struct dev_whitelist_item *wh;
		wh = kmalloc(sizeof(*wh), GFP_KERNEL);
		if (!wh) {
			kfree(dev_cgroup);
			return ERR_PTR(-ENOMEM);
		}
		wh->minor = wh->major = ~0;
		wh->type = DEV_ALL;
Li Zefan's avatar
Li Zefan committed
197
		wh->access = ACC_MASK;
198
		dev_cgroup->deny_all = false;
199 200 201
		list_add(&wh->list, &dev_cgroup->whitelist);
	} else {
		parent_dev_cgroup = cgroup_to_devcgroup(parent_cgroup);
202
		mutex_lock(&devcgroup_mutex);
203 204
		ret = dev_whitelist_copy(&dev_cgroup->whitelist,
				&parent_dev_cgroup->whitelist);
205
		dev_cgroup->deny_all = parent_dev_cgroup->deny_all;
206
		mutex_unlock(&devcgroup_mutex);
207 208 209 210 211 212 213 214 215
		if (ret) {
			kfree(dev_cgroup);
			return ERR_PTR(ret);
		}
	}

	return &dev_cgroup->css;
}

216
static void devcgroup_destroy(struct cgroup *cgroup)
217 218 219 220
{
	struct dev_cgroup *dev_cgroup;

	dev_cgroup = cgroup_to_devcgroup(cgroup);
221
	dev_whitelist_clean(dev_cgroup);
222 223 224 225 226
	kfree(dev_cgroup);
}

#define DEVCG_ALLOW 1
#define DEVCG_DENY 2
227 228
#define DEVCG_LIST 3

229
#define MAJMINLEN 13
230
#define ACCLEN 4
231 232 233 234

static void set_access(char *acc, short access)
{
	int idx = 0;
235
	memset(acc, 0, ACCLEN);
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254
	if (access & ACC_READ)
		acc[idx++] = 'r';
	if (access & ACC_WRITE)
		acc[idx++] = 'w';
	if (access & ACC_MKNOD)
		acc[idx++] = 'm';
}

static char type_to_char(short type)
{
	if (type == DEV_ALL)
		return 'a';
	if (type == DEV_CHAR)
		return 'c';
	if (type == DEV_BLOCK)
		return 'b';
	return 'X';
}

255
static void set_majmin(char *str, unsigned m)
256 257
{
	if (m == ~0)
Li Zefan's avatar
Li Zefan committed
258
		strcpy(str, "*");
259
	else
Li Zefan's avatar
Li Zefan committed
260
		sprintf(str, "%u", m);
261 262
}

263 264
static int devcgroup_seq_read(struct cgroup *cgroup, struct cftype *cft,
				struct seq_file *m)
265
{
266
	struct dev_cgroup *devcgroup = cgroup_to_devcgroup(cgroup);
267
	struct dev_whitelist_item *wh;
268
	char maj[MAJMINLEN], min[MAJMINLEN], acc[ACCLEN];
269

270 271
	rcu_read_lock();
	list_for_each_entry_rcu(wh, &devcgroup->whitelist, list) {
272
		set_access(acc, wh->access);
273 274 275 276
		set_majmin(maj, wh->major);
		set_majmin(min, wh->minor);
		seq_printf(m, "%c %s:%s %s\n", type_to_char(wh->type),
			   maj, min, acc);
277
	}
278
	rcu_read_unlock();
279

280
	return 0;
281 282 283 284 285 286 287
}

/*
 * may_access_whitelist:
 * does the access granted to dev_cgroup c contain the access
 * requested in whitelist item refwh.
 * return 1 if yes, 0 if no.
288
 * call with devcgroup_mutex held
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305
 */
static int may_access_whitelist(struct dev_cgroup *c,
				       struct dev_whitelist_item *refwh)
{
	struct dev_whitelist_item *whitem;

	list_for_each_entry(whitem, &c->whitelist, list) {
		if (whitem->type & DEV_ALL)
			return 1;
		if ((refwh->type & DEV_BLOCK) && !(whitem->type & DEV_BLOCK))
			continue;
		if ((refwh->type & DEV_CHAR) && !(whitem->type & DEV_CHAR))
			continue;
		if (whitem->major != ~0 && whitem->major != refwh->major)
			continue;
		if (whitem->minor != ~0 && whitem->minor != refwh->minor)
			continue;
306
		if (refwh->access & (~whitem->access))
307 308 309 310 311 312 313 314 315 316 317
			continue;
		return 1;
	}
	return 0;
}

/*
 * parent_has_perm:
 * when adding a new allow rule to a device whitelist, the rule
 * must be allowed in the parent device
 */
318
static int parent_has_perm(struct dev_cgroup *childcg,
319 320
				  struct dev_whitelist_item *wh)
{
321
	struct cgroup *pcg = childcg->css.cgroup->parent;
322 323 324 325 326
	struct dev_cgroup *parent;

	if (!pcg)
		return 1;
	parent = cgroup_to_devcgroup(pcg);
327
	return may_access_whitelist(parent, wh);
328 329 330 331 332 333 334 335 336 337 338 339 340 341 342
}

/*
 * Modify the whitelist using allow/deny rules.
 * CAP_SYS_ADMIN is needed for this.  It's at least separate from CAP_MKNOD
 * so we can give a container CAP_MKNOD to let it create devices but not
 * modify the whitelist.
 * It seems likely we'll want to add a CAP_CONTAINER capability to allow
 * us to also grant CAP_SYS_ADMIN to containers without giving away the
 * device whitelist controls, but for now we'll stick with CAP_SYS_ADMIN
 *
 * Taking rules away is always allowed (given CAP_SYS_ADMIN).  Granting
 * new access is only allowed if you're in the top-level cgroup, or your
 * parent cgroup has the access you're asking for.
 */
343 344
static int devcgroup_update_access(struct dev_cgroup *devcgroup,
				   int filetype, const char *buffer)
345
{
346
	const char *b;
Li Zefan's avatar
Li Zefan committed
347
	char *endp;
Li Zefan's avatar
Li Zefan committed
348
	int count;
349 350 351 352 353 354 355 356 357 358 359 360
	struct dev_whitelist_item wh;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	memset(&wh, 0, sizeof(wh));
	b = buffer;

	switch (*b) {
	case 'a':
		wh.type = DEV_ALL;
		wh.access = ACC_MASK;
361 362
		wh.major = ~0;
		wh.minor = ~0;
363 364 365 366 367 368 369 370
		goto handle;
	case 'b':
		wh.type = DEV_BLOCK;
		break;
	case 'c':
		wh.type = DEV_CHAR;
		break;
	default:
371
		return -EINVAL;
372 373
	}
	b++;
374 375
	if (!isspace(*b))
		return -EINVAL;
376 377 378 379 380
	b++;
	if (*b == '*') {
		wh.major = ~0;
		b++;
	} else if (isdigit(*b)) {
Li Zefan's avatar
Li Zefan committed
381 382
		wh.major = simple_strtoul(b, &endp, 10);
		b = endp;
383
	} else {
384
		return -EINVAL;
385
	}
386 387
	if (*b != ':')
		return -EINVAL;
388 389 390 391 392 393 394
	b++;

	/* read minor */
	if (*b == '*') {
		wh.minor = ~0;
		b++;
	} else if (isdigit(*b)) {
Li Zefan's avatar
Li Zefan committed
395 396
		wh.minor = simple_strtoul(b, &endp, 10);
		b = endp;
397
	} else {
398
		return -EINVAL;
399
	}
400 401
	if (!isspace(*b))
		return -EINVAL;
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417
	for (b++, count = 0; count < 3; count++, b++) {
		switch (*b) {
		case 'r':
			wh.access |= ACC_READ;
			break;
		case 'w':
			wh.access |= ACC_WRITE;
			break;
		case 'm':
			wh.access |= ACC_MKNOD;
			break;
		case '\n':
		case '\0':
			count = 3;
			break;
		default:
418
			return -EINVAL;
419 420 421 422 423 424
		}
	}

handle:
	switch (filetype) {
	case DEVCG_ALLOW:
425 426
		if (!parent_has_perm(devcgroup, &wh))
			return -EPERM;
427
		devcgroup->deny_all = false;
428
		return dev_whitelist_add(devcgroup, &wh);
429 430
	case DEVCG_DENY:
		dev_whitelist_rm(devcgroup, &wh);
431
		devcgroup->deny_all = true;
432 433
		break;
	default:
434
		return -EINVAL;
435
	}
436 437
	return 0;
}
438

439 440 441 442
static int devcgroup_access_write(struct cgroup *cgrp, struct cftype *cft,
				  const char *buffer)
{
	int retval;
443 444

	mutex_lock(&devcgroup_mutex);
445 446
	retval = devcgroup_update_access(cgroup_to_devcgroup(cgrp),
					 cft->private, buffer);
447
	mutex_unlock(&devcgroup_mutex);
448 449 450 451 452 453
	return retval;
}

static struct cftype dev_cgroup_files[] = {
	{
		.name = "allow",
454
		.write_string  = devcgroup_access_write,
455 456 457 458
		.private = DEVCG_ALLOW,
	},
	{
		.name = "deny",
459
		.write_string = devcgroup_access_write,
460 461
		.private = DEVCG_DENY,
	},
462 463 464 465 466
	{
		.name = "list",
		.read_seq_string = devcgroup_seq_read,
		.private = DEVCG_LIST,
	},
467
	{ }	/* terminate */
468 469 470 471 472 473
};

struct cgroup_subsys devices_subsys = {
	.name = "devices",
	.can_attach = devcgroup_can_attach,
	.create = devcgroup_create,
474
	.destroy = devcgroup_destroy,
475
	.subsys_id = devices_subsys_id,
476
	.base_cftypes = dev_cgroup_files,
477 478 479 480 481 482 483 484 485

	/*
	 * While devices cgroup has the rudimentary hierarchy support which
	 * checks the parent's restriction, it doesn't properly propagates
	 * config changes in ancestors to their descendents.  A child
	 * should only be allowed to add more restrictions to the parent's
	 * configuration.  Fix it and remove the following.
	 */
	.broken_hierarchy = true,
486 487
};

488
int __devcgroup_inode_permission(struct inode *inode, int mask)
489 490 491 492
{
	struct dev_cgroup *dev_cgroup;
	struct dev_whitelist_item *wh;

493
	rcu_read_lock();
494 495 496

	dev_cgroup = task_devcgroup(current);

497
	list_for_each_entry_rcu(wh, &dev_cgroup->whitelist, list) {
498
		if (wh->type & DEV_ALL)
499
			goto found;
500 501 502 503 504 505 506 507
		if ((wh->type & DEV_BLOCK) && !S_ISBLK(inode->i_mode))
			continue;
		if ((wh->type & DEV_CHAR) && !S_ISCHR(inode->i_mode))
			continue;
		if (wh->major != ~0 && wh->major != imajor(inode))
			continue;
		if (wh->minor != ~0 && wh->minor != iminor(inode))
			continue;
508

509 510 511 512
		if ((mask & MAY_WRITE) && !(wh->access & ACC_WRITE))
			continue;
		if ((mask & MAY_READ) && !(wh->access & ACC_READ))
			continue;
513
found:
514
		rcu_read_unlock();
515 516
		return 0;
	}
517

518
	rcu_read_unlock();
519 520 521 522 523 524 525 526 527

	return -EPERM;
}

int devcgroup_inode_mknod(int mode, dev_t dev)
{
	struct dev_cgroup *dev_cgroup;
	struct dev_whitelist_item *wh;

528 529 530
	if (!S_ISBLK(mode) && !S_ISCHR(mode))
		return 0;

531
	rcu_read_lock();
532 533 534

	dev_cgroup = task_devcgroup(current);

535
	list_for_each_entry_rcu(wh, &dev_cgroup->whitelist, list) {
536
		if (wh->type & DEV_ALL)
537
			goto found;
538 539 540 541 542 543 544 545
		if ((wh->type & DEV_BLOCK) && !S_ISBLK(mode))
			continue;
		if ((wh->type & DEV_CHAR) && !S_ISCHR(mode))
			continue;
		if (wh->major != ~0 && wh->major != MAJOR(dev))
			continue;
		if (wh->minor != ~0 && wh->minor != MINOR(dev))
			continue;
546

547 548
		if (!(wh->access & ACC_MKNOD))
			continue;
549
found:
550
		rcu_read_unlock();
551 552
		return 0;
	}
553

554
	rcu_read_unlock();
555

556 557
	return -EPERM;
}