cpuset.c 78.4 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6
/*
 *  kernel/cpuset.c
 *
 *  Processor and Memory placement constraints for sets of tasks.
 *
 *  Copyright (C) 2003 BULL SA.
7
 *  Copyright (C) 2004-2006 Silicon Graphics, Inc.
Linus Torvalds's avatar
Linus Torvalds committed
8 9 10 11
 *
 *  Portions derived from Patrick Mochel's sysfs code.
 *  sysfs is Copyright (c) 2001-3 Patrick Mochel
 *
12
 *  2003-10-10 Written by Simon Derr.
Linus Torvalds's avatar
Linus Torvalds committed
13
 *  2003-10-22 Updates by Stephen Hemminger.
14
 *  2004 May-July Rework by Paul Jackson.
Linus Torvalds's avatar
Linus Torvalds committed
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
 *
 *  This file is subject to the terms and conditions of the GNU General Public
 *  License.  See the file COPYING in the main directory of the Linux
 *  distribution for more details.
 */

#include <linux/cpu.h>
#include <linux/cpumask.h>
#include <linux/cpuset.h>
#include <linux/err.h>
#include <linux/errno.h>
#include <linux/file.h>
#include <linux/fs.h>
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/kernel.h>
#include <linux/kmod.h>
#include <linux/list.h>
33
#include <linux/mempolicy.h>
Linus Torvalds's avatar
Linus Torvalds committed
34 35 36 37 38 39
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/pagemap.h>
#include <linux/proc_fs.h>
40
#include <linux/rcupdate.h>
Linus Torvalds's avatar
Linus Torvalds committed
41 42
#include <linux/sched.h>
#include <linux/seq_file.h>
43
#include <linux/security.h>
Linus Torvalds's avatar
Linus Torvalds committed
44 45 46 47 48 49 50 51 52 53
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/stat.h>
#include <linux/string.h>
#include <linux/time.h>
#include <linux/backing-dev.h>
#include <linux/sort.h>

#include <asm/uaccess.h>
#include <asm/atomic.h>
54
#include <linux/mutex.h>
Linus Torvalds's avatar
Linus Torvalds committed
55

56
#define CPUSET_SUPER_MAGIC		0x27e0eb
Linus Torvalds's avatar
Linus Torvalds committed
57

58 59 60 61 62
/*
 * Tracks how many cpusets are currently defined in system.
 * When there is only one cpuset (the root cpuset) we can
 * short circuit some hooks.
 */
63
int number_of_cpusets __read_mostly;
64

65 66 67 68 69 70 71 72 73
/* See "Frequency meter" comments, below. */

struct fmeter {
	int cnt;		/* unprocessed events count */
	int val;		/* most recent output value */
	time_t time;		/* clock (secs) when val computed */
	spinlock_t lock;	/* guards read or write of above */
};

Linus Torvalds's avatar
Linus Torvalds committed
74 75 76 77 78
struct cpuset {
	unsigned long flags;		/* "unsigned long" so bitops work */
	cpumask_t cpus_allowed;		/* CPUs allowed to tasks in cpuset */
	nodemask_t mems_allowed;	/* Memory Nodes allowed to tasks */

79 80 81
	/*
	 * Count is atomic so can incr (fork) or decr (exit) without a lock.
	 */
Linus Torvalds's avatar
Linus Torvalds committed
82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97
	atomic_t count;			/* count tasks using this cpuset */

	/*
	 * We link our 'sibling' struct into our parents 'children'.
	 * Our children link their 'sibling' into our 'children'.
	 */
	struct list_head sibling;	/* my parents children */
	struct list_head children;	/* my children */

	struct cpuset *parent;		/* my parent */
	struct dentry *dentry;		/* cpuset fs entry */

	/*
	 * Copy of global cpuset_mems_generation as of the most
	 * recent time this cpuset changed its mems_allowed.
	 */
98 99 100
	int mems_generation;

	struct fmeter fmeter;		/* memory_pressure filter */
Linus Torvalds's avatar
Linus Torvalds committed
101 102 103 104 105 106
};

/* bits in struct cpuset flags field */
typedef enum {
	CS_CPU_EXCLUSIVE,
	CS_MEM_EXCLUSIVE,
107
	CS_MEMORY_MIGRATE,
Linus Torvalds's avatar
Linus Torvalds committed
108
	CS_REMOVED,
109 110 111
	CS_NOTIFY_ON_RELEASE,
	CS_SPREAD_PAGE,
	CS_SPREAD_SLAB,
Linus Torvalds's avatar
Linus Torvalds committed
112 113 114 115 116
} cpuset_flagbits_t;

/* convenient tests for these bits */
static inline int is_cpu_exclusive(const struct cpuset *cs)
{
117
	return test_bit(CS_CPU_EXCLUSIVE, &cs->flags);
Linus Torvalds's avatar
Linus Torvalds committed
118 119 120 121
}

static inline int is_mem_exclusive(const struct cpuset *cs)
{
122
	return test_bit(CS_MEM_EXCLUSIVE, &cs->flags);
Linus Torvalds's avatar
Linus Torvalds committed
123 124 125 126
}

static inline int is_removed(const struct cpuset *cs)
{
127
	return test_bit(CS_REMOVED, &cs->flags);
Linus Torvalds's avatar
Linus Torvalds committed
128 129 130 131
}

static inline int notify_on_release(const struct cpuset *cs)
{
132
	return test_bit(CS_NOTIFY_ON_RELEASE, &cs->flags);
Linus Torvalds's avatar
Linus Torvalds committed
133 134
}

135 136
static inline int is_memory_migrate(const struct cpuset *cs)
{
137
	return test_bit(CS_MEMORY_MIGRATE, &cs->flags);
138 139
}

140 141 142 143 144 145 146 147 148 149
static inline int is_spread_page(const struct cpuset *cs)
{
	return test_bit(CS_SPREAD_PAGE, &cs->flags);
}

static inline int is_spread_slab(const struct cpuset *cs)
{
	return test_bit(CS_SPREAD_SLAB, &cs->flags);
}

Linus Torvalds's avatar
Linus Torvalds committed
150
/*
151
 * Increment this integer everytime any cpuset changes its
Linus Torvalds's avatar
Linus Torvalds committed
152 153 154 155 156 157 158 159 160 161 162 163 164
 * mems_allowed value.  Users of cpusets can track this generation
 * number, and avoid having to lock and reload mems_allowed unless
 * the cpuset they're using changes generation.
 *
 * A single, global generation is needed because attach_task() could
 * reattach a task to a different cpuset, which must not have its
 * generation numbers aliased with those of that tasks previous cpuset.
 *
 * Generations are needed for mems_allowed because one task cannot
 * modify anothers memory placement.  So we must enable every task,
 * on every visit to __alloc_pages(), to efficiently check whether
 * its current->cpuset->mems_allowed has changed, requiring an update
 * of its current->mems_allowed.
165 166 167
 *
 * Since cpuset_mems_generation is guarded by manage_mutex,
 * there is no need to mark it atomic.
Linus Torvalds's avatar
Linus Torvalds committed
168
 */
169
static int cpuset_mems_generation;
Linus Torvalds's avatar
Linus Torvalds committed
170 171 172 173 174 175 176 177 178 179 180

static struct cpuset top_cpuset = {
	.flags = ((1 << CS_CPU_EXCLUSIVE) | (1 << CS_MEM_EXCLUSIVE)),
	.cpus_allowed = CPU_MASK_ALL,
	.mems_allowed = NODE_MASK_ALL,
	.count = ATOMIC_INIT(0),
	.sibling = LIST_HEAD_INIT(top_cpuset.sibling),
	.children = LIST_HEAD_INIT(top_cpuset.children),
};

static struct vfsmount *cpuset_mount;
181
static struct super_block *cpuset_sb;
Linus Torvalds's avatar
Linus Torvalds committed
182 183

/*
184 185
 * We have two global cpuset mutexes below.  They can nest.
 * It is ok to first take manage_mutex, then nest callback_mutex.  We also
186 187 188
 * require taking task_lock() when dereferencing a tasks cpuset pointer.
 * See "The task_lock() exception", at the end of this comment.
 *
189 190 191
 * A task must hold both mutexes to modify cpusets.  If a task
 * holds manage_mutex, then it blocks others wanting that mutex,
 * ensuring that it is the only task able to also acquire callback_mutex
192 193
 * and be able to modify cpusets.  It can perform various checks on
 * the cpuset structure first, knowing nothing will change.  It can
194
 * also allocate memory while just holding manage_mutex.  While it is
195
 * performing these checks, various callback routines can briefly
196 197
 * acquire callback_mutex to query cpusets.  Once it is ready to make
 * the changes, it takes callback_mutex, blocking everyone else.
198 199
 *
 * Calls to the kernel memory allocator can not be made while holding
200
 * callback_mutex, as that would risk double tripping on callback_mutex
201 202 203
 * from one of the callbacks into the cpuset code from within
 * __alloc_pages().
 *
204
 * If a task is only holding callback_mutex, then it has read-only
205 206 207 208 209 210
 * access to cpusets.
 *
 * The task_struct fields mems_allowed and mems_generation may only
 * be accessed in the context of that task, so require no locks.
 *
 * Any task can increment and decrement the count field without lock.
211
 * So in general, code holding manage_mutex or callback_mutex can't rely
212
 * on the count field not changing.  However, if the count goes to
213
 * zero, then only attach_task(), which holds both mutexes, can
214 215 216
 * increment it again.  Because a count of zero means that no tasks
 * are currently attached, therefore there is no way a task attached
 * to that cpuset can fork (the other way to increment the count).
217
 * So code holding manage_mutex or callback_mutex can safely assume that
218
 * if the count is zero, it will stay zero.  Similarly, if a task
219
 * holds manage_mutex or callback_mutex on a cpuset with zero count, it
220
 * knows that the cpuset won't be removed, as cpuset_rmdir() needs
221
 * both of those mutexes.
222 223
 *
 * The cpuset_common_file_write handler for operations that modify
224
 * the cpuset hierarchy holds manage_mutex across the entire operation,
225 226
 * single threading all such cpuset modifications across the system.
 *
227
 * The cpuset_common_file_read() handlers only hold callback_mutex across
228 229 230 231
 * small pieces of code, such as when reading out possibly multi-word
 * cpumasks and nodemasks.
 *
 * The fork and exit callbacks cpuset_fork() and cpuset_exit(), don't
232
 * (usually) take either mutex.  These are the two most performance
233
 * critical pieces of code here.  The exception occurs on cpuset_exit(),
234
 * when a task in a notify_on_release cpuset exits.  Then manage_mutex
235
 * is taken, and if the cpuset count is zero, a usermode call made
Linus Torvalds's avatar
Linus Torvalds committed
236 237 238
 * to /sbin/cpuset_release_agent with the name of the cpuset (path
 * relative to the root of cpuset file system) as the argument.
 *
239 240 241
 * A cpuset can only be deleted if both its 'count' of using tasks
 * is zero, and its list of 'children' cpusets is empty.  Since all
 * tasks in the system use _some_ cpuset, and since there is always at
242
 * least one task in the system (init), therefore, top_cpuset
243 244 245 246 247 248 249 250 251
 * always has either children cpusets and/or using tasks.  So we don't
 * need a special hack to ensure that top_cpuset cannot be deleted.
 *
 * The above "Tale of Two Semaphores" would be complete, but for:
 *
 *	The task_lock() exception
 *
 * The need for this exception arises from the action of attach_task(),
 * which overwrites one tasks cpuset pointer with another.  It does
252
 * so using both mutexes, however there are several performance
253
 * critical places that need to reference task->cpuset without the
254
 * expense of grabbing a system global mutex.  Therefore except as
255 256 257 258
 * noted below, when dereferencing or, as in attach_task(), modifying
 * a tasks cpuset pointer we use task_lock(), which acts on a spinlock
 * (task->alloc_lock) already in the task_struct routinely used for
 * such matters.
259 260 261 262 263
 *
 * P.S.  One more locking exception.  RCU is used to guard the
 * update of a tasks cpuset pointer by attach_task() and the
 * access of task->cpuset->mems_generation via that pointer in
 * the routine cpuset_update_task_memory_state().
Linus Torvalds's avatar
Linus Torvalds committed
264 265
 */

266 267
static DEFINE_MUTEX(manage_mutex);
static DEFINE_MUTEX(callback_mutex);
268

Linus Torvalds's avatar
Linus Torvalds committed
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
/*
 * A couple of forward declarations required, due to cyclic reference loop:
 *  cpuset_mkdir -> cpuset_create -> cpuset_populate_dir -> cpuset_add_file
 *  -> cpuset_create_file -> cpuset_dir_inode_operations -> cpuset_mkdir.
 */

static int cpuset_mkdir(struct inode *dir, struct dentry *dentry, int mode);
static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry);

static struct backing_dev_info cpuset_backing_dev_info = {
	.ra_pages = 0,		/* No readahead */
	.capabilities	= BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
};

static struct inode *cpuset_new_inode(mode_t mode)
{
	struct inode *inode = new_inode(cpuset_sb);

	if (inode) {
		inode->i_mode = mode;
		inode->i_uid = current->fsuid;
		inode->i_gid = current->fsgid;
		inode->i_blocks = 0;
		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
		inode->i_mapping->backing_dev_info = &cpuset_backing_dev_info;
	}
	return inode;
}

static void cpuset_diput(struct dentry *dentry, struct inode *inode)
{
	/* is dentry a directory ? if so, kfree() associated cpuset */
	if (S_ISDIR(inode->i_mode)) {
		struct cpuset *cs = dentry->d_fsdata;
		BUG_ON(!(is_removed(cs)));
		kfree(cs);
	}
	iput(inode);
}

static struct dentry_operations cpuset_dops = {
	.d_iput = cpuset_diput,
};

static struct dentry *cpuset_get_dentry(struct dentry *parent, const char *name)
{
315
	struct dentry *d = lookup_one_len(name, parent, strlen(name));
Linus Torvalds's avatar
Linus Torvalds committed
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
	if (!IS_ERR(d))
		d->d_op = &cpuset_dops;
	return d;
}

static void remove_dir(struct dentry *d)
{
	struct dentry *parent = dget(d->d_parent);

	d_delete(d);
	simple_rmdir(parent->d_inode, d);
	dput(parent);
}

/*
 * NOTE : the dentry must have been dget()'ed
 */
static void cpuset_d_remove_dir(struct dentry *dentry)
{
	struct list_head *node;

	spin_lock(&dcache_lock);
	node = dentry->d_subdirs.next;
	while (node != &dentry->d_subdirs) {
Eric Dumazet's avatar
Eric Dumazet committed
340
		struct dentry *d = list_entry(node, struct dentry, d_u.d_child);
Linus Torvalds's avatar
Linus Torvalds committed
341 342 343 344 345 346 347 348 349 350 351
		list_del_init(node);
		if (d->d_inode) {
			d = dget_locked(d);
			spin_unlock(&dcache_lock);
			d_delete(d);
			simple_unlink(dentry->d_inode, d);
			dput(d);
			spin_lock(&dcache_lock);
		}
		node = dentry->d_subdirs.next;
	}
Eric Dumazet's avatar
Eric Dumazet committed
352
	list_del_init(&dentry->d_u.d_child);
Linus Torvalds's avatar
Linus Torvalds committed
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
	spin_unlock(&dcache_lock);
	remove_dir(dentry);
}

static struct super_operations cpuset_ops = {
	.statfs = simple_statfs,
	.drop_inode = generic_delete_inode,
};

static int cpuset_fill_super(struct super_block *sb, void *unused_data,
							int unused_silent)
{
	struct inode *inode;
	struct dentry *root;

	sb->s_blocksize = PAGE_CACHE_SIZE;
	sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
	sb->s_magic = CPUSET_SUPER_MAGIC;
	sb->s_op = &cpuset_ops;
	cpuset_sb = sb;

	inode = cpuset_new_inode(S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR);
	if (inode) {
		inode->i_op = &simple_dir_inode_operations;
		inode->i_fop = &simple_dir_operations;
		/* directories start off with i_nlink == 2 (for "." entry) */
379
		inc_nlink(inode);
Linus Torvalds's avatar
Linus Torvalds committed
380 381 382 383 384 385 386 387 388 389 390 391 392
	} else {
		return -ENOMEM;
	}

	root = d_alloc_root(inode);
	if (!root) {
		iput(inode);
		return -ENOMEM;
	}
	sb->s_root = root;
	return 0;
}

393 394 395
static int cpuset_get_sb(struct file_system_type *fs_type,
			 int flags, const char *unused_dev_name,
			 void *data, struct vfsmount *mnt)
Linus Torvalds's avatar
Linus Torvalds committed
396
{
397
	return get_sb_single(fs_type, flags, data, cpuset_fill_super, mnt);
Linus Torvalds's avatar
Linus Torvalds committed
398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414
}

static struct file_system_type cpuset_fs_type = {
	.name = "cpuset",
	.get_sb = cpuset_get_sb,
	.kill_sb = kill_litter_super,
};

/* struct cftype:
 *
 * The files in the cpuset filesystem mostly have a very simple read/write
 * handling, some common function will take care of it. Nevertheless some cases
 * (read tasks) are special and therefore I define this structure for every
 * kind of file.
 *
 *
 * When reading/writing to a file:
415 416
 *	- the cpuset to use in file->f_path.dentry->d_parent->d_fsdata
 *	- the 'cftype' of the file is file->f_path.dentry->d_fsdata
Linus Torvalds's avatar
Linus Torvalds committed
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440
 */

struct cftype {
	char *name;
	int private;
	int (*open) (struct inode *inode, struct file *file);
	ssize_t (*read) (struct file *file, char __user *buf, size_t nbytes,
							loff_t *ppos);
	int (*write) (struct file *file, const char __user *buf, size_t nbytes,
							loff_t *ppos);
	int (*release) (struct inode *inode, struct file *file);
};

static inline struct cpuset *__d_cs(struct dentry *dentry)
{
	return dentry->d_fsdata;
}

static inline struct cftype *__d_cft(struct dentry *dentry)
{
	return dentry->d_fsdata;
}

/*
441
 * Call with manage_mutex held.  Writes path of cpuset into buf.
Linus Torvalds's avatar
Linus Torvalds committed
442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484
 * Returns 0 on success, -errno on error.
 */

static int cpuset_path(const struct cpuset *cs, char *buf, int buflen)
{
	char *start;

	start = buf + buflen;

	*--start = '\0';
	for (;;) {
		int len = cs->dentry->d_name.len;
		if ((start -= len) < buf)
			return -ENAMETOOLONG;
		memcpy(start, cs->dentry->d_name.name, len);
		cs = cs->parent;
		if (!cs)
			break;
		if (!cs->parent)
			continue;
		if (--start < buf)
			return -ENAMETOOLONG;
		*start = '/';
	}
	memmove(buf, start, buf + buflen - start);
	return 0;
}

/*
 * Notify userspace when a cpuset is released, by running
 * /sbin/cpuset_release_agent with the name of the cpuset (path
 * relative to the root of cpuset file system) as the argument.
 *
 * Most likely, this user command will try to rmdir this cpuset.
 *
 * This races with the possibility that some other task will be
 * attached to this cpuset before it is removed, or that some other
 * user task will 'mkdir' a child cpuset of this cpuset.  That's ok.
 * The presumed 'rmdir' will fail quietly if this cpuset is no longer
 * unused, and this cpuset will be reprieved from its death sentence,
 * to continue to serve a useful existence.  Next time it's released,
 * we will get notified again, if it still has 'notify_on_release' set.
 *
485 486 487 488 489 490 491 492
 * The final arg to call_usermodehelper() is 0, which means don't
 * wait.  The separate /sbin/cpuset_release_agent task is forked by
 * call_usermodehelper(), then control in this thread returns here,
 * without waiting for the release agent task.  We don't bother to
 * wait because the caller of this routine has no use for the exit
 * status of the /sbin/cpuset_release_agent task, so no sense holding
 * our caller up for that.
 *
493
 * When we had only one cpuset mutex, we had to call this
494 495
 * without holding it, to avoid deadlock when call_usermodehelper()
 * allocated memory.  With two locks, we could now call this while
496 497
 * holding manage_mutex, but we still don't, so as to minimize
 * the time manage_mutex is held.
Linus Torvalds's avatar
Linus Torvalds committed
498 499
 */

500
static void cpuset_release_agent(const char *pathbuf)
Linus Torvalds's avatar
Linus Torvalds committed
501 502 503 504
{
	char *argv[3], *envp[3];
	int i;

505 506 507
	if (!pathbuf)
		return;

Linus Torvalds's avatar
Linus Torvalds committed
508 509
	i = 0;
	argv[i++] = "/sbin/cpuset_release_agent";
510
	argv[i++] = (char *)pathbuf;
Linus Torvalds's avatar
Linus Torvalds committed
511 512 513 514 515 516 517 518
	argv[i] = NULL;

	i = 0;
	/* minimal command environment */
	envp[i++] = "HOME=/";
	envp[i++] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin";
	envp[i] = NULL;

519 520
	call_usermodehelper(argv[0], argv, envp, 0);
	kfree(pathbuf);
Linus Torvalds's avatar
Linus Torvalds committed
521 522 523 524 525 526
}

/*
 * Either cs->count of using tasks transitioned to zero, or the
 * cs->children list of child cpusets just became empty.  If this
 * cs is notify_on_release() and now both the user count is zero and
527 528
 * the list of children is empty, prepare cpuset path in a kmalloc'd
 * buffer, to be returned via ppathbuf, so that the caller can invoke
529 530
 * cpuset_release_agent() with it later on, once manage_mutex is dropped.
 * Call here with manage_mutex held.
531 532 533 534 535
 *
 * This check_for_release() routine is responsible for kmalloc'ing
 * pathbuf.  The above cpuset_release_agent() is responsible for
 * kfree'ing pathbuf.  The caller of these routines is responsible
 * for providing a pathbuf pointer, initialized to NULL, then
536 537
 * calling check_for_release() with manage_mutex held and the address
 * of the pathbuf pointer, then dropping manage_mutex, then calling
538
 * cpuset_release_agent() with pathbuf, as set by check_for_release().
Linus Torvalds's avatar
Linus Torvalds committed
539 540
 */

541
static void check_for_release(struct cpuset *cs, char **ppathbuf)
Linus Torvalds's avatar
Linus Torvalds committed
542 543 544 545 546 547 548 549 550
{
	if (notify_on_release(cs) && atomic_read(&cs->count) == 0 &&
	    list_empty(&cs->children)) {
		char *buf;

		buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
		if (!buf)
			return;
		if (cpuset_path(cs, buf, PAGE_SIZE) < 0)
551 552 553
			kfree(buf);
		else
			*ppathbuf = buf;
Linus Torvalds's avatar
Linus Torvalds committed
554 555 556 557 558 559 560 561 562 563 564 565 566 567
	}
}

/*
 * Return in *pmask the portion of a cpusets's cpus_allowed that
 * are online.  If none are online, walk up the cpuset hierarchy
 * until we find one that does have some online cpus.  If we get
 * all the way to the top and still haven't found any online cpus,
 * return cpu_online_map.  Or if passed a NULL cs from an exit'ing
 * task, return cpu_online_map.
 *
 * One way or another, we guarantee to return some non-empty subset
 * of cpu_online_map.
 *
568
 * Call with callback_mutex held.
Linus Torvalds's avatar
Linus Torvalds committed
569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591
 */

static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask)
{
	while (cs && !cpus_intersects(cs->cpus_allowed, cpu_online_map))
		cs = cs->parent;
	if (cs)
		cpus_and(*pmask, cs->cpus_allowed, cpu_online_map);
	else
		*pmask = cpu_online_map;
	BUG_ON(!cpus_intersects(*pmask, cpu_online_map));
}

/*
 * Return in *pmask the portion of a cpusets's mems_allowed that
 * are online.  If none are online, walk up the cpuset hierarchy
 * until we find one that does have some online mems.  If we get
 * all the way to the top and still haven't found any online mems,
 * return node_online_map.
 *
 * One way or another, we guarantee to return some non-empty subset
 * of node_online_map.
 *
592
 * Call with callback_mutex held.
Linus Torvalds's avatar
Linus Torvalds committed
593 594 595 596 597 598 599 600 601 602 603 604 605
 */

static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask)
{
	while (cs && !nodes_intersects(cs->mems_allowed, node_online_map))
		cs = cs->parent;
	if (cs)
		nodes_and(*pmask, cs->mems_allowed, node_online_map);
	else
		*pmask = node_online_map;
	BUG_ON(!nodes_intersects(*pmask, node_online_map));
}

606 607 608 609 610 611
/**
 * cpuset_update_task_memory_state - update task memory placement
 *
 * If the current tasks cpusets mems_allowed changed behind our
 * backs, update current->mems_allowed, mems_generation and task NUMA
 * mempolicy to the new value.
612
 *
613 614 615 616
 * Task mempolicy is updated by rebinding it relative to the
 * current->cpuset if a task has its memory placement changed.
 * Do not call this routine if in_interrupt().
 *
617 618 619 620
 * Call without callback_mutex or task_lock() held.  May be
 * called with or without manage_mutex held.  Thanks in part to
 * 'the_top_cpuset_hack', the tasks cpuset pointer will never
 * be NULL.  This routine also might acquire callback_mutex and
621
 * current->mm->mmap_sem during call.
622
 *
623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640
 * Reading current->cpuset->mems_generation doesn't need task_lock
 * to guard the current->cpuset derefence, because it is guarded
 * from concurrent freeing of current->cpuset by attach_task(),
 * using RCU.
 *
 * The rcu_dereference() is technically probably not needed,
 * as I don't actually mind if I see a new cpuset pointer but
 * an old value of mems_generation.  However this really only
 * matters on alpha systems using cpusets heavily.  If I dropped
 * that rcu_dereference(), it would save them a memory barrier.
 * For all other arch's, rcu_dereference is a no-op anyway, and for
 * alpha systems not using cpusets, another planned optimization,
 * avoiding the rcu critical section for tasks in the root cpuset
 * which is statically allocated, so can't vanish, will make this
 * irrelevant.  Better to use RCU as intended, than to engage in
 * some cute trick to save a memory barrier that is impossible to
 * test, for alpha systems using cpusets heavily, which might not
 * even exist.
641 642 643 644 645
 *
 * This routine is needed to update the per-task mems_allowed data,
 * within the tasks context, when it is trying to allocate memory
 * (in various mm/mempolicy.c routines) and notices that some other
 * task has been modifying its cpuset.
Linus Torvalds's avatar
Linus Torvalds committed
646 647
 */

648
void cpuset_update_task_memory_state(void)
Linus Torvalds's avatar
Linus Torvalds committed
649
{
650
	int my_cpusets_mem_gen;
651
	struct task_struct *tsk = current;
652
	struct cpuset *cs;
653

654 655 656 657 658 659 660 661 662
	if (tsk->cpuset == &top_cpuset) {
		/* Don't need rcu for top_cpuset.  It's never freed. */
		my_cpusets_mem_gen = top_cpuset.mems_generation;
	} else {
		rcu_read_lock();
		cs = rcu_dereference(tsk->cpuset);
		my_cpusets_mem_gen = cs->mems_generation;
		rcu_read_unlock();
	}
Linus Torvalds's avatar
Linus Torvalds committed
663

664
	if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) {
665
		mutex_lock(&callback_mutex);
666 667 668 669
		task_lock(tsk);
		cs = tsk->cpuset;	/* Maybe changed when task not locked */
		guarantee_online_mems(cs, &tsk->mems_allowed);
		tsk->cpuset_mems_generation = cs->mems_generation;
670 671 672 673 674 675 676 677
		if (is_spread_page(cs))
			tsk->flags |= PF_SPREAD_PAGE;
		else
			tsk->flags &= ~PF_SPREAD_PAGE;
		if (is_spread_slab(cs))
			tsk->flags |= PF_SPREAD_SLAB;
		else
			tsk->flags &= ~PF_SPREAD_SLAB;
678
		task_unlock(tsk);
679
		mutex_unlock(&callback_mutex);
680
		mpol_rebind_task(tsk, &tsk->mems_allowed);
Linus Torvalds's avatar
Linus Torvalds committed
681 682 683 684 685 686 687 688
	}
}

/*
 * is_cpuset_subset(p, q) - Is cpuset p a subset of cpuset q?
 *
 * One cpuset is a subset of another if all its allowed CPUs and
 * Memory Nodes are a subset of the other, and its exclusive flags
689
 * are only set if the other's are set.  Call holding manage_mutex.
Linus Torvalds's avatar
Linus Torvalds committed
690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706
 */

static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
{
	return	cpus_subset(p->cpus_allowed, q->cpus_allowed) &&
		nodes_subset(p->mems_allowed, q->mems_allowed) &&
		is_cpu_exclusive(p) <= is_cpu_exclusive(q) &&
		is_mem_exclusive(p) <= is_mem_exclusive(q);
}

/*
 * validate_change() - Used to validate that any proposed cpuset change
 *		       follows the structural rules for cpusets.
 *
 * If we replaced the flag and mask values of the current cpuset
 * (cur) with those values in the trial cpuset (trial), would
 * our various subset and exclusive rules still be valid?  Presumes
707
 * manage_mutex held.
Linus Torvalds's avatar
Linus Torvalds committed
708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730
 *
 * 'cur' is the address of an actual, in-use cpuset.  Operations
 * such as list traversal that depend on the actual address of the
 * cpuset in the list must use cur below, not trial.
 *
 * 'trial' is the address of bulk structure copy of cur, with
 * perhaps one or more of the fields cpus_allowed, mems_allowed,
 * or flags changed to new, trial values.
 *
 * Return 0 if valid, -errno if not.
 */

static int validate_change(const struct cpuset *cur, const struct cpuset *trial)
{
	struct cpuset *c, *par;

	/* Each of our child cpusets must be a subset of us */
	list_for_each_entry(c, &cur->children, sibling) {
		if (!is_cpuset_subset(c, trial))
			return -EBUSY;
	}

	/* Remaining checks don't apply to root cpuset */
731
	if (cur == &top_cpuset)
Linus Torvalds's avatar
Linus Torvalds committed
732 733
		return 0;

734 735
	par = cur->parent;

Linus Torvalds's avatar
Linus Torvalds committed
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754
	/* We must be a subset of our parent cpuset */
	if (!is_cpuset_subset(trial, par))
		return -EACCES;

	/* If either I or some sibling (!= me) is exclusive, we can't overlap */
	list_for_each_entry(c, &par->children, sibling) {
		if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) &&
		    c != cur &&
		    cpus_intersects(trial->cpus_allowed, c->cpus_allowed))
			return -EINVAL;
		if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) &&
		    c != cur &&
		    nodes_intersects(trial->mems_allowed, c->mems_allowed))
			return -EINVAL;
	}

	return 0;
}

755 756 757 758 759 760 761 762
/*
 * For a given cpuset cur, partition the system as follows
 * a. All cpus in the parent cpuset's cpus_allowed that are not part of any
 *    exclusive child cpusets
 * b. All cpus in the current cpuset's cpus_allowed that are not part of any
 *    exclusive child cpusets
 * Build these two partitions by calling partition_sched_domains
 *
763
 * Call with manage_mutex held.  May nest a call to the
764
 * lock_cpu_hotplug()/unlock_cpu_hotplug() pair.
765 766
 * Must not be called holding callback_mutex, because we must
 * not call lock_cpu_hotplug() while holding callback_mutex.
767
 */
768

769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785
static void update_cpu_domains(struct cpuset *cur)
{
	struct cpuset *c, *par = cur->parent;
	cpumask_t pspan, cspan;

	if (par == NULL || cpus_empty(cur->cpus_allowed))
		return;

	/*
	 * Get all cpus from parent's cpus_allowed not part of exclusive
	 * children
	 */
	pspan = par->cpus_allowed;
	list_for_each_entry(c, &par->children, sibling) {
		if (is_cpu_exclusive(c))
			cpus_andnot(pspan, pspan, c->cpus_allowed);
	}
786
	if (!is_cpu_exclusive(cur)) {
787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809
		cpus_or(pspan, pspan, cur->cpus_allowed);
		if (cpus_equal(pspan, cur->cpus_allowed))
			return;
		cspan = CPU_MASK_NONE;
	} else {
		if (cpus_empty(pspan))
			return;
		cspan = cur->cpus_allowed;
		/*
		 * Get all cpus from current cpuset's cpus_allowed not part
		 * of exclusive children
		 */
		list_for_each_entry(c, &cur->children, sibling) {
			if (is_cpu_exclusive(c))
				cpus_andnot(cspan, cspan, c->cpus_allowed);
		}
	}

	lock_cpu_hotplug();
	partition_sched_domains(&pspan, &cspan);
	unlock_cpu_hotplug();
}

810
/*
811
 * Call with manage_mutex held.  May take callback_mutex during call.
812 813
 */

Linus Torvalds's avatar
Linus Torvalds committed
814 815 816
static int update_cpumask(struct cpuset *cs, char *buf)
{
	struct cpuset trialcs;
817
	int retval, cpus_unchanged;
Linus Torvalds's avatar
Linus Torvalds committed
818

819 820 821 822
	/* top_cpuset.cpus_allowed tracks cpu_online_map; it's read-only */
	if (cs == &top_cpuset)
		return -EACCES;

Linus Torvalds's avatar
Linus Torvalds committed
823 824 825 826 827 828 829 830
	trialcs = *cs;
	retval = cpulist_parse(buf, trialcs.cpus_allowed);
	if (retval < 0)
		return retval;
	cpus_and(trialcs.cpus_allowed, trialcs.cpus_allowed, cpu_online_map);
	if (cpus_empty(trialcs.cpus_allowed))
		return -ENOSPC;
	retval = validate_change(cs, &trialcs);
831 832 833
	if (retval < 0)
		return retval;
	cpus_unchanged = cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed);
834
	mutex_lock(&callback_mutex);
835
	cs->cpus_allowed = trialcs.cpus_allowed;
836
	mutex_unlock(&callback_mutex);
837 838 839
	if (is_cpu_exclusive(cs) && !cpus_unchanged)
		update_cpu_domains(cs);
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
840 841
}

842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890
/*
 * cpuset_migrate_mm
 *
 *    Migrate memory region from one set of nodes to another.
 *
 *    Temporarilly set tasks mems_allowed to target nodes of migration,
 *    so that the migration code can allocate pages on these nodes.
 *
 *    Call holding manage_mutex, so our current->cpuset won't change
 *    during this call, as manage_mutex holds off any attach_task()
 *    calls.  Therefore we don't need to take task_lock around the
 *    call to guarantee_online_mems(), as we know no one is changing
 *    our tasks cpuset.
 *
 *    Hold callback_mutex around the two modifications of our tasks
 *    mems_allowed to synchronize with cpuset_mems_allowed().
 *
 *    While the mm_struct we are migrating is typically from some
 *    other task, the task_struct mems_allowed that we are hacking
 *    is for our current task, which must allocate new pages for that
 *    migrating memory region.
 *
 *    We call cpuset_update_task_memory_state() before hacking
 *    our tasks mems_allowed, so that we are assured of being in
 *    sync with our tasks cpuset, and in particular, callbacks to
 *    cpuset_update_task_memory_state() from nested page allocations
 *    won't see any mismatch of our cpuset and task mems_generation
 *    values, so won't overwrite our hacked tasks mems_allowed
 *    nodemask.
 */

static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from,
							const nodemask_t *to)
{
	struct task_struct *tsk = current;

	cpuset_update_task_memory_state();

	mutex_lock(&callback_mutex);
	tsk->mems_allowed = *to;
	mutex_unlock(&callback_mutex);

	do_migrate_pages(mm, from, to, MPOL_MF_MOVE_ALL);

	mutex_lock(&callback_mutex);
	guarantee_online_mems(tsk->cpuset, &tsk->mems_allowed);
	mutex_unlock(&callback_mutex);
}

891
/*
892 893 894
 * Handle user request to change the 'mems' memory placement
 * of a cpuset.  Needs to validate the request, update the
 * cpusets mems_allowed and mems_generation, and for each
895 896 897
 * task in the cpuset, rebind any vma mempolicies and if
 * the cpuset is marked 'memory_migrate', migrate the tasks
 * pages to the new memory.
898
 *
899
 * Call with manage_mutex held.  May take callback_mutex during call.
900 901 902
 * Will take tasklist_lock, scan tasklist for tasks in cpuset cs,
 * lock each such tasks mm->mmap_sem, scan its vma's and rebind
 * their mempolicies to the cpusets new mems_allowed.
903 904
 */

Linus Torvalds's avatar
Linus Torvalds committed
905 906 907
static int update_nodemask(struct cpuset *cs, char *buf)
{
	struct cpuset trialcs;
908
	nodemask_t oldmem;
909 910 911
	struct task_struct *g, *p;
	struct mm_struct **mmarray;
	int i, n, ntasks;
912
	int migrate;
913
	int fudge;
Linus Torvalds's avatar
Linus Torvalds committed
914 915
	int retval;

916 917 918 919
	/* top_cpuset.mems_allowed tracks node_online_map; it's read-only */
	if (cs == &top_cpuset)
		return -EACCES;

Linus Torvalds's avatar
Linus Torvalds committed
920 921 922
	trialcs = *cs;
	retval = nodelist_parse(buf, trialcs.mems_allowed);
	if (retval < 0)
923
		goto done;
Linus Torvalds's avatar
Linus Torvalds committed
924
	nodes_and(trialcs.mems_allowed, trialcs.mems_allowed, node_online_map);
925 926 927 928 929
	oldmem = cs->mems_allowed;
	if (nodes_equal(oldmem, trialcs.mems_allowed)) {
		retval = 0;		/* Too easy - nothing to do */
		goto done;
	}
930 931 932
	if (nodes_empty(trialcs.mems_allowed)) {
		retval = -ENOSPC;
		goto done;
Linus Torvalds's avatar
Linus Torvalds committed
933
	}
934 935 936 937
	retval = validate_change(cs, &trialcs);
	if (retval < 0)
		goto done;

938
	mutex_lock(&callback_mutex);
939
	cs->mems_allowed = trialcs.mems_allowed;
940
	cs->mems_generation = cpuset_mems_generation++;
941
	mutex_unlock(&callback_mutex);
942

943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996
	set_cpuset_being_rebound(cs);		/* causes mpol_copy() rebind */

	fudge = 10;				/* spare mmarray[] slots */
	fudge += cpus_weight(cs->cpus_allowed);	/* imagine one fork-bomb/cpu */
	retval = -ENOMEM;

	/*
	 * Allocate mmarray[] to hold mm reference for each task
	 * in cpuset cs.  Can't kmalloc GFP_KERNEL while holding
	 * tasklist_lock.  We could use GFP_ATOMIC, but with a
	 * few more lines of code, we can retry until we get a big
	 * enough mmarray[] w/o using GFP_ATOMIC.
	 */
	while (1) {
		ntasks = atomic_read(&cs->count);	/* guess */
		ntasks += fudge;
		mmarray = kmalloc(ntasks * sizeof(*mmarray), GFP_KERNEL);
		if (!mmarray)
			goto done;
		write_lock_irq(&tasklist_lock);		/* block fork */
		if (atomic_read(&cs->count) <= ntasks)
			break;				/* got enough */
		write_unlock_irq(&tasklist_lock);	/* try again */
		kfree(mmarray);
	}

	n = 0;

	/* Load up mmarray[] with mm reference for each task in cpuset. */
	do_each_thread(g, p) {
		struct mm_struct *mm;

		if (n >= ntasks) {
			printk(KERN_WARNING
				"Cpuset mempolicy rebind incomplete.\n");
			continue;
		}
		if (p->cpuset != cs)
			continue;
		mm = get_task_mm(p);
		if (!mm)
			continue;
		mmarray[n++] = mm;
	} while_each_thread(g, p);
	write_unlock_irq(&tasklist_lock);

	/*
	 * Now that we've dropped the tasklist spinlock, we can
	 * rebind the vma mempolicies of each mm in mmarray[] to their
	 * new cpuset, and release that mm.  The mpol_rebind_mm()
	 * call takes mmap_sem, which we couldn't take while holding
	 * tasklist_lock.  Forks can happen again now - the mpol_copy()
	 * cpuset_being_rebound check will catch such forks, and rebind
	 * their vma mempolicies too.  Because we still hold the global
997
	 * cpuset manage_mutex, we know that no other rebind effort will
998 999
	 * be contending for the global variable cpuset_being_rebound.
	 * It's ok if we rebind the same mm twice; mpol_rebind_mm()
1000
	 * is idempotent.  Also migrate pages in each mm to new nodes.
1001
	 */
1002
	migrate = is_memory_migrate(cs);
1003 1004 1005 1006
	for (i = 0; i < n; i++) {
		struct mm_struct *mm = mmarray[i];

		mpol_rebind_mm(mm, &cs->mems_allowed);
1007 1008
		if (migrate)
			cpuset_migrate_mm(mm, &oldmem, &cs->mems_allowed);
1009 1010 1011 1012 1013 1014 1015
		mmput(mm);
	}

	/* We're done rebinding vma's to this cpusets new mems_allowed. */
	kfree(mmarray);
	set_cpuset_being_rebound(NULL);
	retval = 0;
1016
done:
Linus Torvalds's avatar
Linus Torvalds committed
1017 1018 1019
	return retval;
}

1020
/*
1021
 * Call with manage_mutex held.
1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032
 */

static int update_memory_pressure_enabled(struct cpuset *cs, char *buf)
{
	if (simple_strtoul(buf, NULL, 10) != 0)
		cpuset_memory_pressure_enabled = 1;
	else
		cpuset_memory_pressure_enabled = 0;
	return 0;
}

Linus Torvalds's avatar
Linus Torvalds committed
1033 1034 1035
/*
 * update_flag - read a 0 or a 1 in a file and update associated flag
 * bit:	the bit to update (CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE,
1036 1037
 *				CS_NOTIFY_ON_RELEASE, CS_MEMORY_MIGRATE,
 *				CS_SPREAD_PAGE, CS_SPREAD_SLAB)
Linus Torvalds's avatar
Linus Torvalds committed
1038 1039
 * cs:	the cpuset to update
 * buf:	the buffer where we read the 0 or 1
1040
 *
1041
 * Call with manage_mutex held.
Linus Torvalds's avatar
Linus Torvalds committed
1042 1043 1044 1045 1046 1047
 */

static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf)
{
	int turning_on;
	struct cpuset trialcs;
1048
	int err, cpu_exclusive_changed;
Linus Torvalds's avatar
Linus Torvalds committed
1049 1050 1051 1052 1053 1054 1055 1056 1057 1058

	turning_on = (simple_strtoul(buf, NULL, 10) != 0);

	trialcs = *cs;
	if (turning_on)
		set_bit(bit, &trialcs.flags);
	else
		clear_bit(bit, &trialcs.flags);

	err = validate_change(cs, &trialcs);
1059 1060 1061 1062
	if (err < 0)
		return err;
	cpu_exclusive_changed =
		(is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs));
1063
	mutex_lock(&callback_mutex);
1064
	cs->flags = trialcs.flags;
1065
	mutex_unlock(&callback_mutex);
1066 1067 1068 1069

	if (cpu_exclusive_changed)
                update_cpu_domains(cs);
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
1070 1071
}

1072
/*
1073
 * Frequency meter - How fast is some event occurring?
1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169
 *
 * These routines manage a digitally filtered, constant time based,
 * event frequency meter.  There are four routines:
 *   fmeter_init() - initialize a frequency meter.
 *   fmeter_markevent() - called each time the event happens.
 *   fmeter_getrate() - returns the recent rate of such events.
 *   fmeter_update() - internal routine used to update fmeter.
 *
 * A common data structure is passed to each of these routines,
 * which is used to keep track of the state required to manage the
 * frequency meter and its digital filter.
 *
 * The filter works on the number of events marked per unit time.
 * The filter is single-pole low-pass recursive (IIR).  The time unit
 * is 1 second.  Arithmetic is done using 32-bit integers scaled to
 * simulate 3 decimal digits of precision (multiplied by 1000).
 *
 * With an FM_COEF of 933, and a time base of 1 second, the filter
 * has a half-life of 10 seconds, meaning that if the events quit
 * happening, then the rate returned from the fmeter_getrate()
 * will be cut in half each 10 seconds, until it converges to zero.
 *
 * It is not worth doing a real infinitely recursive filter.  If more
 * than FM_MAXTICKS ticks have elapsed since the last filter event,
 * just compute FM_MAXTICKS ticks worth, by which point the level
 * will be stable.
 *
 * Limit the count of unprocessed events to FM_MAXCNT, so as to avoid
 * arithmetic overflow in the fmeter_update() routine.
 *
 * Given the simple 32 bit integer arithmetic used, this meter works
 * best for reporting rates between one per millisecond (msec) and
 * one per 32 (approx) seconds.  At constant rates faster than one
 * per msec it maxes out at values just under 1,000,000.  At constant
 * rates between one per msec, and one per second it will stabilize
 * to a value N*1000, where N is the rate of events per second.
 * At constant rates between one per second and one per 32 seconds,
 * it will be choppy, moving up on the seconds that have an event,
 * and then decaying until the next event.  At rates slower than
 * about one in 32 seconds, it decays all the way back to zero between
 * each event.
 */

#define FM_COEF 933		/* coefficient for half-life of 10 secs */
#define FM_MAXTICKS ((time_t)99) /* useless computing more ticks than this */
#define FM_MAXCNT 1000000	/* limit cnt to avoid overflow */
#define FM_SCALE 1000		/* faux fixed point scale */

/* Initialize a frequency meter */
static void fmeter_init(struct fmeter *fmp)
{
	fmp->cnt = 0;
	fmp->val = 0;
	fmp->time = 0;
	spin_lock_init(&fmp->lock);
}

/* Internal meter update - process cnt events and update value */
static void fmeter_update(struct fmeter *fmp)
{
	time_t now = get_seconds();
	time_t ticks = now - fmp->time;

	if (ticks == 0)
		return;

	ticks = min(FM_MAXTICKS, ticks);
	while (ticks-- > 0)
		fmp->val = (FM_COEF * fmp->val) / FM_SCALE;
	fmp->time = now;

	fmp->val += ((FM_SCALE - FM_COEF) * fmp->cnt) / FM_SCALE;
	fmp->cnt = 0;
}

/* Process any previous ticks, then bump cnt by one (times scale). */
static void fmeter_markevent(struct fmeter *fmp)
{
	spin_lock(&fmp->lock);
	fmeter_update(fmp);
	fmp->cnt = min(FM_MAXCNT, fmp->cnt + FM_SCALE);
	spin_unlock(&fmp->lock);
}

/* Process any previous ticks, then return current value. */
static int fmeter_getrate(struct fmeter *fmp)
{
	int val;

	spin_lock(&fmp->lock);
	fmeter_update(fmp);
	val = fmp->val;
	spin_unlock(&fmp->lock);
	return val;
}

1170 1171 1172 1173 1174
/*
 * Attack task specified by pid in 'pidbuf' to cpuset 'cs', possibly
 * writing the path of the old cpuset in 'ppathbuf' if it needs to be
 * notified on release.
 *
1175
 * Call holding manage_mutex.  May take callback_mutex and task_lock of
1176 1177 1178
 * the task 'pid' during call.
 */

1179
static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf)
Linus Torvalds's avatar
Linus Torvalds committed
1180 1181 1182 1183 1184
{
	pid_t pid;
	struct task_struct *tsk;
	struct cpuset *oldcs;
	cpumask_t cpus;
1185
	nodemask_t from, to;
1186
	struct mm_struct *mm;
1187
	int retval;
Linus Torvalds's avatar
Linus Torvalds committed
1188

1189
	if (sscanf(pidbuf, "%d", &pid) != 1)
Linus Torvalds's avatar
Linus Torvalds committed
1190 1191 1192 1193 1194 1195 1196 1197
		return -EIO;
	if (cpus_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))
		return -ENOSPC;

	if (pid) {
		read_lock(&tasklist_lock);

		tsk = find_task_by_pid(pid);
1198
		if (!tsk || tsk->flags & PF_EXITING) {
Linus Torvalds's avatar
Linus Torvalds committed
1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215
			read_unlock(&tasklist_lock);
			return -ESRCH;
		}

		get_task_struct(tsk);
		read_unlock(&tasklist_lock);

		if ((current->euid) && (current->euid != tsk->uid)
		    && (current->euid != tsk->suid)) {
			put_task_struct(tsk);
			return -EACCES;
		}
	} else {
		tsk = current;
		get_task_struct(tsk);
	}

1216 1217 1218 1219 1220 1221
	retval = security_task_setscheduler(tsk, 0, NULL);
	if (retval) {
		put_task_struct(tsk);
		return retval;
	}

1222
	mutex_lock(&callback_mutex);
1223

Linus Torvalds's avatar
Linus Torvalds committed
1224 1225
	task_lock(tsk);
	oldcs = tsk->cpuset;
1226 1227 1228 1229 1230 1231
	/*
	 * After getting 'oldcs' cpuset ptr, be sure still not exiting.
	 * If 'oldcs' might be the top_cpuset due to the_top_cpuset_hack
	 * then fail this attach_task(), to avoid breaking top_cpuset.count.
	 */
	if (tsk->flags & PF_EXITING) {
Linus Torvalds's avatar
Linus Torvalds committed
1232
		task_unlock(tsk);
1233
		mutex_unlock(&callback_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
1234 1235 1236 1237
		put_task_struct(tsk);
		return -ESRCH;
	}
	atomic_inc(&cs->count);
1238
	rcu_assign_pointer(tsk->cpuset, cs);
Linus Torvalds's avatar
Linus Torvalds committed
1239 1240 1241 1242 1243
	task_unlock(tsk);

	guarantee_online_cpus(cs, &cpus);
	set_cpus_allowed(tsk, cpus);

1244 1245 1246
	from = oldcs->mems_allowed;
	to = cs->mems_allowed;

1247
	mutex_unlock(&callback_mutex);
1248 1249 1250 1251

	mm = get_task_mm(tsk);
	if (mm) {
		mpol_rebind_mm(mm, &to);
1252
		if (is_memory_migrate(cs))
1253
			cpuset_migrate_mm(mm, &from, &to);
1254 1255 1256
		mmput(mm);
	}

Linus Torvalds's avatar
Linus Torvalds committed
1257
	put_task_struct(tsk);
1258
	synchronize_rcu();
Linus Torvalds's avatar
Linus Torvalds committed
1259
	if (atomic_dec_and_test(&oldcs->count))
1260
		check_for_release(oldcs, ppathbuf);
Linus Torvalds's avatar
Linus Torvalds committed
1261 1262 1263 1264 1265 1266 1267 1268
	return 0;
}

/* The various types of files and directories in a cpuset file system */

typedef enum {
	FILE_ROOT,
	FILE_DIR,
1269
	FILE_MEMORY_MIGRATE,
Linus Torvalds's avatar
Linus Torvalds committed
1270 1271 1272 1273 1274
	FILE_CPULIST,
	FILE_MEMLIST,
	FILE_CPU_EXCLUSIVE,
	FILE_MEM_EXCLUSIVE,
	FILE_NOTIFY_ON_RELEASE,
1275 1276
	FILE_MEMORY_PRESSURE_ENABLED,
	FILE_MEMORY_PRESSURE,
1277 1278
	FILE_SPREAD_PAGE,
	FILE_SPREAD_SLAB,
Linus Torvalds's avatar
Linus Torvalds committed
1279 1280 1281
	FILE_TASKLIST,
} cpuset_filetype_t;

1282 1283
static ssize_t cpuset_common_file_write(struct file *file,
					const char __user *userbuf,
Linus Torvalds's avatar
Linus Torvalds committed
1284 1285
					size_t nbytes, loff_t *unused_ppos)
{
1286 1287
	struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent);
	struct cftype *cft = __d_cft(file->f_path.dentry);
Linus Torvalds's avatar
Linus Torvalds committed
1288 1289
	cpuset_filetype_t type = cft->private;
	char *buffer;
1290
	char *pathbuf = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
1291 1292 1293
	int retval = 0;

	/* Crude upper limit on largest legitimate cpulist user might write. */
1294
	if (nbytes > 100 + 6 * max(NR_CPUS, MAX_NUMNODES))
Linus Torvalds's avatar
Linus Torvalds committed
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
		return -E2BIG;

	/* +1 for nul-terminator */
	if ((buffer = kmalloc(nbytes + 1, GFP_KERNEL)) == 0)
		return -ENOMEM;

	if (copy_from_user(buffer, userbuf, nbytes)) {
		retval = -EFAULT;
		goto out1;
	}
	buffer[nbytes] = 0;	/* nul-terminate */

1307
	mutex_lock(&manage_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329

	if (is_removed(cs)) {
		retval = -ENODEV;
		goto out2;
	}

	switch (type) {
	case FILE_CPULIST:
		retval = update_cpumask(cs, buffer);
		break;
	case FILE_MEMLIST:
		retval = update_nodemask(cs, buffer);
		break;
	case FILE_CPU_EXCLUSIVE:
		retval = update_flag(CS_CPU_EXCLUSIVE, cs, buffer);
		break;
	case FILE_MEM_EXCLUSIVE:
		retval = update_flag(CS_MEM_EXCLUSIVE, cs, buffer);
		break;
	case FILE_NOTIFY_ON_RELEASE:
		retval = update_flag(CS_NOTIFY_ON_RELEASE, cs, buffer);
		break;
1330 1331 1332
	case FILE_MEMORY_MIGRATE:
		retval = update_flag(CS_MEMORY_MIGRATE, cs, buffer);
		break;
1333 1334 1335 1336 1337 1338
	case FILE_MEMORY_PRESSURE_ENABLED:
		retval = update_memory_pressure_enabled(cs, buffer);
		break;
	case FILE_MEMORY_PRESSURE:
		retval = -EACCES;
		break;
1339 1340
	case FILE_SPREAD_PAGE:
		retval = update_flag(CS_SPREAD_PAGE, cs, buffer);
1341
		cs->mems_generation = cpuset_mems_generation++;
1342 1343 1344
		break;
	case FILE_SPREAD_SLAB:
		retval = update_flag(CS_SPREAD_SLAB, cs, buffer);
1345
		cs->mems_generation = cpuset_mems_generation++;
1346
		break;
Linus Torvalds's avatar
Linus Torvalds committed
1347
	case FILE_TASKLIST:
1348
		retval = attach_task(cs, buffer, &pathbuf);
Linus Torvalds's avatar
Linus Torvalds committed
1349 1350 1351 1352 1353 1354 1355 1356 1357
		break;
	default:
		retval = -EINVAL;
		goto out2;
	}

	if (retval == 0)
		retval = nbytes;
out2:
1358
	mutex_unlock(&manage_mutex);
1359
	cpuset_release_agent(pathbuf);
Linus Torvalds's avatar
Linus Torvalds committed
1360 1361 1362 1363 1364 1365 1366 1367 1368
out1:
	kfree(buffer);
	return retval;
}

static ssize_t cpuset_file_write(struct file *file, const char __user *buf,
						size_t nbytes, loff_t *ppos)
{
	ssize_t retval = 0;
1369
	struct cftype *cft = __d_cft(file->f_path.dentry);
Linus Torvalds's avatar
Linus Torvalds committed
1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397
	if (!cft)
		return -ENODEV;

	/* special function ? */
	if (cft->write)
		retval = cft->write(file, buf, nbytes, ppos);
	else
		retval = cpuset_common_file_write(file, buf, nbytes, ppos);

	return retval;
}

/*
 * These ascii lists should be read in a single call, by using a user
 * buffer large enough to hold the entire map.  If read in smaller
 * chunks, there is no guarantee of atomicity.  Since the display format
 * used, list of ranges of sequential numbers, is variable length,
 * and since these maps can change value dynamically, one could read
 * gibberish by doing partial reads while a list was changing.
 * A single large read to a buffer that crosses a page boundary is
 * ok, because the result being copied to user land is not recomputed
 * across a page fault.
 */

static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs)
{
	cpumask_t mask;

1398
	mutex_lock(&callback_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
1399
	mask = cs->cpus_allowed;
1400
	mutex_unlock(&callback_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
1401 1402 1403 1404 1405 1406 1407 1408

	return cpulist_scnprintf(page, PAGE_SIZE, mask);
}

static int cpuset_sprintf_memlist(char *page, struct cpuset *cs)
{
	nodemask_t mask;

1409
	mutex_lock(&callback_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
1410
	mask = cs->mems_allowed;
1411
	mutex_unlock(&callback_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
1412 1413 1414 1415 1416 1417 1418

	return nodelist_scnprintf(page, PAGE_SIZE, mask);
}

static ssize_t cpuset_common_file_read(struct file *file, char __user *buf,
				size_t nbytes, loff_t *ppos)
{
1419 1420
	struct cftype *cft = __d_cft(file->f_path.dentry);
	struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent);
Linus Torvalds's avatar
Linus Torvalds committed
1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446
	cpuset_filetype_t type = cft->private;
	char *page;
	ssize_t retval = 0;
	char *s;

	if (!(page = (char *)__get_free_page(GFP_KERNEL)))
		return -ENOMEM;

	s = page;

	switch (type) {
	case FILE_CPULIST:
		s += cpuset_sprintf_cpulist(s, cs);
		break;
	case FILE_MEMLIST:
		s += cpuset_sprintf_memlist(s, cs);
		break;
	case FILE_CPU_EXCLUSIVE:
		*s++ = is_cpu_exclusive(cs) ? '1' : '0';
		break;
	case FILE_MEM_EXCLUSIVE:
		*s++ = is_mem_exclusive(cs) ? '1' : '0';
		break;
	case FILE_NOTIFY_ON_RELEASE:
		*s++ = notify_on_release(cs) ? '1' : '0';
		break;
1447 1448 1449
	case FILE_MEMORY_MIGRATE:
		*s++ = is_memory_migrate(cs) ? '1' : '0';
		break;
1450 1451 1452 1453 1454 1455
	case FILE_MEMORY_PRESSURE_ENABLED:
		*s++ = cpuset_memory_pressure_enabled ? '1' : '0';
		break;
	case FILE_MEMORY_PRESSURE:
		s += sprintf(s, "%d", fmeter_getrate(&cs->fmeter));
		break;
1456 1457 1458 1459 1460 1461
	case FILE_SPREAD_PAGE:
		*s++ = is_spread_page(cs) ? '1' : '0';
		break;
	case FILE_SPREAD_SLAB:
		*s++ = is_spread_slab(cs) ? '1' : '0';
		break;
Linus Torvalds's avatar
Linus Torvalds committed
1462 1463 1464 1465