cgroup.h 13.4 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14
#ifndef _LINUX_CGROUP_H
#define _LINUX_CGROUP_H
/*
 *  cgroup interface
 *
 *  Copyright (C) 2003 BULL SA
 *  Copyright (C) 2004-2006 Silicon Graphics, Inc.
 *
 */

#include <linux/sched.h>
#include <linux/cpumask.h>
#include <linux/nodemask.h>
#include <linux/rcupdate.h>
B
Balbir Singh 已提交
15
#include <linux/cgroupstats.h>
16
#include <linux/prio_heap.h>
17
#include <linux/rwsem.h>
18 19 20 21 22 23

#ifdef CONFIG_CGROUPS

struct cgroupfs_root;
struct cgroup_subsys;
struct inode;
24
struct cgroup;
25 26 27 28

extern int cgroup_init_early(void);
extern int cgroup_init(void);
extern void cgroup_lock(void);
29
extern bool cgroup_lock_live_group(struct cgroup *cgrp);
30
extern void cgroup_unlock(void);
31 32
extern void cgroup_fork(struct task_struct *p);
extern void cgroup_fork_callbacks(struct task_struct *p);
33
extern void cgroup_post_fork(struct task_struct *p);
34
extern void cgroup_exit(struct task_struct *p, int run_callbacks);
B
Balbir Singh 已提交
35 36
extern int cgroupstats_build(struct cgroupstats *stats,
				struct dentry *dentry);
37

38 39
extern struct file_operations proc_cgroup_operations;

40 41 42 43 44 45 46 47
/* Define the enumeration of all cgroup subsystems */
#define SUBSYS(_x) _x ## _subsys_id,
enum cgroup_subsys_id {
#include <linux/cgroup_subsys.h>
	CGROUP_SUBSYS_COUNT
};
#undef SUBSYS

48 49 50 51 52 53 54
/* Per-subsystem/per-cgroup state maintained by the system. */
struct cgroup_subsys_state {
	/* The cgroup that this subsystem is attached to. Useful
	 * for subsystems that want to know about the cgroup
	 * hierarchy structure */
	struct cgroup *cgroup;

P
Paul Menage 已提交
55 56 57
	/* State maintained by the cgroup system to allow subsystems
	 * to be "busy". Should be accessed via css_get(),
	 * css_tryget() and and css_put(). */
58 59 60 61 62 63 64 65 66

	atomic_t refcnt;

	unsigned long flags;
};

/* bits in struct cgroup_subsys_state flags field */
enum {
	CSS_ROOT, /* This CSS is the root of the subsystem */
P
Paul Menage 已提交
67
	CSS_REMOVED, /* This CSS is dead */
68 69 70
};

/*
P
Paul Menage 已提交
71 72 73 74
 * Call css_get() to hold a reference on the css; it can be used
 * for a reference obtained via:
 * - an existing ref-counted reference to the css
 * - task->cgroups for a locked task
75 76 77 78 79 80 81 82
 */

static inline void css_get(struct cgroup_subsys_state *css)
{
	/* We don't need to reference count the root state */
	if (!test_bit(CSS_ROOT, &css->flags))
		atomic_inc(&css->refcnt);
}
P
Paul Menage 已提交
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101

static inline bool css_is_removed(struct cgroup_subsys_state *css)
{
	return test_bit(CSS_REMOVED, &css->flags);
}

/*
 * Call css_tryget() to take a reference on a css if your existing
 * (known-valid) reference isn't already ref-counted. Returns false if
 * the css has been destroyed.
 */

static inline bool css_tryget(struct cgroup_subsys_state *css)
{
	if (test_bit(CSS_ROOT, &css->flags))
		return true;
	while (!atomic_inc_not_zero(&css->refcnt)) {
		if (test_bit(CSS_REMOVED, &css->flags))
			return false;
102
		cpu_relax();
P
Paul Menage 已提交
103 104 105 106
	}
	return true;
}

107 108
/*
 * css_put() should be called to release a reference taken by
P
Paul Menage 已提交
109
 * css_get() or css_tryget()
110 111
 */

112
extern void __css_put(struct cgroup_subsys_state *css);
113 114 115
static inline void css_put(struct cgroup_subsys_state *css)
{
	if (!test_bit(CSS_ROOT, &css->flags))
116
		__css_put(css);
117 118
}

119 120 121 122 123 124 125 126 127 128 129
/* bits in struct cgroup flags field */
enum {
	/* Control Group is dead */
	CGRP_REMOVED,
	/* Control Group has previously had a child cgroup or a task,
	 * but no longer (only if CGRP_NOTIFY_ON_RELEASE is set) */
	CGRP_RELEASABLE,
	/* Control Group requires release notifications to userspace */
	CGRP_NOTIFY_ON_RELEASE,
};

130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145
struct cgroup {
	unsigned long flags;		/* "unsigned long" so bitops work */

	/* count users of this cgroup. >0 means busy, but doesn't
	 * necessarily indicate the number of tasks in the
	 * cgroup */
	atomic_t count;

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

	struct cgroup *parent;	/* my parent */
146
	struct dentry *dentry;	  	/* cgroup fs entry, RCU protected */
147 148 149 150 151 152

	/* Private pointers for each registered subsystem */
	struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT];

	struct cgroupfs_root *root;
	struct cgroup *top_cgroup;
153 154 155 156 157 158

	/*
	 * List of cg_cgroup_links pointing at css_sets with
	 * tasks in this cgroup. Protected by css_set_lock
	 */
	struct list_head css_sets;
159 160 161 162 163 164 165

	/*
	 * Linked list running through all cgroups that can
	 * potentially be reaped by the release agent. Protected by
	 * release_list_lock
	 */
	struct list_head release_list;
166 167 168 169 170 171 172 173 174

	/* pids_mutex protects the fields below */
	struct rw_semaphore pids_mutex;
	/* Array of process ids in the cgroup */
	pid_t *tasks_pids;
	/* How many files are using the current tasks_pids array */
	int pids_use_count;
	/* Length of the current tasks_pids array */
	int pids_length;
175 176 177

	/* For RCU-protected deletion */
	struct rcu_head rcu_head;
178 179 180 181 182 183 184 185 186 187 188 189
};

/* A css_set is a structure holding pointers to a set of
 * cgroup_subsys_state objects. This saves space in the task struct
 * object and speeds up fork()/exit(), since a single inc/dec and a
 * list_add()/del() can bump the reference count on the entire
 * cgroup set for a task.
 */

struct css_set {

	/* Reference count */
190
	atomic_t refcount;
191

192 193 194 195 196 197
	/*
	 * List running through all cgroup groups in the same hash
	 * slot. Protected by css_set_lock
	 */
	struct hlist_node hlist;

198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216
	/*
	 * List running through all tasks using this cgroup
	 * group. Protected by css_set_lock
	 */
	struct list_head tasks;

	/*
	 * List of cg_cgroup_link objects on link chains from
	 * cgroups referenced from this css_set. Protected by
	 * css_set_lock
	 */
	struct list_head cg_links;

	/*
	 * Set of subsystem states, one for each subsystem. This array
	 * is immutable after creation apart from the init_css_set
	 * during subsystem registration (at boot time).
	 */
	struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT];
217 218
};

219 220 221 222 223 224 225 226 227 228
/*
 * cgroup_map_cb is an abstract callback API for reporting map-valued
 * control files
 */

struct cgroup_map_cb {
	int (*fill)(struct cgroup_map_cb *cb, const char *key, u64 value);
	void *state;
};

229 230 231 232 233 234 235 236 237
/* struct cftype:
 *
 * The files in the cgroup 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:
L
Li Zefan 已提交
238
 *	- the cgroup to use is file->f_dentry->d_parent->d_fsdata
239 240 241 242 243 244 245 246 247
 *	- the 'cftype' of the file is file->f_dentry->d_fsdata
 */

#define MAX_CFTYPE_NAME 64
struct cftype {
	/* By convention, the name should begin with the name of the
	 * subsystem, followed by a period */
	char name[MAX_CFTYPE_NAME];
	int private;
248 249 250 251 252 253 254

	/*
	 * If non-zero, defines the maximum length of string that can
	 * be passed to write_string; defaults to 64
	 */
	size_t max_write_len;

255 256 257 258
	int (*open)(struct inode *inode, struct file *file);
	ssize_t (*read)(struct cgroup *cgrp, struct cftype *cft,
			struct file *file,
			char __user *buf, size_t nbytes, loff_t *ppos);
259
	/*
260
	 * read_u64() is a shortcut for the common case of returning a
261 262
	 * single integer. Use it in place of read()
	 */
263
	u64 (*read_u64)(struct cgroup *cgrp, struct cftype *cft);
264 265 266
	/*
	 * read_s64() is a signed version of read_u64()
	 */
267
	s64 (*read_s64)(struct cgroup *cgrp, struct cftype *cft);
268 269 270 271 272 273
	/*
	 * read_map() is used for defining a map of key/value
	 * pairs. It should call cb->fill(cb, key, value) for each
	 * entry. The key/value pairs (and their ordering) should not
	 * change between reboots.
	 */
274 275
	int (*read_map)(struct cgroup *cont, struct cftype *cft,
			struct cgroup_map_cb *cb);
276 277 278 279
	/*
	 * read_seq_string() is used for outputting a simple sequence
	 * using seqfile.
	 */
280 281
	int (*read_seq_string)(struct cgroup *cont, struct cftype *cft,
			       struct seq_file *m);
282

283 284 285
	ssize_t (*write)(struct cgroup *cgrp, struct cftype *cft,
			 struct file *file,
			 const char __user *buf, size_t nbytes, loff_t *ppos);
286 287

	/*
288
	 * write_u64() is a shortcut for the common case of accepting
289 290 291
	 * a single integer (as parsed by simple_strtoull) from
	 * userspace. Use in place of write(); return 0 or error.
	 */
292
	int (*write_u64)(struct cgroup *cgrp, struct cftype *cft, u64 val);
293 294 295
	/*
	 * write_s64() is a signed version of write_u64()
	 */
296
	int (*write_s64)(struct cgroup *cgrp, struct cftype *cft, s64 val);
297

298 299 300 301 302 303 304
	/*
	 * write_string() is passed a nul-terminated kernelspace
	 * buffer of maximum length determined by max_write_len.
	 * Returns 0 or -ve error code.
	 */
	int (*write_string)(struct cgroup *cgrp, struct cftype *cft,
			    const char *buffer);
305 306 307 308 309 310 311 312
	/*
	 * trigger() callback can be used to get some kick from the
	 * userspace, when the actual string written is not important
	 * at all. The private field can be used to determine the
	 * kick type for multiplexing.
	 */
	int (*trigger)(struct cgroup *cgrp, unsigned int event);

313
	int (*release)(struct inode *inode, struct file *file);
314 315
};

316 317 318 319 320 321 322 323
struct cgroup_scanner {
	struct cgroup *cg;
	int (*test_task)(struct task_struct *p, struct cgroup_scanner *scan);
	void (*process_task)(struct task_struct *p,
			struct cgroup_scanner *scan);
	struct ptr_heap *heap;
};

324 325
/* Add a new file to the given cgroup directory. Should only be
 * called by subsystems from within a populate() method */
L
Li Zefan 已提交
326
int cgroup_add_file(struct cgroup *cgrp, struct cgroup_subsys *subsys,
327 328 329 330
		       const struct cftype *cft);

/* Add a set of new files to the given cgroup directory. Should
 * only be called by subsystems from within a populate() method */
L
Li Zefan 已提交
331
int cgroup_add_files(struct cgroup *cgrp,
332 333 334 335
			struct cgroup_subsys *subsys,
			const struct cftype cft[],
			int count);

L
Li Zefan 已提交
336
int cgroup_is_removed(const struct cgroup *cgrp);
337

L
Li Zefan 已提交
338
int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen);
339

L
Li Zefan 已提交
340
int cgroup_task_count(const struct cgroup *cgrp);
341

342
/* Return true if the cgroup is a descendant of the current cgroup */
L
Li Zefan 已提交
343
int cgroup_is_descendant(const struct cgroup *cgrp);
344 345 346 347 348

/* Control Group subsystem type. See Documentation/cgroups.txt for details */

struct cgroup_subsys {
	struct cgroup_subsys_state *(*create)(struct cgroup_subsys *ss,
L
Li Zefan 已提交
349 350 351
						  struct cgroup *cgrp);
	void (*pre_destroy)(struct cgroup_subsys *ss, struct cgroup *cgrp);
	void (*destroy)(struct cgroup_subsys *ss, struct cgroup *cgrp);
352
	int (*can_attach)(struct cgroup_subsys *ss,
L
Li Zefan 已提交
353 354 355
			  struct cgroup *cgrp, struct task_struct *tsk);
	void (*attach)(struct cgroup_subsys *ss, struct cgroup *cgrp,
			struct cgroup *old_cgrp, struct task_struct *tsk);
356 357 358
	void (*fork)(struct cgroup_subsys *ss, struct task_struct *task);
	void (*exit)(struct cgroup_subsys *ss, struct task_struct *task);
	int (*populate)(struct cgroup_subsys *ss,
L
Li Zefan 已提交
359 360
			struct cgroup *cgrp);
	void (*post_clone)(struct cgroup_subsys *ss, struct cgroup *cgrp);
361
	void (*bind)(struct cgroup_subsys *ss, struct cgroup *root);
362

363 364
	int subsys_id;
	int active;
365
	int disabled;
366 367 368 369
	int early_init;
#define MAX_CGROUP_TYPE_NAMELEN 32
	const char *name;

370 371 372 373 374 375 376 377 378 379 380
	/*
	 * Protects sibling/children links of cgroups in this
	 * hierarchy, plus protects which hierarchy (or none) the
	 * subsystem is a part of (i.e. root/sibling).  To avoid
	 * potential deadlocks, the following operations should not be
	 * undertaken while holding any hierarchy_mutex:
	 *
	 * - allocating memory
	 * - initiating hotplug events
	 */
	struct mutex hierarchy_mutex;
381
	struct lock_class_key subsys_key;
382

383 384 385 386 387
	/*
	 * Link to parent, and list entry in parent's children.
	 * Protected by this->hierarchy_mutex and cgroup_lock()
	 */
	struct cgroupfs_root *root;
388 389 390 391 392 393 394 395
	struct list_head sibling;
};

#define SUBSYS(_x) extern struct cgroup_subsys _x ## _subsys;
#include <linux/cgroup_subsys.h>
#undef SUBSYS

static inline struct cgroup_subsys_state *cgroup_subsys_state(
L
Li Zefan 已提交
396
	struct cgroup *cgrp, int subsys_id)
397
{
L
Li Zefan 已提交
398
	return cgrp->subsys[subsys_id];
399 400 401 402 403
}

static inline struct cgroup_subsys_state *task_subsys_state(
	struct task_struct *task, int subsys_id)
{
404
	return rcu_dereference(task->cgroups->subsys[subsys_id]);
405 406 407 408 409 410 411 412
}

static inline struct cgroup* task_cgroup(struct task_struct *task,
					       int subsys_id)
{
	return task_subsys_state(task, subsys_id)->cgroup;
}

413 414
int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *ss,
							char *nodename);
415

416 417 418 419 420 421 422 423 424 425 426 427 428 429
/* A cgroup_iter should be treated as an opaque object */
struct cgroup_iter {
	struct list_head *cg_link;
	struct list_head *task;
};

/* To iterate across the tasks in a cgroup:
 *
 * 1) call cgroup_iter_start to intialize an iterator
 *
 * 2) call cgroup_iter_next() to retrieve member tasks until it
 *    returns NULL or until you want to end the iteration
 *
 * 3) call cgroup_iter_end() to destroy the iterator.
430 431 432 433
 *
 * Or, call cgroup_scan_tasks() to iterate through every task in a cpuset.
 *    - cgroup_scan_tasks() holds the css_set_lock when calling the test_task()
 *      callback, but not while calling the process_task() callback.
434
 */
L
Li Zefan 已提交
435 436
void cgroup_iter_start(struct cgroup *cgrp, struct cgroup_iter *it);
struct task_struct *cgroup_iter_next(struct cgroup *cgrp,
437
					struct cgroup_iter *it);
L
Li Zefan 已提交
438
void cgroup_iter_end(struct cgroup *cgrp, struct cgroup_iter *it);
439
int cgroup_scan_tasks(struct cgroup_scanner *scan);
440
int cgroup_attach_task(struct cgroup *, struct task_struct *);
441

442 443 444 445
#else /* !CONFIG_CGROUPS */

static inline int cgroup_init_early(void) { return 0; }
static inline int cgroup_init(void) { return 0; }
446 447
static inline void cgroup_fork(struct task_struct *p) {}
static inline void cgroup_fork_callbacks(struct task_struct *p) {}
448
static inline void cgroup_post_fork(struct task_struct *p) {}
449
static inline void cgroup_exit(struct task_struct *p, int callbacks) {}
450 451 452

static inline void cgroup_lock(void) {}
static inline void cgroup_unlock(void) {}
B
Balbir Singh 已提交
453 454 455 456 457
static inline int cgroupstats_build(struct cgroupstats *stats,
					struct dentry *dentry)
{
	return -EINVAL;
}
458 459 460 461

#endif /* !CONFIG_CGROUPS */

#endif /* _LINUX_CGROUP_H */