memcontrol.h 24.1 KB
Newer Older
B
Balbir Singh 已提交
1 2 3 4 5
/* memcontrol.h - Memory Controller
 *
 * Copyright IBM Corporation, 2007
 * Author Balbir Singh <balbir@linux.vnet.ibm.com>
 *
6 7 8
 * Copyright 2007 OpenVZ SWsoft Inc
 * Author: Pavel Emelianov <xemul@openvz.org>
 *
B
Balbir Singh 已提交
9 10 11 12 13 14 15 16 17 18 19 20 21
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 */

#ifndef _LINUX_MEMCONTROL_H
#define _LINUX_MEMCONTROL_H
22
#include <linux/cgroup.h>
23
#include <linux/vm_event_item.h>
24
#include <linux/hardirq.h>
25
#include <linux/jump_label.h>
M
Michal Hocko 已提交
26 27 28 29 30
#include <linux/page_counter.h>
#include <linux/vmpressure.h>
#include <linux/eventfd.h>
#include <linux/mmzone.h>
#include <linux/writeback.h>
31
#include <linux/page-flags.h>
32

33
struct mem_cgroup;
34 35
struct page;
struct mm_struct;
36
struct kmem_cache;
37

S
Sha Zhengju 已提交
38 39 40 41 42 43 44 45 46 47 48
/*
 * The corresponding mem_cgroup_stat_names is defined in mm/memcontrol.c,
 * These two lists should keep in accord with each other.
 */
enum mem_cgroup_stat_index {
	/*
	 * For MEM_CONTAINER_TYPE_ALL, usage = pagecache + rss.
	 */
	MEM_CGROUP_STAT_CACHE,		/* # of pages charged as cache */
	MEM_CGROUP_STAT_RSS,		/* # of pages charged as anon rss */
	MEM_CGROUP_STAT_RSS_HUGE,	/* # of pages charged as anon huge */
49
	MEM_CGROUP_STAT_SHMEM,		/* # of pages charged as shmem */
S
Sha Zhengju 已提交
50
	MEM_CGROUP_STAT_FILE_MAPPED,	/* # of pages charged as file rss */
51
	MEM_CGROUP_STAT_DIRTY,          /* # of dirty pages in page cache */
52
	MEM_CGROUP_STAT_WRITEBACK,	/* # of pages under writeback */
S
Sha Zhengju 已提交
53 54
	MEM_CGROUP_STAT_SWAP,		/* # of pages, swapped out */
	MEM_CGROUP_STAT_NSTATS,
55
	/* default hierarchy stats */
56
	MEMCG_KERNEL_STACK_KB = MEM_CGROUP_STAT_NSTATS,
57 58
	MEMCG_SLAB_RECLAIMABLE,
	MEMCG_SLAB_UNRECLAIMABLE,
59
	MEMCG_SOCK,
60 61 62
	MEMCG_WORKINGSET_REFAULT,
	MEMCG_WORKINGSET_ACTIVATE,
	MEMCG_WORKINGSET_NODERECLAIM,
63
	MEMCG_NR_STAT,
64 65
};

66
struct mem_cgroup_reclaim_cookie {
67
	pg_data_t *pgdat;
68 69 70 71
	int priority;
	unsigned int generation;
};

72 73 74 75 76 77 78 79 80 81 82 83 84 85
enum mem_cgroup_events_index {
	MEM_CGROUP_EVENTS_PGPGIN,	/* # of pages paged in */
	MEM_CGROUP_EVENTS_PGPGOUT,	/* # of pages paged out */
	MEM_CGROUP_EVENTS_PGFAULT,	/* # of page-faults */
	MEM_CGROUP_EVENTS_PGMAJFAULT,	/* # of major page-faults */
	MEM_CGROUP_EVENTS_NSTATS,
	/* default hierarchy events */
	MEMCG_LOW = MEM_CGROUP_EVENTS_NSTATS,
	MEMCG_HIGH,
	MEMCG_MAX,
	MEMCG_OOM,
	MEMCG_NR_EVENTS,
};

M
Michal Hocko 已提交
86 87 88 89 90 91 92 93 94 95 96 97 98
/*
 * Per memcg event counter is incremented at every pagein/pageout. With THP,
 * it will be incremated by the number of pages. This counter is used for
 * for trigger some periodic events. This is straightforward and better
 * than using jiffies etc. to handle periodic memcg event.
 */
enum mem_cgroup_events_target {
	MEM_CGROUP_TARGET_THRESH,
	MEM_CGROUP_TARGET_SOFTLIMIT,
	MEM_CGROUP_TARGET_NUMAINFO,
	MEM_CGROUP_NTARGETS,
};

A
Andrew Morton 已提交
99
#ifdef CONFIG_MEMCG
100 101 102 103

#define MEM_CGROUP_ID_SHIFT	16
#define MEM_CGROUP_ID_MAX	USHRT_MAX

104 105 106 107 108
struct mem_cgroup_id {
	int id;
	atomic_t ref;
};

M
Michal Hocko 已提交
109
struct mem_cgroup_stat_cpu {
110
	long count[MEMCG_NR_STAT];
M
Michal Hocko 已提交
111 112 113 114 115 116 117 118 119 120 121 122 123 124
	unsigned long events[MEMCG_NR_EVENTS];
	unsigned long nr_page_events;
	unsigned long targets[MEM_CGROUP_NTARGETS];
};

struct mem_cgroup_reclaim_iter {
	struct mem_cgroup *position;
	/* scan generation, increased every round-trip */
	unsigned int generation;
};

/*
 * per-zone information in memory controller.
 */
125
struct mem_cgroup_per_node {
M
Michal Hocko 已提交
126
	struct lruvec		lruvec;
127
	unsigned long		lru_zone_size[MAX_NR_ZONES][NR_LRU_LISTS];
M
Michal Hocko 已提交
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

	struct mem_cgroup_reclaim_iter	iter[DEF_PRIORITY + 1];

	struct rb_node		tree_node;	/* RB tree node */
	unsigned long		usage_in_excess;/* Set to the value by which */
						/* the soft limit is exceeded*/
	bool			on_tree;
	struct mem_cgroup	*memcg;		/* Back pointer, we cannot */
						/* use container_of	   */
};

struct mem_cgroup_threshold {
	struct eventfd_ctx *eventfd;
	unsigned long threshold;
};

/* For threshold */
struct mem_cgroup_threshold_ary {
	/* An array index points to threshold just below or equal to usage. */
	int current_threshold;
	/* Size of entries[] */
	unsigned int size;
	/* Array of thresholds */
	struct mem_cgroup_threshold entries[0];
};

struct mem_cgroup_thresholds {
	/* Primary thresholds array */
	struct mem_cgroup_threshold_ary *primary;
	/*
	 * Spare threshold array.
	 * This is needed to make mem_cgroup_unregister_event() "never fail".
	 * It must be able to store at least primary->size - 1 entries.
	 */
	struct mem_cgroup_threshold_ary *spare;
};

165 166 167 168 169 170
enum memcg_kmem_state {
	KMEM_NONE,
	KMEM_ALLOCATED,
	KMEM_ONLINE,
};

M
Michal Hocko 已提交
171 172 173 174 175 176 177 178 179
/*
 * The memory controller data structure. The memory controller controls both
 * page cache and RSS per cgroup. We would eventually like to provide
 * statistics based on the statistics developed by Rik Van Riel for clock-pro,
 * to help the administrator determine what knobs to tune.
 */
struct mem_cgroup {
	struct cgroup_subsys_state css;

180 181 182
	/* Private memcg ID. Used to ID objects that outlive the cgroup */
	struct mem_cgroup_id id;

M
Michal Hocko 已提交
183 184
	/* Accounted resources */
	struct page_counter memory;
185
	struct page_counter swap;
186 187

	/* Legacy consumer-oriented counters */
M
Michal Hocko 已提交
188 189
	struct page_counter memsw;
	struct page_counter kmem;
190
	struct page_counter tcpmem;
M
Michal Hocko 已提交
191 192 193 194 195

	/* Normal memory consumption range */
	unsigned long low;
	unsigned long high;

196 197 198
	/* Range enforcement for interrupt charges */
	struct work_struct high_work;

M
Michal Hocko 已提交
199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216
	unsigned long soft_limit;

	/* vmpressure notifications */
	struct vmpressure vmpressure;

	/*
	 * Should the accounting and control be hierarchical, per subtree?
	 */
	bool use_hierarchy;

	/* protected by memcg_oom_lock */
	bool		oom_lock;
	int		under_oom;

	int	swappiness;
	/* OOM-Killer disable */
	int		oom_kill_disable;

217 218 219
	/* handle for "memory.events" */
	struct cgroup_file events_file;

M
Michal Hocko 已提交
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
	/* protect arrays of thresholds */
	struct mutex thresholds_lock;

	/* thresholds for memory usage. RCU-protected */
	struct mem_cgroup_thresholds thresholds;

	/* thresholds for mem+swap usage. RCU-protected */
	struct mem_cgroup_thresholds memsw_thresholds;

	/* For oom notifier event fd */
	struct list_head oom_notify;

	/*
	 * Should we move charges of a task when a task is moved into this
	 * mem_cgroup ? And what type of charges should we move ?
	 */
	unsigned long move_charge_at_immigrate;
	/*
	 * set > 0 if pages under this cgroup are moving to other cgroup.
	 */
	atomic_t		moving_account;
	/* taken only while moving_account > 0 */
	spinlock_t		move_lock;
	struct task_struct	*move_lock_task;
	unsigned long		move_lock_flags;
	/*
	 * percpu counter.
	 */
	struct mem_cgroup_stat_cpu __percpu *stat;

250 251 252
	unsigned long		socket_pressure;

	/* Legacy tcp memory accounting */
253 254
	bool			tcpmem_active;
	int			tcpmem_pressure;
255

256
#ifndef CONFIG_SLOB
M
Michal Hocko 已提交
257 258
        /* Index in the kmem_cache->memcg_params.memcg_caches array */
	int kmemcg_id;
259
	enum memcg_kmem_state kmem_state;
260
	struct list_head kmem_caches;
M
Michal Hocko 已提交
261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
#endif

	int last_scanned_node;
#if MAX_NUMNODES > 1
	nodemask_t	scan_nodes;
	atomic_t	numainfo_events;
	atomic_t	numainfo_updating;
#endif

#ifdef CONFIG_CGROUP_WRITEBACK
	struct list_head cgwb_list;
	struct wb_domain cgwb_domain;
#endif

	/* List of events which userspace want to receive */
	struct list_head event_list;
	spinlock_t event_list_lock;

	struct mem_cgroup_per_node *nodeinfo[0];
	/* WARNING: nodeinfo must be the last member here */
};
282 283

extern struct mem_cgroup *root_mem_cgroup;
T
Tejun Heo 已提交
284

285 286 287 288 289
static inline bool mem_cgroup_disabled(void)
{
	return !cgroup_subsys_enabled(memory_cgrp_subsys);
}

M
Michal Hocko 已提交
290 291 292 293 294 295 296
/**
 * mem_cgroup_events - count memory events against a cgroup
 * @memcg: the memory cgroup
 * @idx: the event index
 * @nr: the number of events to account for
 */
static inline void mem_cgroup_events(struct mem_cgroup *memcg,
297
		       enum mem_cgroup_events_index idx,
M
Michal Hocko 已提交
298 299 300
		       unsigned int nr)
{
	this_cpu_add(memcg->stat->events[idx], nr);
301
	cgroup_file_notify(&memcg->events_file);
M
Michal Hocko 已提交
302
}
303 304 305

bool mem_cgroup_low(struct mem_cgroup *root, struct mem_cgroup *memcg);

306
int mem_cgroup_try_charge(struct page *page, struct mm_struct *mm,
307 308
			  gfp_t gfp_mask, struct mem_cgroup **memcgp,
			  bool compound);
309
void mem_cgroup_commit_charge(struct page *page, struct mem_cgroup *memcg,
310 311 312
			      bool lrucare, bool compound);
void mem_cgroup_cancel_charge(struct page *page, struct mem_cgroup *memcg,
		bool compound);
313
void mem_cgroup_uncharge(struct page *page);
314
void mem_cgroup_uncharge_list(struct list_head *page_list);
315

316
void mem_cgroup_migrate(struct page *oldpage, struct page *newpage);
317

318 319
static struct mem_cgroup_per_node *
mem_cgroup_nodeinfo(struct mem_cgroup *memcg, int nid)
320
{
321
	return memcg->nodeinfo[nid];
322 323 324
}

/**
325 326
 * mem_cgroup_lruvec - get the lru list vector for a node or a memcg zone
 * @node: node of the wanted lruvec
327 328
 * @memcg: memcg of the wanted lruvec
 *
329 330
 * Returns the lru list vector holding pages for a given @node or a given
 * @memcg and @zone. This can be the node lruvec, if the memory controller
331 332
 * is disabled.
 */
333
static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat,
334
				struct mem_cgroup *memcg)
335
{
336
	struct mem_cgroup_per_node *mz;
337 338 339
	struct lruvec *lruvec;

	if (mem_cgroup_disabled()) {
340
		lruvec = node_lruvec(pgdat);
341 342 343
		goto out;
	}

344
	mz = mem_cgroup_nodeinfo(memcg, pgdat->node_id);
345 346 347 348
	lruvec = &mz->lruvec;
out:
	/*
	 * Since a node can be onlined after the mem_cgroup was created,
M
Mel Gorman 已提交
349
	 * we have to be prepared to initialize lruvec->pgdat here;
350 351
	 * and if offlined then reonlined, we need to reinitialize it.
	 */
352 353
	if (unlikely(lruvec->pgdat != pgdat))
		lruvec->pgdat = pgdat;
354 355 356
	return lruvec;
}

M
Mel Gorman 已提交
357
struct lruvec *mem_cgroup_page_lruvec(struct page *, struct pglist_data *);
358

359
bool task_in_mem_cgroup(struct task_struct *task, struct mem_cgroup *memcg);
360
struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p);
361

M
Michal Hocko 已提交
362 363 364 365 366
static inline
struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *css){
	return css ? container_of(css, struct mem_cgroup, css) : NULL;
}

367 368 369
#define mem_cgroup_from_counter(counter, member)	\
	container_of(counter, struct mem_cgroup, member)

M
Michal Hocko 已提交
370 371 372 373
struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *,
				   struct mem_cgroup *,
				   struct mem_cgroup_reclaim_cookie *);
void mem_cgroup_iter_break(struct mem_cgroup *, struct mem_cgroup *);
374 375
int mem_cgroup_scan_tasks(struct mem_cgroup *,
			  int (*)(struct task_struct *, void *), void *);
M
Michal Hocko 已提交
376

377 378 379 380 381
static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
{
	if (mem_cgroup_disabled())
		return 0;

382
	return memcg->id.id;
383
}
384
struct mem_cgroup *mem_cgroup_from_id(unsigned short id);
385

386 387 388 389 390 391 392 393 394 395 396 397 398 399
/**
 * parent_mem_cgroup - find the accounting parent of a memcg
 * @memcg: memcg whose parent to find
 *
 * Returns the parent memcg, or NULL if this is the root or the memory
 * controller is in legacy no-hierarchy mode.
 */
static inline struct mem_cgroup *parent_mem_cgroup(struct mem_cgroup *memcg)
{
	if (!memcg->memory.parent)
		return NULL;
	return mem_cgroup_from_counter(memcg->memory.parent, memory);
}

M
Michal Hocko 已提交
400 401 402 403 404 405 406 407 408
static inline bool mem_cgroup_is_descendant(struct mem_cgroup *memcg,
			      struct mem_cgroup *root)
{
	if (root == memcg)
		return true;
	if (!root->use_hierarchy)
		return false;
	return cgroup_is_descendant(memcg->css.cgroup, root->css.cgroup);
}
G
Glauber Costa 已提交
409

410 411
static inline bool mm_match_cgroup(struct mm_struct *mm,
				   struct mem_cgroup *memcg)
412
{
413
	struct mem_cgroup *task_memcg;
414
	bool match = false;
415

416
	rcu_read_lock();
417
	task_memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
418
	if (task_memcg)
419
		match = mem_cgroup_is_descendant(task_memcg, memcg);
420
	rcu_read_unlock();
421
	return match;
422
}
423

424
struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page);
425
ino_t page_cgroup_ino(struct page *page);
426

427 428 429 430 431 432 433
static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
{
	if (mem_cgroup_disabled())
		return true;
	return !!(memcg->css.flags & CSS_ONLINE);
}

434 435 436
/*
 * For memory reclaim.
 */
437
int mem_cgroup_select_victim_node(struct mem_cgroup *memcg);
M
Michal Hocko 已提交
438 439

void mem_cgroup_update_lru_size(struct lruvec *lruvec, enum lru_list lru,
440
		int zid, int nr_pages);
M
Michal Hocko 已提交
441

442 443 444
unsigned long mem_cgroup_node_nr_lru_pages(struct mem_cgroup *memcg,
					   int nid, unsigned int lru_mask);

M
Michal Hocko 已提交
445 446 447
static inline
unsigned long mem_cgroup_get_lru_size(struct lruvec *lruvec, enum lru_list lru)
{
448
	struct mem_cgroup_per_node *mz;
449 450
	unsigned long nr_pages = 0;
	int zid;
M
Michal Hocko 已提交
451

452
	mz = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
453 454 455 456 457 458 459 460 461 462 463 464 465
	for (zid = 0; zid < MAX_NR_ZONES; zid++)
		nr_pages += mz->lru_zone_size[zid][lru];
	return nr_pages;
}

static inline
unsigned long mem_cgroup_get_zone_lru_size(struct lruvec *lruvec,
		enum lru_list lru, int zone_idx)
{
	struct mem_cgroup_per_node *mz;

	mz = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
	return mz->lru_zone_size[zone_idx][lru];
M
Michal Hocko 已提交
466 467
}

468 469
void mem_cgroup_handle_over_high(void);

470 471
unsigned long mem_cgroup_get_limit(struct mem_cgroup *memcg);

472 473
void mem_cgroup_print_oom_info(struct mem_cgroup *memcg,
				struct task_struct *p);
474

475
static inline void mem_cgroup_oom_enable(void)
476
{
T
Tejun Heo 已提交
477 478
	WARN_ON(current->memcg_may_oom);
	current->memcg_may_oom = 1;
479 480
}

481
static inline void mem_cgroup_oom_disable(void)
482
{
T
Tejun Heo 已提交
483 484
	WARN_ON(!current->memcg_may_oom);
	current->memcg_may_oom = 0;
485 486
}

487 488
static inline bool task_in_memcg_oom(struct task_struct *p)
{
T
Tejun Heo 已提交
489
	return p->memcg_in_oom;
490 491
}

492
bool mem_cgroup_oom_synchronize(bool wait);
493

A
Andrew Morton 已提交
494
#ifdef CONFIG_MEMCG_SWAP
495 496
extern int do_swap_account;
#endif
497

J
Johannes Weiner 已提交
498 499
void lock_page_memcg(struct page *page);
void unlock_page_memcg(struct page *page);
500

501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg,
						 enum mem_cgroup_stat_index idx)
{
	long val = 0;
	int cpu;

	for_each_possible_cpu(cpu)
		val += per_cpu(memcg->stat->count[idx], cpu);

	if (val < 0)
		val = 0;

	return val;
}

static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg,
				   enum mem_cgroup_stat_index idx, int val)
{
	if (!mem_cgroup_disabled())
		this_cpu_add(memcg->stat->count[idx], val);
}

static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg,
				   enum mem_cgroup_stat_index idx)
{
	mem_cgroup_update_stat(memcg, idx, 1);
}

static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg,
				   enum mem_cgroup_stat_index idx)
{
	mem_cgroup_update_stat(memcg, idx, -1);
}

M
Michal Hocko 已提交
535 536
/**
 * mem_cgroup_update_page_stat - update page state statistics
J
Johannes Weiner 已提交
537
 * @page: the page
M
Michal Hocko 已提交
538 539 540
 * @idx: page state item to account
 * @val: number of pages (positive or negative)
 *
541 542 543
 * The @page must be locked or the caller must use lock_page_memcg()
 * to prevent double accounting when the page is concurrently being
 * moved to another memcg:
544
 *
545
 *   lock_page(page) or lock_page_memcg(page)
546
 *   if (TestClearPageState(page))
J
Johannes Weiner 已提交
547
 *     mem_cgroup_update_page_stat(page, state, -1);
548
 *   unlock_page(page) or unlock_page_memcg(page)
549 550
 *
 * Kernel pages are an exception to this, since they'll never move.
M
Michal Hocko 已提交
551
 */
J
Johannes Weiner 已提交
552
static inline void mem_cgroup_update_page_stat(struct page *page,
M
Michal Hocko 已提交
553 554
				 enum mem_cgroup_stat_index idx, int val)
{
J
Johannes Weiner 已提交
555
	if (page->mem_cgroup)
556
		mem_cgroup_update_stat(page->mem_cgroup, idx, val);
M
Michal Hocko 已提交
557 558
}

J
Johannes Weiner 已提交
559
static inline void mem_cgroup_inc_page_stat(struct page *page,
S
Sha Zhengju 已提交
560
					    enum mem_cgroup_stat_index idx)
561
{
J
Johannes Weiner 已提交
562
	mem_cgroup_update_page_stat(page, idx, 1);
563 564
}

J
Johannes Weiner 已提交
565
static inline void mem_cgroup_dec_page_stat(struct page *page,
S
Sha Zhengju 已提交
566
					    enum mem_cgroup_stat_index idx)
567
{
J
Johannes Weiner 已提交
568
	mem_cgroup_update_page_stat(page, idx, -1);
569 570
}

571
unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
572 573
						gfp_t gfp_mask,
						unsigned long *total_scanned);
D
David Rientjes 已提交
574

575 576 577
static inline void mem_cgroup_count_vm_event(struct mm_struct *mm,
					     enum vm_event_item idx)
{
M
Michal Hocko 已提交
578 579
	struct mem_cgroup *memcg;

580 581
	if (mem_cgroup_disabled())
		return;
M
Michal Hocko 已提交
582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599

	rcu_read_lock();
	memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
	if (unlikely(!memcg))
		goto out;

	switch (idx) {
	case PGFAULT:
		this_cpu_inc(memcg->stat->events[MEM_CGROUP_EVENTS_PGFAULT]);
		break;
	case PGMAJFAULT:
		this_cpu_inc(memcg->stat->events[MEM_CGROUP_EVENTS_PGMAJFAULT]);
		break;
	default:
		BUG();
	}
out:
	rcu_read_unlock();
600
}
601
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
602
void mem_cgroup_split_huge_fixup(struct page *head);
603 604
#endif

A
Andrew Morton 已提交
605
#else /* CONFIG_MEMCG */
606 607 608 609

#define MEM_CGROUP_ID_SHIFT	0
#define MEM_CGROUP_ID_MAX	0

610 611
struct mem_cgroup;

612 613 614 615 616
static inline bool mem_cgroup_disabled(void)
{
	return true;
}

617 618 619 620 621 622 623 624 625 626 627 628
static inline void mem_cgroup_events(struct mem_cgroup *memcg,
				     enum mem_cgroup_events_index idx,
				     unsigned int nr)
{
}

static inline bool mem_cgroup_low(struct mem_cgroup *root,
				  struct mem_cgroup *memcg)
{
	return false;
}

629 630
static inline int mem_cgroup_try_charge(struct page *page, struct mm_struct *mm,
					gfp_t gfp_mask,
631 632
					struct mem_cgroup **memcgp,
					bool compound)
633
{
634
	*memcgp = NULL;
635 636 637
	return 0;
}

638 639
static inline void mem_cgroup_commit_charge(struct page *page,
					    struct mem_cgroup *memcg,
640
					    bool lrucare, bool compound)
641 642 643
{
}

644
static inline void mem_cgroup_cancel_charge(struct page *page,
645 646
					    struct mem_cgroup *memcg,
					    bool compound)
647 648 649
{
}

650
static inline void mem_cgroup_uncharge(struct page *page)
651 652 653
{
}

654
static inline void mem_cgroup_uncharge_list(struct list_head *page_list)
655 656 657
{
}

658
static inline void mem_cgroup_migrate(struct page *old, struct page *new)
659 660 661
{
}

662
static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat,
663
				struct mem_cgroup *memcg)
K
KAMEZAWA Hiroyuki 已提交
664
{
665
	return node_lruvec(pgdat);
K
KAMEZAWA Hiroyuki 已提交
666 667
}

668
static inline struct lruvec *mem_cgroup_page_lruvec(struct page *page,
M
Mel Gorman 已提交
669
						    struct pglist_data *pgdat)
670
{
M
Mel Gorman 已提交
671
	return &pgdat->lruvec;
672 673
}

674
static inline bool mm_match_cgroup(struct mm_struct *mm,
675
		struct mem_cgroup *memcg)
676
{
677
	return true;
678 679
}

680 681
static inline bool task_in_mem_cgroup(struct task_struct *task,
				      const struct mem_cgroup *memcg)
682
{
683
	return true;
684 685
}

686 687 688 689 690 691 692 693 694 695 696 697 698
static inline struct mem_cgroup *
mem_cgroup_iter(struct mem_cgroup *root,
		struct mem_cgroup *prev,
		struct mem_cgroup_reclaim_cookie *reclaim)
{
	return NULL;
}

static inline void mem_cgroup_iter_break(struct mem_cgroup *root,
					 struct mem_cgroup *prev)
{
}

699 700 701 702 703 704
static inline int mem_cgroup_scan_tasks(struct mem_cgroup *memcg,
		int (*fn)(struct task_struct *, void *), void *arg)
{
	return 0;
}

705
static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
706
{
707 708 709 710 711 712 713 714
	return 0;
}

static inline struct mem_cgroup *mem_cgroup_from_id(unsigned short id)
{
	WARN_ON_ONCE(id);
	/* XXX: This should always return root_mem_cgroup */
	return NULL;
715
}
716

717
static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
718
{
719
	return true;
720 721
}

722
static inline unsigned long
723
mem_cgroup_get_lru_size(struct lruvec *lruvec, enum lru_list lru)
724 725 726
{
	return 0;
}
727 728 729 730 731 732
static inline
unsigned long mem_cgroup_get_zone_lru_size(struct lruvec *lruvec,
		enum lru_list lru, int zone_idx)
{
	return 0;
}
733

734 735 736 737 738 739 740
static inline unsigned long
mem_cgroup_node_nr_lru_pages(struct mem_cgroup *memcg,
			     int nid, unsigned int lru_mask)
{
	return 0;
}

741 742 743 744 745
static inline unsigned long mem_cgroup_get_limit(struct mem_cgroup *memcg)
{
	return 0;
}

746 747 748 749 750
static inline void
mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p)
{
}

J
Johannes Weiner 已提交
751
static inline void lock_page_memcg(struct page *page)
752 753 754
{
}

J
Johannes Weiner 已提交
755
static inline void unlock_page_memcg(struct page *page)
756 757 758
{
}

759 760 761 762
static inline void mem_cgroup_handle_over_high(void)
{
}

763
static inline void mem_cgroup_oom_enable(void)
764 765 766
{
}

767
static inline void mem_cgroup_oom_disable(void)
768 769 770
{
}

771 772 773 774 775
static inline bool task_in_memcg_oom(struct task_struct *p)
{
	return false;
}

776
static inline bool mem_cgroup_oom_synchronize(bool wait)
777 778 779 780
{
	return false;
}

781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801
static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg,
						 enum mem_cgroup_stat_index idx)
{
	return 0;
}

static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg,
				   enum mem_cgroup_stat_index idx, int val)
{
}

static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg,
				   enum mem_cgroup_stat_index idx)
{
}

static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg,
				   enum mem_cgroup_stat_index idx)
{
}

802 803 804 805 806 807
static inline void mem_cgroup_update_page_stat(struct page *page,
					       enum mem_cgroup_stat_index idx,
					       int nr)
{
}

J
Johannes Weiner 已提交
808
static inline void mem_cgroup_inc_page_stat(struct page *page,
S
Sha Zhengju 已提交
809
					    enum mem_cgroup_stat_index idx)
810 811 812
{
}

J
Johannes Weiner 已提交
813
static inline void mem_cgroup_dec_page_stat(struct page *page,
S
Sha Zhengju 已提交
814
					    enum mem_cgroup_stat_index idx)
815 816 817
{
}

818
static inline
819
unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
820 821
					    gfp_t gfp_mask,
					    unsigned long *total_scanned)
822
{
823
	return 0;
824 825
}

826
static inline void mem_cgroup_split_huge_fixup(struct page *head)
827 828 829
{
}

830 831 832 833
static inline
void mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx)
{
}
A
Andrew Morton 已提交
834
#endif /* CONFIG_MEMCG */
835

836
#ifdef CONFIG_CGROUP_WRITEBACK
T
Tejun Heo 已提交
837

838
struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg);
T
Tejun Heo 已提交
839
struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb);
840 841 842
void mem_cgroup_wb_stats(struct bdi_writeback *wb, unsigned long *pfilepages,
			 unsigned long *pheadroom, unsigned long *pdirty,
			 unsigned long *pwriteback);
T
Tejun Heo 已提交
843 844 845 846 847 848 849 850

#else	/* CONFIG_CGROUP_WRITEBACK */

static inline struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb)
{
	return NULL;
}

851
static inline void mem_cgroup_wb_stats(struct bdi_writeback *wb,
852 853
				       unsigned long *pfilepages,
				       unsigned long *pheadroom,
854 855 856 857 858
				       unsigned long *pdirty,
				       unsigned long *pwriteback)
{
}

T
Tejun Heo 已提交
859
#endif	/* CONFIG_CGROUP_WRITEBACK */
860

G
Glauber Costa 已提交
861
struct sock;
862 863
bool mem_cgroup_charge_skmem(struct mem_cgroup *memcg, unsigned int nr_pages);
void mem_cgroup_uncharge_skmem(struct mem_cgroup *memcg, unsigned int nr_pages);
864
#ifdef CONFIG_MEMCG
865 866
extern struct static_key_false memcg_sockets_enabled_key;
#define mem_cgroup_sockets_enabled static_branch_unlikely(&memcg_sockets_enabled_key)
867 868
void mem_cgroup_sk_alloc(struct sock *sk);
void mem_cgroup_sk_free(struct sock *sk);
869
static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg)
870
{
871
	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && memcg->tcpmem_pressure)
872 873 874 875 876 877
		return true;
	do {
		if (time_before(jiffies, memcg->socket_pressure))
			return true;
	} while ((memcg = parent_mem_cgroup(memcg)));
	return false;
878 879
}
#else
880
#define mem_cgroup_sockets_enabled 0
881 882
static inline void mem_cgroup_sk_alloc(struct sock *sk) { };
static inline void mem_cgroup_sk_free(struct sock *sk) { };
883
static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg)
884 885 886 887
{
	return false;
}
#endif
888

889 890 891 892 893 894 895
struct kmem_cache *memcg_kmem_get_cache(struct kmem_cache *cachep);
void memcg_kmem_put_cache(struct kmem_cache *cachep);
int memcg_kmem_charge_memcg(struct page *page, gfp_t gfp, int order,
			    struct mem_cgroup *memcg);
int memcg_kmem_charge(struct page *page, gfp_t gfp, int order);
void memcg_kmem_uncharge(struct page *page, int order);

896
#if defined(CONFIG_MEMCG) && !defined(CONFIG_SLOB)
897
extern struct static_key_false memcg_kmem_enabled_key;
898
extern struct workqueue_struct *memcg_kmem_cache_wq;
899

900
extern int memcg_nr_cache_ids;
901 902
void memcg_get_cache_ids(void);
void memcg_put_cache_ids(void);
903 904 905 906 907 908

/*
 * Helper macro to loop through all memcg-specific caches. Callers must still
 * check if the cache is valid (it is either valid or NULL).
 * the slab_mutex must be held when looping through those caches
 */
909
#define for_each_memcg_cache_index(_idx)	\
910
	for ((_idx) = 0; (_idx) < memcg_nr_cache_ids; (_idx)++)
911

912 913
static inline bool memcg_kmem_enabled(void)
{
914
	return static_branch_unlikely(&memcg_kmem_enabled_key);
915 916
}

M
Michal Hocko 已提交
917
/*
J
Jesper Dangaard Brouer 已提交
918
 * helper for accessing a memcg's index. It will be used as an index in the
M
Michal Hocko 已提交
919 920 921 922 923 924 925
 * child cache array in kmem_cache, and also to derive its name. This function
 * will return -1 when this is not a kmem-limited memcg.
 */
static inline int memcg_cache_id(struct mem_cgroup *memcg)
{
	return memcg ? memcg->kmemcg_id : -1;
}
926

927 928 929 930 931 932 933 934 935 936 937 938 939
/**
 * memcg_kmem_update_page_stat - update kmem page state statistics
 * @page: the page
 * @idx: page state item to account
 * @val: number of pages (positive or negative)
 */
static inline void memcg_kmem_update_page_stat(struct page *page,
				enum mem_cgroup_stat_index idx, int val)
{
	if (memcg_kmem_enabled() && page->mem_cgroup)
		this_cpu_add(page->mem_cgroup->stat->count[idx], val);
}

940
#else
941 942 943
#define for_each_memcg_cache_index(_idx)	\
	for (; NULL; )

944 945 946 947 948
static inline bool memcg_kmem_enabled(void)
{
	return false;
}

949 950 951 952 953
static inline int memcg_cache_id(struct mem_cgroup *memcg)
{
	return -1;
}

954 955 956 957 958 959 960 961
static inline void memcg_get_cache_ids(void)
{
}

static inline void memcg_put_cache_ids(void)
{
}

962 963 964 965
static inline void memcg_kmem_update_page_stat(struct page *page,
				enum mem_cgroup_stat_index idx, int val)
{
}
966 967
#endif /* CONFIG_MEMCG && !CONFIG_SLOB */

B
Balbir Singh 已提交
968
#endif /* _LINUX_MEMCONTROL_H */