memcontrol.h 43.0 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0-or-later */
B
Balbir Singh 已提交
2 3 4 5 6
/* memcontrol.h - Memory Controller
 *
 * Copyright IBM Corporation, 2007
 * Author Balbir Singh <balbir@linux.vnet.ibm.com>
 *
7 8
 * Copyright 2007 OpenVZ SWsoft Inc
 * Author: Pavel Emelianov <xemul@openvz.org>
B
Balbir Singh 已提交
9 10 11 12
 */

#ifndef _LINUX_MEMCONTROL_H
#define _LINUX_MEMCONTROL_H
13
#include <linux/cgroup.h>
14
#include <linux/vm_event_item.h>
15
#include <linux/hardirq.h>
16
#include <linux/jump_label.h>
M
Michal Hocko 已提交
17 18 19
#include <linux/page_counter.h>
#include <linux/vmpressure.h>
#include <linux/eventfd.h>
20 21
#include <linux/mm.h>
#include <linux/vmstat.h>
M
Michal Hocko 已提交
22
#include <linux/writeback.h>
23
#include <linux/page-flags.h>
24

25
struct mem_cgroup;
R
Roman Gushchin 已提交
26
struct obj_cgroup;
27 28
struct page;
struct mm_struct;
29
struct kmem_cache;
30

31 32
/* Cgroup-specific page state, on top of universal node page state */
enum memcg_stat_item {
33
	MEMCG_SWAP = NR_VM_NODE_STAT_ITEMS,
34
	MEMCG_SOCK,
35
	MEMCG_PERCPU_B,
36
	MEMCG_NR_STAT,
37 38
};

39 40
enum memcg_memory_event {
	MEMCG_LOW,
41 42 43
	MEMCG_HIGH,
	MEMCG_MAX,
	MEMCG_OOM,
R
Roman Gushchin 已提交
44
	MEMCG_OOM_KILL,
45
	MEMCG_SWAP_HIGH,
46 47
	MEMCG_SWAP_MAX,
	MEMCG_SWAP_FAIL,
48
	MEMCG_NR_MEMORY_EVENTS,
49 50
};

51
struct mem_cgroup_reclaim_cookie {
52
	pg_data_t *pgdat;
53 54 55
	unsigned int generation;
};

56 57 58 59 60 61 62
#ifdef CONFIG_MEMCG

#define MEM_CGROUP_ID_SHIFT	16
#define MEM_CGROUP_ID_MAX	USHRT_MAX

struct mem_cgroup_id {
	int id;
63
	refcount_t ref;
64 65
};

M
Michal Hocko 已提交
66 67
/*
 * Per memcg event counter is incremented at every pagein/pageout. With THP,
68 69
 * it will be incremented by the number of pages. This counter is used
 * to trigger some periodic events. This is straightforward and better
M
Michal Hocko 已提交
70 71 72 73 74 75 76 77
 * than using jiffies etc. to handle periodic memcg event.
 */
enum mem_cgroup_events_target {
	MEM_CGROUP_TARGET_THRESH,
	MEM_CGROUP_TARGET_SOFTLIMIT,
	MEM_CGROUP_NTARGETS,
};

78
struct memcg_vmstats_percpu {
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
	/* Local (CPU and cgroup) page state & events */
	long			state[MEMCG_NR_STAT];
	unsigned long		events[NR_VM_EVENT_ITEMS];

	/* Delta calculation for lockless upward propagation */
	long			state_prev[MEMCG_NR_STAT];
	unsigned long		events_prev[NR_VM_EVENT_ITEMS];

	/* Cgroup1: threshold notifications & softlimit tree updates */
	unsigned long		nr_page_events;
	unsigned long		targets[MEM_CGROUP_NTARGETS];
};

struct memcg_vmstats {
	/* Aggregated (CPU and subtree) page state & events */
	long			state[MEMCG_NR_STAT];
	unsigned long		events[NR_VM_EVENT_ITEMS];

	/* Pending child counts during tree propagation */
	long			state_pending[MEMCG_NR_STAT];
	unsigned long		events_pending[NR_VM_EVENT_ITEMS];
M
Michal Hocko 已提交
100 101 102 103 104 105 106 107
};

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

108 109 110 111
struct lruvec_stat {
	long count[NR_VM_NODE_STAT_ITEMS];
};

112 113 114 115
struct batched_lruvec_stat {
	s32 count[NR_VM_NODE_STAT_ITEMS];
};

116
/*
117 118
 * Bitmap and deferred work of shrinker::id corresponding to memcg-aware
 * shrinkers, which have elements charged to this memcg.
119
 */
120
struct shrinker_info {
121
	struct rcu_head rcu;
122 123
	atomic_long_t *nr_deferred;
	unsigned long *map;
124 125
};

M
Michal Hocko 已提交
126
/*
127
 * per-node information in memory controller.
M
Michal Hocko 已提交
128
 */
129
struct mem_cgroup_per_node {
M
Michal Hocko 已提交
130
	struct lruvec		lruvec;
131

132 133 134 135 136 137 138
	/*
	 * Legacy local VM stats. This should be struct lruvec_stat and
	 * cannot be optimized to struct batched_lruvec_stat. Because
	 * the threshold of the lruvec_stat_cpu can be as big as
	 * MEMCG_CHARGE_BATCH * PAGE_SIZE. It can fit into s32. But this
	 * filed has no upper limit.
	 */
139 140 141
	struct lruvec_stat __percpu *lruvec_stat_local;

	/* Subtree VM stats (batched updates) */
142
	struct batched_lruvec_stat __percpu *lruvec_stat_cpu;
143 144
	atomic_long_t		lruvec_stat[NR_VM_NODE_STAT_ITEMS];

145
	unsigned long		lru_zone_size[MAX_NR_ZONES][NR_LRU_LISTS];
M
Michal Hocko 已提交
146

147
	struct mem_cgroup_reclaim_iter	iter;
M
Michal Hocko 已提交
148

149
	struct shrinker_info __rcu	*shrinker_info;
150

M
Michal Hocko 已提交
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
	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 */
171
	struct mem_cgroup_threshold entries[];
M
Michal Hocko 已提交
172 173 174 175 176 177 178 179 180 181 182 183 184
};

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;
};

185 186 187 188 189 190
enum memcg_kmem_state {
	KMEM_NONE,
	KMEM_ALLOCATED,
	KMEM_ONLINE,
};

191 192 193 194 195 196 197 198 199
#if defined(CONFIG_SMP)
struct memcg_padding {
	char x[0];
} ____cacheline_internodealigned_in_smp;
#define MEMCG_PADDING(name)      struct memcg_padding name;
#else
#define MEMCG_PADDING(name)
#endif

200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216
/*
 * Remember four most recent foreign writebacks with dirty pages in this
 * cgroup.  Inode sharing is expected to be uncommon and, even if we miss
 * one in a given round, we're likely to catch it later if it keeps
 * foreign-dirtying, so a fairly low count should be enough.
 *
 * See mem_cgroup_track_foreign_dirty_slowpath() for details.
 */
#define MEMCG_CGWB_FRN_CNT	4

struct memcg_cgwb_frn {
	u64 bdi_id;			/* bdi->id of the foreign inode */
	int memcg_id;			/* memcg->css.id of foreign inode */
	u64 at;				/* jiffies_64 at the time of dirtying */
	struct wb_completion done;	/* tracks in-flight foreign writebacks */
};

R
Roman Gushchin 已提交
217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
/*
 * Bucket for arbitrarily byte-sized objects charged to a memory
 * cgroup. The bucket can be reparented in one piece when the cgroup
 * is destroyed, without having to round up the individual references
 * of all live memory objects in the wild.
 */
struct obj_cgroup {
	struct percpu_ref refcnt;
	struct mem_cgroup *memcg;
	atomic_t nr_charged_bytes;
	union {
		struct list_head list;
		struct rcu_head rcu;
	};
};

M
Michal Hocko 已提交
233 234 235 236 237 238 239 240 241
/*
 * 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;

242 243 244
	/* Private memcg ID. Used to ID objects that outlive the cgroup */
	struct mem_cgroup_id id;

M
Michal Hocko 已提交
245
	/* Accounted resources */
246 247 248 249 250 251
	struct page_counter memory;		/* Both v1 & v2 */

	union {
		struct page_counter swap;	/* v2 only */
		struct page_counter memsw;	/* v1 only */
	};
252 253

	/* Legacy consumer-oriented counters */
254 255
	struct page_counter kmem;		/* v1 only */
	struct page_counter tcpmem;		/* v1 only */
M
Michal Hocko 已提交
256

257 258 259
	/* Range enforcement for interrupt charges */
	struct work_struct high_work;

M
Michal Hocko 已提交
260 261 262 263 264
	unsigned long soft_limit;

	/* vmpressure notifications */
	struct vmpressure vmpressure;

265 266 267 268 269
	/*
	 * Should the OOM killer kill all belonging tasks, had it kill one?
	 */
	bool oom_group;

M
Michal Hocko 已提交
270 271 272 273 274 275 276 277
	/* protected by memcg_oom_lock */
	bool		oom_lock;
	int		under_oom;

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

278
	/* memory.events and memory.events.local */
279
	struct cgroup_file events_file;
280
	struct cgroup_file events_local_file;
281

282 283 284
	/* handle for "memory.swap.events" */
	struct cgroup_file swap_events_file;

M
Michal Hocko 已提交
285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301
	/* 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;
302 303 304 305 306 307
	/* taken only while moving_account > 0 */
	spinlock_t		move_lock;
	unsigned long		move_lock_flags;

	MEMCG_PADDING(_pad1_);

308 309
	/* memory.stat */
	struct memcg_vmstats	vmstats;
310

311
	/* memory.events */
312
	atomic_long_t		memory_events[MEMCG_NR_MEMORY_EVENTS];
313
	atomic_long_t		memory_events_local[MEMCG_NR_MEMORY_EVENTS];
M
Michal Hocko 已提交
314

315 316 317
	unsigned long		socket_pressure;

	/* Legacy tcp memory accounting */
318 319
	bool			tcpmem_active;
	int			tcpmem_pressure;
320

321
#ifdef CONFIG_MEMCG_KMEM
M
Michal Hocko 已提交
322
	int kmemcg_id;
323
	enum memcg_kmem_state kmem_state;
R
Roman Gushchin 已提交
324 325
	struct obj_cgroup __rcu *objcg;
	struct list_head objcg_list; /* list of inherited objcgs */
M
Michal Hocko 已提交
326 327
#endif

328 329 330 331 332 333 334 335 336 337
	MEMCG_PADDING(_pad2_);

	/*
	 * set > 0 if pages under this cgroup are moving to other cgroup.
	 */
	atomic_t		moving_account;
	struct task_struct	*move_lock_task;

	struct memcg_vmstats_percpu __percpu *vmstats_percpu;

M
Michal Hocko 已提交
338 339 340
#ifdef CONFIG_CGROUP_WRITEBACK
	struct list_head cgwb_list;
	struct wb_domain cgwb_domain;
341
	struct memcg_cgwb_frn cgwb_frn[MEMCG_CGWB_FRN_CNT];
M
Michal Hocko 已提交
342 343 344 345 346 347
#endif

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

348 349 350 351
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	struct deferred_split deferred_split_queue;
#endif

M
Michal Hocko 已提交
352 353 354
	struct mem_cgroup_per_node *nodeinfo[0];
	/* WARNING: nodeinfo must be the last member here */
};
355

356 357 358 359 360 361
/*
 * size of first charge trial. "32" comes from vmscan.c's magic value.
 * TODO: maybe necessary to use big numbers in big irons.
 */
#define MEMCG_CHARGE_BATCH 32U

362
extern struct mem_cgroup *root_mem_cgroup;
T
Tejun Heo 已提交
363

R
Roman Gushchin 已提交
364 365 366
enum page_memcg_data_flags {
	/* page->memcg_data is a pointer to an objcgs vector */
	MEMCG_DATA_OBJCGS = (1UL << 0),
367 368
	/* page has been accounted as a non-slab kernel page */
	MEMCG_DATA_KMEM = (1UL << 1),
R
Roman Gushchin 已提交
369
	/* the next bit after the last actual flag */
370
	__NR_MEMCG_DATA_FLAGS  = (1UL << 2),
R
Roman Gushchin 已提交
371 372 373 374
};

#define MEMCG_DATA_FLAGS_MASK (__NR_MEMCG_DATA_FLAGS - 1)

375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
static inline bool PageMemcgKmem(struct page *page);

/*
 * After the initialization objcg->memcg is always pointing at
 * a valid memcg, but can be atomically swapped to the parent memcg.
 *
 * The caller must ensure that the returned memcg won't be released:
 * e.g. acquire the rcu_read_lock or css_set_lock.
 */
static inline struct mem_cgroup *obj_cgroup_memcg(struct obj_cgroup *objcg)
{
	return READ_ONCE(objcg->memcg);
}

/*
 * __page_memcg - get the memory cgroup associated with a non-kmem page
 * @page: a pointer to the page struct
 *
 * Returns a pointer to the memory cgroup associated with the page,
 * or NULL. This function assumes that the page is known to have a
 * proper memory cgroup pointer. It's not safe to call this function
 * against some type of pages, e.g. slab pages or ex-slab pages or
 * kmem pages.
 */
static inline struct mem_cgroup *__page_memcg(struct page *page)
{
	unsigned long memcg_data = page->memcg_data;

	VM_BUG_ON_PAGE(PageSlab(page), page);
	VM_BUG_ON_PAGE(memcg_data & MEMCG_DATA_OBJCGS, page);
	VM_BUG_ON_PAGE(memcg_data & MEMCG_DATA_KMEM, page);

	return (struct mem_cgroup *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
}

/*
 * __page_objcg - get the object cgroup associated with a kmem page
 * @page: a pointer to the page struct
 *
 * Returns a pointer to the object cgroup associated with the page,
 * or NULL. This function assumes that the page is known to have a
 * proper object cgroup pointer. It's not safe to call this function
 * against some type of pages, e.g. slab pages or ex-slab pages or
 * LRU pages.
 */
static inline struct obj_cgroup *__page_objcg(struct page *page)
{
	unsigned long memcg_data = page->memcg_data;

	VM_BUG_ON_PAGE(PageSlab(page), page);
	VM_BUG_ON_PAGE(memcg_data & MEMCG_DATA_OBJCGS, page);
	VM_BUG_ON_PAGE(!(memcg_data & MEMCG_DATA_KMEM), page);

	return (struct obj_cgroup *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
}

431 432 433 434 435 436 437 438 439
/*
 * page_memcg - get the memory cgroup associated with a page
 * @page: a pointer to the page struct
 *
 * Returns a pointer to the memory cgroup associated with the page,
 * or NULL. This function assumes that the page is known to have a
 * proper memory cgroup pointer. It's not safe to call this function
 * against some type of pages, e.g. slab pages or ex-slab pages.
 *
440 441 442
 * For a non-kmem page any of the following ensures page and memcg binding
 * stability:
 *
443 444 445 446
 * - the page lock
 * - LRU isolation
 * - lock_page_memcg()
 * - exclusive reference
447 448 449
 *
 * For a kmem page a caller should hold an rcu read lock to protect memcg
 * associated with a kmem page from being released.
450 451 452
 */
static inline struct mem_cgroup *page_memcg(struct page *page)
{
453 454 455 456
	if (PageMemcgKmem(page))
		return obj_cgroup_memcg(__page_objcg(page));
	else
		return __page_memcg(page);
457 458 459 460 461 462 463 464 465 466 467 468 469
}

/*
 * page_memcg_rcu - locklessly get the memory cgroup associated with a page
 * @page: a pointer to the page struct
 *
 * Returns a pointer to the memory cgroup associated with the page,
 * or NULL. This function assumes that the page is known to have a
 * proper memory cgroup pointer. It's not safe to call this function
 * against some type of pages, e.g. slab pages or ex-slab pages.
 */
static inline struct mem_cgroup *page_memcg_rcu(struct page *page)
{
470 471
	unsigned long memcg_data = READ_ONCE(page->memcg_data);

472 473 474
	VM_BUG_ON_PAGE(PageSlab(page), page);
	WARN_ON_ONCE(!rcu_read_lock_held());

475 476 477 478 479 480 481 482
	if (memcg_data & MEMCG_DATA_KMEM) {
		struct obj_cgroup *objcg;

		objcg = (void *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
		return obj_cgroup_memcg(objcg);
	}

	return (struct mem_cgroup *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
483 484 485 486 487 488 489
}

/*
 * page_memcg_check - get the memory cgroup associated with a page
 * @page: a pointer to the page struct
 *
 * Returns a pointer to the memory cgroup associated with the page,
490
 * or NULL. This function unlike page_memcg() can take any page
491
 * as an argument. It has to be used in cases when it's not known if a page
492 493 494 495 496
 * has an associated memory cgroup pointer or an object cgroups vector or
 * an object cgroup.
 *
 * For a non-kmem page any of the following ensures page and memcg binding
 * stability:
497 498 499 500 501
 *
 * - the page lock
 * - LRU isolation
 * - lock_page_memcg()
 * - exclusive reference
502 503 504
 *
 * For a kmem page a caller should hold an rcu read lock to protect memcg
 * associated with a kmem page from being released.
505 506 507 508 509 510 511 512 513
 */
static inline struct mem_cgroup *page_memcg_check(struct page *page)
{
	/*
	 * Because page->memcg_data might be changed asynchronously
	 * for slab pages, READ_ONCE() should be used here.
	 */
	unsigned long memcg_data = READ_ONCE(page->memcg_data);

R
Roman Gushchin 已提交
514
	if (memcg_data & MEMCG_DATA_OBJCGS)
515 516
		return NULL;

517 518 519 520 521 522 523
	if (memcg_data & MEMCG_DATA_KMEM) {
		struct obj_cgroup *objcg;

		objcg = (void *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
		return obj_cgroup_memcg(objcg);
	}

524 525 526
	return (struct mem_cgroup *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
}

527
#ifdef CONFIG_MEMCG_KMEM
528 529 530 531 532 533 534 535 536 537 538 539
/*
 * PageMemcgKmem - check if the page has MemcgKmem flag set
 * @page: a pointer to the page struct
 *
 * Checks if the page has MemcgKmem flag set. The caller must ensure that
 * the page has an associated memory cgroup. It's not safe to call this function
 * against some types of pages, e.g. slab pages.
 */
static inline bool PageMemcgKmem(struct page *page)
{
	VM_BUG_ON_PAGE(page->memcg_data & MEMCG_DATA_OBJCGS, page);
	return page->memcg_data & MEMCG_DATA_KMEM;
540 541
}

542 543 544 545 546 547 548 549 550 551 552 553
/*
 * page_objcgs - get the object cgroups vector associated with a page
 * @page: a pointer to the page struct
 *
 * Returns a pointer to the object cgroups vector associated with the page,
 * or NULL. This function assumes that the page is known to have an
 * associated object cgroups vector. It's not safe to call this function
 * against pages, which might have an associated memory cgroup: e.g.
 * kernel stack pages.
 */
static inline struct obj_cgroup **page_objcgs(struct page *page)
{
R
Roman Gushchin 已提交
554 555 556
	unsigned long memcg_data = READ_ONCE(page->memcg_data);

	VM_BUG_ON_PAGE(memcg_data && !(memcg_data & MEMCG_DATA_OBJCGS), page);
557
	VM_BUG_ON_PAGE(memcg_data & MEMCG_DATA_KMEM, page);
R
Roman Gushchin 已提交
558 559

	return (struct obj_cgroup **)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
560 561 562 563 564 565 566 567 568 569 570 571 572 573
}

/*
 * page_objcgs_check - get the object cgroups vector associated with a page
 * @page: a pointer to the page struct
 *
 * Returns a pointer to the object cgroups vector associated with the page,
 * or NULL. This function is safe to use if the page can be directly associated
 * with a memory cgroup.
 */
static inline struct obj_cgroup **page_objcgs_check(struct page *page)
{
	unsigned long memcg_data = READ_ONCE(page->memcg_data);

R
Roman Gushchin 已提交
574 575
	if (!memcg_data || !(memcg_data & MEMCG_DATA_OBJCGS))
		return NULL;
576

577 578
	VM_BUG_ON_PAGE(memcg_data & MEMCG_DATA_KMEM, page);

R
Roman Gushchin 已提交
579
	return (struct obj_cgroup **)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
580 581 582
}

#else
583 584 585 586 587
static inline bool PageMemcgKmem(struct page *page)
{
	return false;
}

588 589 590 591 592 593 594 595 596 597 598
static inline struct obj_cgroup **page_objcgs(struct page *page)
{
	return NULL;
}

static inline struct obj_cgroup **page_objcgs_check(struct page *page)
{
	return NULL;
}
#endif

599 600 601 602 603 604 605
static __always_inline bool memcg_stat_item_in_bytes(int idx)
{
	if (idx == MEMCG_PERCPU_B)
		return true;
	return vmstat_item_in_bytes(idx);
}

606 607 608 609 610
static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
{
	return (memcg == root_mem_cgroup);
}

611 612 613 614 615
static inline bool mem_cgroup_disabled(void)
{
	return !cgroup_subsys_enabled(memory_cgrp_subsys);
}

616 617
static inline unsigned long mem_cgroup_protection(struct mem_cgroup *root,
						  struct mem_cgroup *memcg,
618
						  bool in_low_reclaim)
619
{
620 621 622
	if (mem_cgroup_disabled())
		return 0;

623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658
	/*
	 * There is no reclaim protection applied to a targeted reclaim.
	 * We are special casing this specific case here because
	 * mem_cgroup_protected calculation is not robust enough to keep
	 * the protection invariant for calculated effective values for
	 * parallel reclaimers with different reclaim target. This is
	 * especially a problem for tail memcgs (as they have pages on LRU)
	 * which would want to have effective values 0 for targeted reclaim
	 * but a different value for external reclaim.
	 *
	 * Example
	 * Let's have global and A's reclaim in parallel:
	 *  |
	 *  A (low=2G, usage = 3G, max = 3G, children_low_usage = 1.5G)
	 *  |\
	 *  | C (low = 1G, usage = 2.5G)
	 *  B (low = 1G, usage = 0.5G)
	 *
	 * For the global reclaim
	 * A.elow = A.low
	 * B.elow = min(B.usage, B.low) because children_low_usage <= A.elow
	 * C.elow = min(C.usage, C.low)
	 *
	 * With the effective values resetting we have A reclaim
	 * A.elow = 0
	 * B.elow = B.low
	 * C.elow = C.low
	 *
	 * If the global reclaim races with A's reclaim then
	 * B.elow = C.elow = 0 because children_low_usage > A.elow)
	 * is possible and reclaiming B would be violating the protection.
	 *
	 */
	if (root == memcg)
		return 0;

659 660
	if (in_low_reclaim)
		return READ_ONCE(memcg->memory.emin);
661

662 663
	return max(READ_ONCE(memcg->memory.emin),
		   READ_ONCE(memcg->memory.elow));
664 665
}

666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695
void mem_cgroup_calculate_protection(struct mem_cgroup *root,
				     struct mem_cgroup *memcg);

static inline bool mem_cgroup_supports_protection(struct mem_cgroup *memcg)
{
	/*
	 * The root memcg doesn't account charges, and doesn't support
	 * protection.
	 */
	return !mem_cgroup_disabled() && !mem_cgroup_is_root(memcg);

}

static inline bool mem_cgroup_below_low(struct mem_cgroup *memcg)
{
	if (!mem_cgroup_supports_protection(memcg))
		return false;

	return READ_ONCE(memcg->memory.elow) >=
		page_counter_read(&memcg->memory);
}

static inline bool mem_cgroup_below_min(struct mem_cgroup *memcg)
{
	if (!mem_cgroup_supports_protection(memcg))
		return false;

	return READ_ONCE(memcg->memory.emin) >=
		page_counter_read(&memcg->memory);
}
696

697
int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask);
698 699 700
int mem_cgroup_swapin_charge_page(struct page *page, struct mm_struct *mm,
				  gfp_t gfp, swp_entry_t entry);
void mem_cgroup_swapin_uncharge_swap(swp_entry_t entry);
701

702
void mem_cgroup_uncharge(struct page *page);
703
void mem_cgroup_uncharge_list(struct list_head *page_list);
704

705
void mem_cgroup_migrate(struct page *oldpage, struct page *newpage);
706

707
/**
708
 * mem_cgroup_lruvec - get the lru list vector for a memcg & node
709
 * @memcg: memcg of the wanted lruvec
710
 * @pgdat: pglist_data
711
 *
712
 * Returns the lru list vector holding pages for a given @memcg &
713
 * @pgdat combination. This can be the node lruvec, if the memory
714
 * controller is disabled.
715
 */
716 717
static inline struct lruvec *mem_cgroup_lruvec(struct mem_cgroup *memcg,
					       struct pglist_data *pgdat)
718
{
719
	struct mem_cgroup_per_node *mz;
720 721 722
	struct lruvec *lruvec;

	if (mem_cgroup_disabled()) {
723
		lruvec = &pgdat->__lruvec;
724 725 726
		goto out;
	}

727 728 729
	if (!memcg)
		memcg = root_mem_cgroup;

730
	mz = memcg->nodeinfo[pgdat->node_id];
731 732 733 734
	lruvec = &mz->lruvec;
out:
	/*
	 * Since a node can be onlined after the mem_cgroup was created,
M
Mel Gorman 已提交
735
	 * we have to be prepared to initialize lruvec->pgdat here;
736 737
	 * and if offlined then reonlined, we need to reinitialize it.
	 */
738 739
	if (unlikely(lruvec->pgdat != pgdat))
		lruvec->pgdat = pgdat;
740 741 742
	return lruvec;
}

743 744 745 746 747 748 749 750 751 752 753 754
/**
 * mem_cgroup_page_lruvec - return lruvec for isolating/putting an LRU page
 * @page: the page
 * @pgdat: pgdat of the page
 *
 * This function relies on page->mem_cgroup being stable.
 */
static inline struct lruvec *mem_cgroup_page_lruvec(struct page *page,
						struct pglist_data *pgdat)
{
	struct mem_cgroup *memcg = page_memcg(page);

755
	VM_WARN_ON_ONCE_PAGE(!memcg && !mem_cgroup_disabled(), page);
756 757
	return mem_cgroup_lruvec(memcg, pgdat);
}
758

759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774
static inline bool lruvec_holds_page_lru_lock(struct page *page,
					      struct lruvec *lruvec)
{
	pg_data_t *pgdat = page_pgdat(page);
	const struct mem_cgroup *memcg;
	struct mem_cgroup_per_node *mz;

	if (mem_cgroup_disabled())
		return lruvec == &pgdat->__lruvec;

	mz = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
	memcg = page_memcg(page) ? : root_mem_cgroup;

	return lruvec->pgdat == pgdat && mz->memcg == memcg;
}

775
struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p);
776

777 778
struct mem_cgroup *get_mem_cgroup_from_mm(struct mm_struct *mm);

779 780 781 782 783 784 785 786 787 788 789 790 791
struct lruvec *lock_page_lruvec(struct page *page);
struct lruvec *lock_page_lruvec_irq(struct page *page);
struct lruvec *lock_page_lruvec_irqsave(struct page *page,
						unsigned long *flags);

#ifdef CONFIG_DEBUG_VM
void lruvec_memcg_debug(struct lruvec *lruvec, struct page *page);
#else
static inline void lruvec_memcg_debug(struct lruvec *lruvec, struct page *page)
{
}
#endif

M
Michal Hocko 已提交
792 793 794 795 796
static inline
struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *css){
	return css ? container_of(css, struct mem_cgroup, css) : NULL;
}

R
Roman Gushchin 已提交
797 798 799 800 801 802 803 804 805 806
static inline bool obj_cgroup_tryget(struct obj_cgroup *objcg)
{
	return percpu_ref_tryget(&objcg->refcnt);
}

static inline void obj_cgroup_get(struct obj_cgroup *objcg)
{
	percpu_ref_get(&objcg->refcnt);
}

807 808
static inline void obj_cgroup_get_many(struct obj_cgroup *objcg,
				       unsigned long nr)
R
Roman Gushchin 已提交
809
{
810
	percpu_ref_get_many(&objcg->refcnt, nr);
R
Roman Gushchin 已提交
811 812
}

813
static inline void obj_cgroup_put(struct obj_cgroup *objcg)
R
Roman Gushchin 已提交
814
{
815
	percpu_ref_put(&objcg->refcnt);
R
Roman Gushchin 已提交
816 817
}

818 819
static inline void mem_cgroup_put(struct mem_cgroup *memcg)
{
820 821
	if (memcg)
		css_put(&memcg->css);
822 823
}

824 825 826
#define mem_cgroup_from_counter(counter, member)	\
	container_of(counter, struct mem_cgroup, member)

M
Michal Hocko 已提交
827 828 829 830
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 *);
831 832
int mem_cgroup_scan_tasks(struct mem_cgroup *,
			  int (*)(struct task_struct *, void *), void *);
M
Michal Hocko 已提交
833

834 835 836 837 838
static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
{
	if (mem_cgroup_disabled())
		return 0;

839
	return memcg->id.id;
840
}
841
struct mem_cgroup *mem_cgroup_from_id(unsigned short id);
842

843 844 845 846 847
static inline struct mem_cgroup *mem_cgroup_from_seq(struct seq_file *m)
{
	return mem_cgroup_from_css(seq_css(m));
}

848 849 850 851 852 853 854 855 856 857 858
static inline struct mem_cgroup *lruvec_memcg(struct lruvec *lruvec)
{
	struct mem_cgroup_per_node *mz;

	if (mem_cgroup_disabled())
		return NULL;

	mz = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
	return mz->memcg;
}

859 860 861 862 863 864 865 866 867 868 869 870 871 872
/**
 * 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
873 874 875 876 877 878 879
static inline bool mem_cgroup_is_descendant(struct mem_cgroup *memcg,
			      struct mem_cgroup *root)
{
	if (root == memcg)
		return true;
	return cgroup_is_descendant(memcg->css.cgroup, root->css.cgroup);
}
G
Glauber Costa 已提交
880

881 882
static inline bool mm_match_cgroup(struct mm_struct *mm,
				   struct mem_cgroup *memcg)
883
{
884
	struct mem_cgroup *task_memcg;
885
	bool match = false;
886

887
	rcu_read_lock();
888
	task_memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
889
	if (task_memcg)
890
		match = mem_cgroup_is_descendant(task_memcg, memcg);
891
	rcu_read_unlock();
892
	return match;
893
}
894

895
struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page);
896
ino_t page_cgroup_ino(struct page *page);
897

898 899 900 901 902 903 904
static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
{
	if (mem_cgroup_disabled())
		return true;
	return !!(memcg->css.flags & CSS_ONLINE);
}

905 906 907
/*
 * For memory reclaim.
 */
908
int mem_cgroup_select_victim_node(struct mem_cgroup *memcg);
M
Michal Hocko 已提交
909 910

void mem_cgroup_update_lru_size(struct lruvec *lruvec, enum lru_list lru,
911
		int zid, int nr_pages);
M
Michal Hocko 已提交
912

913 914 915 916 917 918 919
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);
920
	return READ_ONCE(mz->lru_zone_size[zone_idx][lru]);
M
Michal Hocko 已提交
921 922
}

923 924
void mem_cgroup_handle_over_high(void);

925
unsigned long mem_cgroup_get_max(struct mem_cgroup *memcg);
926

927 928
unsigned long mem_cgroup_size(struct mem_cgroup *memcg);

929
void mem_cgroup_print_oom_context(struct mem_cgroup *memcg,
930
				struct task_struct *p);
931

932 933
void mem_cgroup_print_oom_meminfo(struct mem_cgroup *memcg);

934
static inline void mem_cgroup_enter_user_fault(void)
935
{
936 937
	WARN_ON(current->in_user_fault);
	current->in_user_fault = 1;
938 939
}

940
static inline void mem_cgroup_exit_user_fault(void)
941
{
942 943
	WARN_ON(!current->in_user_fault);
	current->in_user_fault = 0;
944 945
}

946 947
static inline bool task_in_memcg_oom(struct task_struct *p)
{
T
Tejun Heo 已提交
948
	return p->memcg_in_oom;
949 950
}

951
bool mem_cgroup_oom_synchronize(bool wait);
952 953 954
struct mem_cgroup *mem_cgroup_get_oom_group(struct task_struct *victim,
					    struct mem_cgroup *oom_domain);
void mem_cgroup_print_oom_group(struct mem_cgroup *memcg);
955

A
Andrew Morton 已提交
956
#ifdef CONFIG_MEMCG_SWAP
957
extern bool cgroup_memory_noswap;
958
#endif
959

960
void lock_page_memcg(struct page *page);
J
Johannes Weiner 已提交
961
void unlock_page_memcg(struct page *page);
962

963
void __mod_memcg_state(struct mem_cgroup *memcg, int idx, int val);
964

965
/* idx can be of type enum memcg_stat_item or node_stat_item */
966
static inline void mod_memcg_state(struct mem_cgroup *memcg,
967
				   int idx, int val)
968
{
969 970 971
	unsigned long flags;

	local_irq_save(flags);
972
	__mod_memcg_state(memcg, idx, val);
973
	local_irq_restore(flags);
974 975
}

976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993
static inline unsigned long lruvec_page_state(struct lruvec *lruvec,
					      enum node_stat_item idx)
{
	struct mem_cgroup_per_node *pn;
	long x;

	if (mem_cgroup_disabled())
		return node_page_state(lruvec_pgdat(lruvec), idx);

	pn = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
	x = atomic_long_read(&pn->lruvec_stat[idx]);
#ifdef CONFIG_SMP
	if (x < 0)
		x = 0;
#endif
	return x;
}

994 995
static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
						    enum node_stat_item idx)
996
{
997
	struct mem_cgroup_per_node *pn;
998 999
	long x = 0;
	int cpu;
1000 1001 1002 1003 1004

	if (mem_cgroup_disabled())
		return node_page_state(lruvec_pgdat(lruvec), idx);

	pn = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
1005 1006
	for_each_possible_cpu(cpu)
		x += per_cpu(pn->lruvec_stat_local->count[idx], cpu);
1007 1008 1009 1010 1011
#ifdef CONFIG_SMP
	if (x < 0)
		x = 0;
#endif
	return x;
1012 1013
}

1014 1015
void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
			      int val);
1016
void __mod_lruvec_kmem_state(void *p, enum node_stat_item idx, int val);
1017

1018
static inline void mod_lruvec_kmem_state(void *p, enum node_stat_item idx,
1019 1020 1021 1022 1023
					 int val)
{
	unsigned long flags;

	local_irq_save(flags);
1024
	__mod_lruvec_kmem_state(p, idx, val);
1025 1026 1027
	local_irq_restore(flags);
}

1028 1029 1030 1031 1032 1033 1034 1035 1036 1037
static inline void mod_memcg_lruvec_state(struct lruvec *lruvec,
					  enum node_stat_item idx, int val)
{
	unsigned long flags;

	local_irq_save(flags);
	__mod_memcg_lruvec_state(lruvec, idx, val);
	local_irq_restore(flags);
}

1038 1039
void __count_memcg_events(struct mem_cgroup *memcg, enum vm_event_item idx,
			  unsigned long count);
1040

1041
static inline void count_memcg_events(struct mem_cgroup *memcg,
1042 1043
				      enum vm_event_item idx,
				      unsigned long count)
1044
{
1045 1046 1047
	unsigned long flags;

	local_irq_save(flags);
1048
	__count_memcg_events(memcg, idx, count);
1049
	local_irq_restore(flags);
1050 1051 1052
}

static inline void count_memcg_page_event(struct page *page,
1053
					  enum vm_event_item idx)
1054
{
1055 1056 1057 1058
	struct mem_cgroup *memcg = page_memcg(page);

	if (memcg)
		count_memcg_events(memcg, idx, 1);
1059 1060 1061 1062
}

static inline void count_memcg_event_mm(struct mm_struct *mm,
					enum vm_event_item idx)
1063
{
M
Michal Hocko 已提交
1064 1065
	struct mem_cgroup *memcg;

1066 1067
	if (mem_cgroup_disabled())
		return;
M
Michal Hocko 已提交
1068 1069 1070

	rcu_read_lock();
	memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
R
Roman Gushchin 已提交
1071
	if (likely(memcg))
1072
		count_memcg_events(memcg, idx, 1);
M
Michal Hocko 已提交
1073
	rcu_read_unlock();
1074
}
1075

1076 1077
static inline void memcg_memory_event(struct mem_cgroup *memcg,
				      enum memcg_memory_event event)
1078
{
1079 1080 1081
	bool swap_event = event == MEMCG_SWAP_HIGH || event == MEMCG_SWAP_MAX ||
			  event == MEMCG_SWAP_FAIL;

1082
	atomic_long_inc(&memcg->memory_events_local[event]);
1083 1084
	if (!swap_event)
		cgroup_file_notify(&memcg->events_local_file);
1085

1086 1087
	do {
		atomic_long_inc(&memcg->memory_events[event]);
1088 1089 1090 1091
		if (swap_event)
			cgroup_file_notify(&memcg->swap_events_file);
		else
			cgroup_file_notify(&memcg->events_file);
1092

1093 1094
		if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
			break;
1095 1096 1097 1098
		if (cgrp_dfl_root.flags & CGRP_ROOT_MEMORY_LOCAL_EVENTS)
			break;
	} while ((memcg = parent_mem_cgroup(memcg)) &&
		 !mem_cgroup_is_root(memcg));
1099 1100
}

R
Roman Gushchin 已提交
1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115
static inline void memcg_memory_event_mm(struct mm_struct *mm,
					 enum memcg_memory_event event)
{
	struct mem_cgroup *memcg;

	if (mem_cgroup_disabled())
		return;

	rcu_read_lock();
	memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
	if (likely(memcg))
		memcg_memory_event(memcg, event);
	rcu_read_unlock();
}

1116
void split_page_memcg(struct page *head, unsigned int nr);
1117

1118 1119 1120 1121
unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
						gfp_t gfp_mask,
						unsigned long *total_scanned);

A
Andrew Morton 已提交
1122
#else /* CONFIG_MEMCG */
1123 1124 1125 1126

#define MEM_CGROUP_ID_SHIFT	0
#define MEM_CGROUP_ID_MAX	0

1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142
static inline struct mem_cgroup *page_memcg(struct page *page)
{
	return NULL;
}

static inline struct mem_cgroup *page_memcg_rcu(struct page *page)
{
	WARN_ON_ONCE(!rcu_read_lock_held());
	return NULL;
}

static inline struct mem_cgroup *page_memcg_check(struct page *page)
{
	return NULL;
}

1143 1144 1145 1146 1147
static inline bool PageMemcgKmem(struct page *page)
{
	return false;
}

1148 1149 1150 1151 1152
static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
{
	return true;
}

1153 1154 1155 1156 1157
static inline bool mem_cgroup_disabled(void)
{
	return true;
}

1158 1159
static inline void memcg_memory_event(struct mem_cgroup *memcg,
				      enum memcg_memory_event event)
1160 1161 1162
{
}

R
Roman Gushchin 已提交
1163 1164 1165 1166 1167
static inline void memcg_memory_event_mm(struct mm_struct *mm,
					 enum memcg_memory_event event)
{
}

1168 1169
static inline unsigned long mem_cgroup_protection(struct mem_cgroup *root,
						  struct mem_cgroup *memcg,
1170
						  bool in_low_reclaim)
1171
{
1172
	return 0;
1173 1174
}

1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185
static inline void mem_cgroup_calculate_protection(struct mem_cgroup *root,
						   struct mem_cgroup *memcg)
{
}

static inline bool mem_cgroup_below_low(struct mem_cgroup *memcg)
{
	return false;
}

static inline bool mem_cgroup_below_min(struct mem_cgroup *memcg)
1186
{
1187
	return false;
1188 1189
}

1190
static inline int mem_cgroup_charge(struct page *page, struct mm_struct *mm,
1191
				    gfp_t gfp_mask)
1192 1193 1194 1195
{
	return 0;
}

1196 1197 1198 1199 1200 1201 1202 1203 1204 1205
static inline int mem_cgroup_swapin_charge_page(struct page *page,
			struct mm_struct *mm, gfp_t gfp, swp_entry_t entry)
{
	return 0;
}

static inline void mem_cgroup_swapin_uncharge_swap(swp_entry_t entry)
{
}

1206
static inline void mem_cgroup_uncharge(struct page *page)
1207 1208 1209
{
}

1210
static inline void mem_cgroup_uncharge_list(struct list_head *page_list)
1211 1212 1213
{
}

1214
static inline void mem_cgroup_migrate(struct page *old, struct page *new)
1215 1216 1217
{
}

1218 1219
static inline struct lruvec *mem_cgroup_lruvec(struct mem_cgroup *memcg,
					       struct pglist_data *pgdat)
K
KAMEZAWA Hiroyuki 已提交
1220
{
1221
	return &pgdat->__lruvec;
K
KAMEZAWA Hiroyuki 已提交
1222 1223
}

1224
static inline struct lruvec *mem_cgroup_page_lruvec(struct page *page,
M
Mel Gorman 已提交
1225
						    struct pglist_data *pgdat)
1226
{
1227
	return &pgdat->__lruvec;
1228 1229
}

1230 1231 1232 1233 1234 1235 1236 1237
static inline bool lruvec_holds_page_lru_lock(struct page *page,
					      struct lruvec *lruvec)
{
	pg_data_t *pgdat = page_pgdat(page);

	return lruvec == &pgdat->__lruvec;
}

1238 1239 1240 1241
static inline void lruvec_memcg_debug(struct lruvec *lruvec, struct page *page)
{
}

1242 1243 1244 1245 1246
static inline struct mem_cgroup *parent_mem_cgroup(struct mem_cgroup *memcg)
{
	return NULL;
}

1247
static inline bool mm_match_cgroup(struct mm_struct *mm,
1248
		struct mem_cgroup *memcg)
1249
{
1250
	return true;
1251 1252
}

1253 1254 1255 1256 1257
static inline struct mem_cgroup *get_mem_cgroup_from_mm(struct mm_struct *mm)
{
	return NULL;
}

1258 1259 1260 1261
static inline void mem_cgroup_put(struct mem_cgroup *memcg)
{
}

1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286
static inline struct lruvec *lock_page_lruvec(struct page *page)
{
	struct pglist_data *pgdat = page_pgdat(page);

	spin_lock(&pgdat->__lruvec.lru_lock);
	return &pgdat->__lruvec;
}

static inline struct lruvec *lock_page_lruvec_irq(struct page *page)
{
	struct pglist_data *pgdat = page_pgdat(page);

	spin_lock_irq(&pgdat->__lruvec.lru_lock);
	return &pgdat->__lruvec;
}

static inline struct lruvec *lock_page_lruvec_irqsave(struct page *page,
		unsigned long *flagsp)
{
	struct pglist_data *pgdat = page_pgdat(page);

	spin_lock_irqsave(&pgdat->__lruvec.lru_lock, *flagsp);
	return &pgdat->__lruvec;
}

1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299
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)
{
}

1300 1301 1302 1303 1304 1305
static inline int mem_cgroup_scan_tasks(struct mem_cgroup *memcg,
		int (*fn)(struct task_struct *, void *), void *arg)
{
	return 0;
}

1306
static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
1307
{
1308 1309 1310 1311 1312 1313 1314 1315
	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;
1316
}
1317

1318 1319 1320 1321 1322
static inline struct mem_cgroup *mem_cgroup_from_seq(struct seq_file *m)
{
	return NULL;
}

1323 1324 1325 1326 1327
static inline struct mem_cgroup *lruvec_memcg(struct lruvec *lruvec)
{
	return NULL;
}

1328
static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
1329
{
1330
	return true;
1331 1332
}

1333 1334 1335 1336 1337 1338
static inline
unsigned long mem_cgroup_get_zone_lru_size(struct lruvec *lruvec,
		enum lru_list lru, int zone_idx)
{
	return 0;
}
1339

1340
static inline unsigned long mem_cgroup_get_max(struct mem_cgroup *memcg)
1341 1342 1343 1344
{
	return 0;
}

1345 1346 1347 1348 1349
static inline unsigned long mem_cgroup_size(struct mem_cgroup *memcg)
{
	return 0;
}

1350
static inline void
1351 1352 1353 1354 1355 1356
mem_cgroup_print_oom_context(struct mem_cgroup *memcg, struct task_struct *p)
{
}

static inline void
mem_cgroup_print_oom_meminfo(struct mem_cgroup *memcg)
1357 1358 1359
{
}

1360
static inline void lock_page_memcg(struct page *page)
1361 1362 1363
{
}

J
Johannes Weiner 已提交
1364
static inline void unlock_page_memcg(struct page *page)
1365 1366 1367
{
}

1368 1369 1370 1371
static inline void mem_cgroup_handle_over_high(void)
{
}

1372
static inline void mem_cgroup_enter_user_fault(void)
1373 1374 1375
{
}

1376
static inline void mem_cgroup_exit_user_fault(void)
1377 1378 1379
{
}

1380 1381 1382 1383 1384
static inline bool task_in_memcg_oom(struct task_struct *p)
{
	return false;
}

1385
static inline bool mem_cgroup_oom_synchronize(bool wait)
1386 1387 1388 1389
{
	return false;
}

1390 1391 1392 1393 1394 1395 1396 1397 1398 1399
static inline struct mem_cgroup *mem_cgroup_get_oom_group(
	struct task_struct *victim, struct mem_cgroup *oom_domain)
{
	return NULL;
}

static inline void mem_cgroup_print_oom_group(struct mem_cgroup *memcg)
{
}

1400
static inline void __mod_memcg_state(struct mem_cgroup *memcg,
1401
				     int idx,
1402
				     int nr)
1403 1404 1405
{
}

1406
static inline void mod_memcg_state(struct mem_cgroup *memcg,
1407
				   int idx,
1408
				   int nr)
1409 1410 1411
{
}

1412 1413 1414 1415 1416 1417
static inline unsigned long lruvec_page_state(struct lruvec *lruvec,
					      enum node_stat_item idx)
{
	return node_page_state(lruvec_pgdat(lruvec), idx);
}

1418 1419
static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
						    enum node_stat_item idx)
1420
{
1421
	return node_page_state(lruvec_pgdat(lruvec), idx);
1422 1423
}

1424 1425 1426 1427 1428
static inline void __mod_memcg_lruvec_state(struct lruvec *lruvec,
					    enum node_stat_item idx, int val)
{
}

1429
static inline void __mod_lruvec_kmem_state(void *p, enum node_stat_item idx,
1430 1431 1432 1433 1434 1435 1436
					   int val)
{
	struct page *page = virt_to_head_page(p);

	__mod_node_page_state(page_pgdat(page), idx, val);
}

1437
static inline void mod_lruvec_kmem_state(void *p, enum node_stat_item idx,
1438 1439 1440 1441 1442 1443 1444
					 int val)
{
	struct page *page = virt_to_head_page(p);

	mod_node_page_state(page_pgdat(page), idx, val);
}

1445 1446 1447 1448 1449 1450
static inline void count_memcg_events(struct mem_cgroup *memcg,
				      enum vm_event_item idx,
				      unsigned long count)
{
}

1451 1452 1453 1454 1455 1456
static inline void __count_memcg_events(struct mem_cgroup *memcg,
					enum vm_event_item idx,
					unsigned long count)
{
}

1457
static inline void count_memcg_page_event(struct page *page,
1458
					  int idx)
1459 1460 1461
{
}

1462
static inline
1463
void count_memcg_event_mm(struct mm_struct *mm, enum vm_event_item idx)
1464 1465
{
}
1466

1467 1468 1469 1470 1471 1472 1473 1474
static inline void split_page_memcg(struct page *head, unsigned int nr)
{
}

static inline
unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
					    gfp_t gfp_mask,
					    unsigned long *total_scanned)
1475
{
1476
	return 0;
1477
}
A
Andrew Morton 已提交
1478
#endif /* CONFIG_MEMCG */
1479

1480
static inline void __inc_lruvec_kmem_state(void *p, enum node_stat_item idx)
1481
{
1482
	__mod_lruvec_kmem_state(p, idx, 1);
1483 1484
}

1485
static inline void __dec_lruvec_kmem_state(void *p, enum node_stat_item idx)
1486
{
1487
	__mod_lruvec_kmem_state(p, idx, -1);
1488 1489
}

1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502
static inline struct lruvec *parent_lruvec(struct lruvec *lruvec)
{
	struct mem_cgroup *memcg;

	memcg = lruvec_memcg(lruvec);
	if (!memcg)
		return NULL;
	memcg = parent_mem_cgroup(memcg);
	if (!memcg)
		return NULL;
	return mem_cgroup_lruvec(memcg, lruvec_pgdat(lruvec));
}

1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518
static inline void unlock_page_lruvec(struct lruvec *lruvec)
{
	spin_unlock(&lruvec->lru_lock);
}

static inline void unlock_page_lruvec_irq(struct lruvec *lruvec)
{
	spin_unlock_irq(&lruvec->lru_lock);
}

static inline void unlock_page_lruvec_irqrestore(struct lruvec *lruvec,
		unsigned long flags)
{
	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
}

1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546
/* Don't lock again iff page's lruvec locked */
static inline struct lruvec *relock_page_lruvec_irq(struct page *page,
		struct lruvec *locked_lruvec)
{
	if (locked_lruvec) {
		if (lruvec_holds_page_lru_lock(page, locked_lruvec))
			return locked_lruvec;

		unlock_page_lruvec_irq(locked_lruvec);
	}

	return lock_page_lruvec_irq(page);
}

/* Don't lock again iff page's lruvec locked */
static inline struct lruvec *relock_page_lruvec_irqsave(struct page *page,
		struct lruvec *locked_lruvec, unsigned long *flags)
{
	if (locked_lruvec) {
		if (lruvec_holds_page_lru_lock(page, locked_lruvec))
			return locked_lruvec;

		unlock_page_lruvec_irqrestore(locked_lruvec, *flags);
	}

	return lock_page_lruvec_irqsave(page, flags);
}

1547
#ifdef CONFIG_CGROUP_WRITEBACK
T
Tejun Heo 已提交
1548 1549

struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb);
1550 1551 1552
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
1553

1554 1555 1556 1557 1558 1559
void mem_cgroup_track_foreign_dirty_slowpath(struct page *page,
					     struct bdi_writeback *wb);

static inline void mem_cgroup_track_foreign_dirty(struct page *page,
						  struct bdi_writeback *wb)
{
1560 1561 1562
	if (mem_cgroup_disabled())
		return;

1563
	if (unlikely(&page_memcg(page)->css != wb->memcg_css))
1564 1565 1566 1567 1568
		mem_cgroup_track_foreign_dirty_slowpath(page, wb);
}

void mem_cgroup_flush_foreign(struct bdi_writeback *wb);

T
Tejun Heo 已提交
1569 1570 1571 1572 1573 1574 1575
#else	/* CONFIG_CGROUP_WRITEBACK */

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

1576
static inline void mem_cgroup_wb_stats(struct bdi_writeback *wb,
1577 1578
				       unsigned long *pfilepages,
				       unsigned long *pheadroom,
1579 1580 1581 1582 1583
				       unsigned long *pdirty,
				       unsigned long *pwriteback)
{
}

1584 1585 1586 1587 1588 1589 1590 1591 1592
static inline void mem_cgroup_track_foreign_dirty(struct page *page,
						  struct bdi_writeback *wb)
{
}

static inline void mem_cgroup_flush_foreign(struct bdi_writeback *wb)
{
}

T
Tejun Heo 已提交
1593
#endif	/* CONFIG_CGROUP_WRITEBACK */
1594

G
Glauber Costa 已提交
1595
struct sock;
1596 1597
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);
1598
#ifdef CONFIG_MEMCG
1599 1600
extern struct static_key_false memcg_sockets_enabled_key;
#define mem_cgroup_sockets_enabled static_branch_unlikely(&memcg_sockets_enabled_key)
1601 1602
void mem_cgroup_sk_alloc(struct sock *sk);
void mem_cgroup_sk_free(struct sock *sk);
1603
static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg)
1604
{
1605
	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && memcg->tcpmem_pressure)
1606 1607 1608 1609 1610 1611
		return true;
	do {
		if (time_before(jiffies, memcg->socket_pressure))
			return true;
	} while ((memcg = parent_mem_cgroup(memcg)));
	return false;
1612
}
1613

1614 1615
int alloc_shrinker_info(struct mem_cgroup *memcg);
void free_shrinker_info(struct mem_cgroup *memcg);
1616
void set_shrinker_bit(struct mem_cgroup *memcg, int nid, int shrinker_id);
1617
void reparent_shrinker_deferred(struct mem_cgroup *memcg);
1618
#else
1619
#define mem_cgroup_sockets_enabled 0
1620 1621
static inline void mem_cgroup_sk_alloc(struct sock *sk) { };
static inline void mem_cgroup_sk_free(struct sock *sk) { };
1622
static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg)
1623 1624 1625
{
	return false;
}
1626

1627 1628
static inline void set_shrinker_bit(struct mem_cgroup *memcg,
				    int nid, int shrinker_id)
1629 1630
{
}
1631
#endif
1632

1633
#ifdef CONFIG_MEMCG_KMEM
1634 1635
int __memcg_kmem_charge_page(struct page *page, gfp_t gfp, int order);
void __memcg_kmem_uncharge_page(struct page *page, int order);
1636

R
Roman Gushchin 已提交
1637 1638 1639 1640 1641
struct obj_cgroup *get_obj_cgroup_from_current(void);

int obj_cgroup_charge(struct obj_cgroup *objcg, gfp_t gfp, size_t size);
void obj_cgroup_uncharge(struct obj_cgroup *objcg, size_t size);

1642
extern struct static_key_false memcg_kmem_enabled_key;
1643

1644
extern int memcg_nr_cache_ids;
1645 1646
void memcg_get_cache_ids(void);
void memcg_put_cache_ids(void);
1647 1648 1649 1650 1651 1652

/*
 * 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
 */
1653
#define for_each_memcg_cache_index(_idx)	\
1654
	for ((_idx) = 0; (_idx) < memcg_nr_cache_ids; (_idx)++)
1655

1656 1657
static inline bool memcg_kmem_enabled(void)
{
1658
	return static_branch_likely(&memcg_kmem_enabled_key);
1659 1660
}

1661 1662
static inline int memcg_kmem_charge_page(struct page *page, gfp_t gfp,
					 int order)
1663 1664
{
	if (memcg_kmem_enabled())
1665
		return __memcg_kmem_charge_page(page, gfp, order);
1666 1667 1668
	return 0;
}

1669
static inline void memcg_kmem_uncharge_page(struct page *page, int order)
1670 1671
{
	if (memcg_kmem_enabled())
1672
		__memcg_kmem_uncharge_page(page, order);
1673 1674
}

M
Michal Hocko 已提交
1675
/*
1676 1677
 * A helper for accessing memcg's kmem_id, used for getting
 * corresponding LRU lists.
M
Michal Hocko 已提交
1678 1679 1680 1681 1682
 */
static inline int memcg_cache_id(struct mem_cgroup *memcg)
{
	return memcg ? memcg->kmemcg_id : -1;
}
1683

1684 1685
struct mem_cgroup *mem_cgroup_from_obj(void *p);

1686
#else
1687

1688 1689
static inline int memcg_kmem_charge_page(struct page *page, gfp_t gfp,
					 int order)
1690 1691 1692 1693
{
	return 0;
}

1694
static inline void memcg_kmem_uncharge_page(struct page *page, int order)
1695 1696 1697
{
}

1698 1699
static inline int __memcg_kmem_charge_page(struct page *page, gfp_t gfp,
					   int order)
1700 1701 1702 1703
{
	return 0;
}

1704
static inline void __memcg_kmem_uncharge_page(struct page *page, int order)
1705 1706 1707
{
}

1708 1709 1710
#define for_each_memcg_cache_index(_idx)	\
	for (; NULL; )

1711 1712 1713 1714 1715
static inline bool memcg_kmem_enabled(void)
{
	return false;
}

1716 1717 1718 1719 1720
static inline int memcg_cache_id(struct mem_cgroup *memcg)
{
	return -1;
}

1721 1722 1723 1724 1725 1726 1727 1728
static inline void memcg_get_cache_ids(void)
{
}

static inline void memcg_put_cache_ids(void)
{
}

1729 1730 1731 1732 1733
static inline struct mem_cgroup *mem_cgroup_from_obj(void *p)
{
       return NULL;
}

1734
#endif /* CONFIG_MEMCG_KMEM */
1735

B
Balbir Singh 已提交
1736
#endif /* _LINUX_MEMCONTROL_H */