memcontrol.h 42.9 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];
};

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

M
Michal Hocko 已提交
125
/*
126
 * per-node information in memory controller.
M
Michal Hocko 已提交
127
 */
128
struct mem_cgroup_per_node {
M
Michal Hocko 已提交
129
	struct lruvec		lruvec;
130

131 132 133 134 135 136 137
	/*
	 * 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.
	 */
138 139 140
	struct lruvec_stat __percpu *lruvec_stat_local;

	/* Subtree VM stats (batched updates) */
141
	struct batched_lruvec_stat __percpu *lruvec_stat_cpu;
142 143
	atomic_long_t		lruvec_stat[NR_VM_NODE_STAT_ITEMS];

144
	unsigned long		lru_zone_size[MAX_NR_ZONES][NR_LRU_LISTS];
M
Michal Hocko 已提交
145

146
	struct mem_cgroup_reclaim_iter	iter;
M
Michal Hocko 已提交
147

148
	struct shrinker_info __rcu	*shrinker_info;
149

M
Michal Hocko 已提交
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
	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 */
170
	struct mem_cgroup_threshold entries[];
M
Michal Hocko 已提交
171 172 173 174 175 176 177 178 179 180 181 182 183
};

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;
};

184 185 186 187 188 189
enum memcg_kmem_state {
	KMEM_NONE,
	KMEM_ALLOCATED,
	KMEM_ONLINE,
};

190 191 192 193 194 195 196 197 198
#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

199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
/*
 * 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
216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
/*
 * 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
232 233 234 235 236 237 238 239 240
/*
 * 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;

241 242 243
	/* Private memcg ID. Used to ID objects that outlive the cgroup */
	struct mem_cgroup_id id;

M
Michal Hocko 已提交
244
	/* Accounted resources */
245 246 247 248 249 250
	struct page_counter memory;		/* Both v1 & v2 */

	union {
		struct page_counter swap;	/* v2 only */
		struct page_counter memsw;	/* v1 only */
	};
251 252

	/* Legacy consumer-oriented counters */
253 254
	struct page_counter kmem;		/* v1 only */
	struct page_counter tcpmem;		/* v1 only */
M
Michal Hocko 已提交
255

256 257 258
	/* Range enforcement for interrupt charges */
	struct work_struct high_work;

M
Michal Hocko 已提交
259 260 261 262 263
	unsigned long soft_limit;

	/* vmpressure notifications */
	struct vmpressure vmpressure;

264 265 266 267 268
	/*
	 * Should the OOM killer kill all belonging tasks, had it kill one?
	 */
	bool oom_group;

M
Michal Hocko 已提交
269 270 271 272 273 274 275 276
	/* protected by memcg_oom_lock */
	bool		oom_lock;
	int		under_oom;

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

277
	/* memory.events and memory.events.local */
278
	struct cgroup_file events_file;
279
	struct cgroup_file events_local_file;
280

281 282 283
	/* handle for "memory.swap.events" */
	struct cgroup_file swap_events_file;

M
Michal Hocko 已提交
284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300
	/* 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;
301 302 303 304 305 306
	/* taken only while moving_account > 0 */
	spinlock_t		move_lock;
	unsigned long		move_lock_flags;

	MEMCG_PADDING(_pad1_);

307 308
	/* memory.stat */
	struct memcg_vmstats	vmstats;
309

310
	/* memory.events */
311
	atomic_long_t		memory_events[MEMCG_NR_MEMORY_EVENTS];
312
	atomic_long_t		memory_events_local[MEMCG_NR_MEMORY_EVENTS];
M
Michal Hocko 已提交
313

314 315 316
	unsigned long		socket_pressure;

	/* Legacy tcp memory accounting */
317 318
	bool			tcpmem_active;
	int			tcpmem_pressure;
319

320
#ifdef CONFIG_MEMCG_KMEM
M
Michal Hocko 已提交
321
	int kmemcg_id;
322
	enum memcg_kmem_state kmem_state;
R
Roman Gushchin 已提交
323 324
	struct obj_cgroup __rcu *objcg;
	struct list_head objcg_list; /* list of inherited objcgs */
M
Michal Hocko 已提交
325 326
#endif

327 328 329 330 331 332 333 334 335 336
	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
337 338 339
#ifdef CONFIG_CGROUP_WRITEBACK
	struct list_head cgwb_list;
	struct wb_domain cgwb_domain;
340
	struct memcg_cgwb_frn cgwb_frn[MEMCG_CGWB_FRN_CNT];
M
Michal Hocko 已提交
341 342 343 344 345 346
#endif

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

347 348 349 350
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	struct deferred_split deferred_split_queue;
#endif

M
Michal Hocko 已提交
351 352 353
	struct mem_cgroup_per_node *nodeinfo[0];
	/* WARNING: nodeinfo must be the last member here */
};
354

355 356 357 358 359 360
/*
 * 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

361
extern struct mem_cgroup *root_mem_cgroup;
T
Tejun Heo 已提交
362

R
Roman Gushchin 已提交
363 364 365
enum page_memcg_data_flags {
	/* page->memcg_data is a pointer to an objcgs vector */
	MEMCG_DATA_OBJCGS = (1UL << 0),
366 367
	/* page has been accounted as a non-slab kernel page */
	MEMCG_DATA_KMEM = (1UL << 1),
R
Roman Gushchin 已提交
368
	/* the next bit after the last actual flag */
369
	__NR_MEMCG_DATA_FLAGS  = (1UL << 2),
R
Roman Gushchin 已提交
370 371 372 373
};

#define MEMCG_DATA_FLAGS_MASK (__NR_MEMCG_DATA_FLAGS - 1)

374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
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);
}

430 431 432 433 434 435 436 437 438
/*
 * 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.
 *
439 440 441
 * For a non-kmem page any of the following ensures page and memcg binding
 * stability:
 *
442 443 444 445
 * - the page lock
 * - LRU isolation
 * - lock_page_memcg()
 * - exclusive reference
446 447 448
 *
 * For a kmem page a caller should hold an rcu read lock to protect memcg
 * associated with a kmem page from being released.
449 450 451
 */
static inline struct mem_cgroup *page_memcg(struct page *page)
{
452 453 454 455
	if (PageMemcgKmem(page))
		return obj_cgroup_memcg(__page_objcg(page));
	else
		return __page_memcg(page);
456 457 458 459 460 461 462 463 464 465 466 467 468
}

/*
 * 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)
{
469 470
	unsigned long memcg_data = READ_ONCE(page->memcg_data);

471 472 473
	VM_BUG_ON_PAGE(PageSlab(page), page);
	WARN_ON_ONCE(!rcu_read_lock_held());

474 475 476 477 478 479 480 481
	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);
482 483 484 485 486 487 488
}

/*
 * 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,
489
 * or NULL. This function unlike page_memcg() can take any page
490
 * as an argument. It has to be used in cases when it's not known if a page
491 492 493 494 495
 * 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:
496 497 498 499 500
 *
 * - the page lock
 * - LRU isolation
 * - lock_page_memcg()
 * - exclusive reference
501 502 503
 *
 * For a kmem page a caller should hold an rcu read lock to protect memcg
 * associated with a kmem page from being released.
504 505 506 507 508 509 510 511 512
 */
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
513
	if (memcg_data & MEMCG_DATA_OBJCGS)
514 515
		return NULL;

516 517 518 519 520 521 522
	if (memcg_data & MEMCG_DATA_KMEM) {
		struct obj_cgroup *objcg;

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

523 524 525
	return (struct mem_cgroup *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
}

526
#ifdef CONFIG_MEMCG_KMEM
527 528 529 530 531 532 533 534 535 536 537 538
/*
 * 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;
539 540
}

541 542 543 544 545 546 547 548 549 550 551 552
/*
 * 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
553 554 555
	unsigned long memcg_data = READ_ONCE(page->memcg_data);

	VM_BUG_ON_PAGE(memcg_data && !(memcg_data & MEMCG_DATA_OBJCGS), page);
556
	VM_BUG_ON_PAGE(memcg_data & MEMCG_DATA_KMEM, page);
R
Roman Gushchin 已提交
557 558

	return (struct obj_cgroup **)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
559 560 561 562 563 564 565 566 567 568 569 570 571 572
}

/*
 * 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
573 574
	if (!memcg_data || !(memcg_data & MEMCG_DATA_OBJCGS))
		return NULL;
575

576 577
	VM_BUG_ON_PAGE(memcg_data & MEMCG_DATA_KMEM, page);

R
Roman Gushchin 已提交
578
	return (struct obj_cgroup **)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
579 580 581
}

#else
582 583 584 585 586
static inline bool PageMemcgKmem(struct page *page)
{
	return false;
}

587 588 589 590 591 592 593 594 595 596 597
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

598 599 600 601 602 603 604
static __always_inline bool memcg_stat_item_in_bytes(int idx)
{
	if (idx == MEMCG_PERCPU_B)
		return true;
	return vmstat_item_in_bytes(idx);
}

605 606 607 608 609
static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
{
	return (memcg == root_mem_cgroup);
}

610 611 612 613 614
static inline bool mem_cgroup_disabled(void)
{
	return !cgroup_subsys_enabled(memory_cgrp_subsys);
}

615 616
static inline unsigned long mem_cgroup_protection(struct mem_cgroup *root,
						  struct mem_cgroup *memcg,
617
						  bool in_low_reclaim)
618
{
619 620 621
	if (mem_cgroup_disabled())
		return 0;

622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657
	/*
	 * 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;

658 659
	if (in_low_reclaim)
		return READ_ONCE(memcg->memory.emin);
660

661 662
	return max(READ_ONCE(memcg->memory.emin),
		   READ_ONCE(memcg->memory.elow));
663 664
}

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
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);
}
695

696
int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask);
697 698 699
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);
700

701
void mem_cgroup_uncharge(struct page *page);
702
void mem_cgroup_uncharge_list(struct list_head *page_list);
703

704
void mem_cgroup_migrate(struct page *oldpage, struct page *newpage);
705

706
/**
707
 * mem_cgroup_lruvec - get the lru list vector for a memcg & node
708
 * @memcg: memcg of the wanted lruvec
709
 * @pgdat: pglist_data
710
 *
711
 * Returns the lru list vector holding pages for a given @memcg &
712
 * @pgdat combination. This can be the node lruvec, if the memory
713
 * controller is disabled.
714
 */
715 716
static inline struct lruvec *mem_cgroup_lruvec(struct mem_cgroup *memcg,
					       struct pglist_data *pgdat)
717
{
718
	struct mem_cgroup_per_node *mz;
719 720 721
	struct lruvec *lruvec;

	if (mem_cgroup_disabled()) {
722
		lruvec = &pgdat->__lruvec;
723 724 725
		goto out;
	}

726 727 728
	if (!memcg)
		memcg = root_mem_cgroup;

729
	mz = memcg->nodeinfo[pgdat->node_id];
730 731 732 733
	lruvec = &mz->lruvec;
out:
	/*
	 * Since a node can be onlined after the mem_cgroup was created,
M
Mel Gorman 已提交
734
	 * we have to be prepared to initialize lruvec->pgdat here;
735 736
	 * and if offlined then reonlined, we need to reinitialize it.
	 */
737 738
	if (unlikely(lruvec->pgdat != pgdat))
		lruvec->pgdat = pgdat;
739 740 741
	return lruvec;
}

742 743 744 745 746 747 748 749 750 751 752 753
/**
 * 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);

754
	VM_WARN_ON_ONCE_PAGE(!memcg && !mem_cgroup_disabled(), page);
755 756
	return mem_cgroup_lruvec(memcg, pgdat);
}
757

758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773
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;
}

774
struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p);
775

776 777
struct mem_cgroup *get_mem_cgroup_from_mm(struct mm_struct *mm);

778 779 780 781 782 783 784 785 786 787 788 789 790
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
791 792 793 794 795
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
796 797 798 799 800 801 802 803 804 805
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);
}

806 807
static inline void obj_cgroup_get_many(struct obj_cgroup *objcg,
				       unsigned long nr)
R
Roman Gushchin 已提交
808
{
809
	percpu_ref_get_many(&objcg->refcnt, nr);
R
Roman Gushchin 已提交
810 811
}

812
static inline void obj_cgroup_put(struct obj_cgroup *objcg)
R
Roman Gushchin 已提交
813
{
814
	percpu_ref_put(&objcg->refcnt);
R
Roman Gushchin 已提交
815 816
}

817 818
static inline void mem_cgroup_put(struct mem_cgroup *memcg)
{
819 820
	if (memcg)
		css_put(&memcg->css);
821 822
}

823 824 825
#define mem_cgroup_from_counter(counter, member)	\
	container_of(counter, struct mem_cgroup, member)

M
Michal Hocko 已提交
826 827 828 829
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 *);
830 831
int mem_cgroup_scan_tasks(struct mem_cgroup *,
			  int (*)(struct task_struct *, void *), void *);
M
Michal Hocko 已提交
832

833 834 835 836 837
static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
{
	if (mem_cgroup_disabled())
		return 0;

838
	return memcg->id.id;
839
}
840
struct mem_cgroup *mem_cgroup_from_id(unsigned short id);
841

842 843 844 845 846
static inline struct mem_cgroup *mem_cgroup_from_seq(struct seq_file *m)
{
	return mem_cgroup_from_css(seq_css(m));
}

847 848 849 850 851 852 853 854 855 856 857
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;
}

858 859 860 861 862 863 864 865 866 867 868 869 870 871
/**
 * 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
872 873 874 875 876 877 878
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
879

880 881
static inline bool mm_match_cgroup(struct mm_struct *mm,
				   struct mem_cgroup *memcg)
882
{
883
	struct mem_cgroup *task_memcg;
884
	bool match = false;
885

886
	rcu_read_lock();
887
	task_memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
888
	if (task_memcg)
889
		match = mem_cgroup_is_descendant(task_memcg, memcg);
890
	rcu_read_unlock();
891
	return match;
892
}
893

894
struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page);
895
ino_t page_cgroup_ino(struct page *page);
896

897 898 899 900 901 902 903
static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
{
	if (mem_cgroup_disabled())
		return true;
	return !!(memcg->css.flags & CSS_ONLINE);
}

904 905 906
/*
 * For memory reclaim.
 */
907
int mem_cgroup_select_victim_node(struct mem_cgroup *memcg);
M
Michal Hocko 已提交
908 909

void mem_cgroup_update_lru_size(struct lruvec *lruvec, enum lru_list lru,
910
		int zid, int nr_pages);
M
Michal Hocko 已提交
911

912 913 914 915 916 917 918
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);
919
	return READ_ONCE(mz->lru_zone_size[zone_idx][lru]);
M
Michal Hocko 已提交
920 921
}

922 923
void mem_cgroup_handle_over_high(void);

924
unsigned long mem_cgroup_get_max(struct mem_cgroup *memcg);
925

926 927
unsigned long mem_cgroup_size(struct mem_cgroup *memcg);

928
void mem_cgroup_print_oom_context(struct mem_cgroup *memcg,
929
				struct task_struct *p);
930

931 932
void mem_cgroup_print_oom_meminfo(struct mem_cgroup *memcg);

933
static inline void mem_cgroup_enter_user_fault(void)
934
{
935 936
	WARN_ON(current->in_user_fault);
	current->in_user_fault = 1;
937 938
}

939
static inline void mem_cgroup_exit_user_fault(void)
940
{
941 942
	WARN_ON(!current->in_user_fault);
	current->in_user_fault = 0;
943 944
}

945 946
static inline bool task_in_memcg_oom(struct task_struct *p)
{
T
Tejun Heo 已提交
947
	return p->memcg_in_oom;
948 949
}

950
bool mem_cgroup_oom_synchronize(bool wait);
951 952 953
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);
954

A
Andrew Morton 已提交
955
#ifdef CONFIG_MEMCG_SWAP
956
extern bool cgroup_memory_noswap;
957
#endif
958

959
void lock_page_memcg(struct page *page);
J
Johannes Weiner 已提交
960
void unlock_page_memcg(struct page *page);
961

962
void __mod_memcg_state(struct mem_cgroup *memcg, int idx, int val);
963

964
/* idx can be of type enum memcg_stat_item or node_stat_item */
965
static inline void mod_memcg_state(struct mem_cgroup *memcg,
966
				   int idx, int val)
967
{
968 969 970
	unsigned long flags;

	local_irq_save(flags);
971
	__mod_memcg_state(memcg, idx, val);
972
	local_irq_restore(flags);
973 974
}

975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
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;
}

993 994
static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
						    enum node_stat_item idx)
995
{
996
	struct mem_cgroup_per_node *pn;
997 998
	long x = 0;
	int cpu;
999 1000 1001 1002 1003

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

	pn = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
1004 1005
	for_each_possible_cpu(cpu)
		x += per_cpu(pn->lruvec_stat_local->count[idx], cpu);
1006 1007 1008 1009 1010
#ifdef CONFIG_SMP
	if (x < 0)
		x = 0;
#endif
	return x;
1011 1012
}

1013 1014
void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
			      int val);
1015
void __mod_lruvec_kmem_state(void *p, enum node_stat_item idx, int val);
1016

1017
static inline void mod_lruvec_kmem_state(void *p, enum node_stat_item idx,
1018 1019 1020 1021 1022
					 int val)
{
	unsigned long flags;

	local_irq_save(flags);
1023
	__mod_lruvec_kmem_state(p, idx, val);
1024 1025 1026
	local_irq_restore(flags);
}

1027 1028 1029 1030 1031 1032 1033 1034 1035 1036
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);
}

1037 1038
void __count_memcg_events(struct mem_cgroup *memcg, enum vm_event_item idx,
			  unsigned long count);
1039

1040
static inline void count_memcg_events(struct mem_cgroup *memcg,
1041 1042
				      enum vm_event_item idx,
				      unsigned long count)
1043
{
1044 1045 1046
	unsigned long flags;

	local_irq_save(flags);
1047
	__count_memcg_events(memcg, idx, count);
1048
	local_irq_restore(flags);
1049 1050 1051
}

static inline void count_memcg_page_event(struct page *page,
1052
					  enum vm_event_item idx)
1053
{
1054 1055 1056 1057
	struct mem_cgroup *memcg = page_memcg(page);

	if (memcg)
		count_memcg_events(memcg, idx, 1);
1058 1059 1060 1061
}

static inline void count_memcg_event_mm(struct mm_struct *mm,
					enum vm_event_item idx)
1062
{
M
Michal Hocko 已提交
1063 1064
	struct mem_cgroup *memcg;

1065 1066
	if (mem_cgroup_disabled())
		return;
M
Michal Hocko 已提交
1067 1068 1069

	rcu_read_lock();
	memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
R
Roman Gushchin 已提交
1070
	if (likely(memcg))
1071
		count_memcg_events(memcg, idx, 1);
M
Michal Hocko 已提交
1072
	rcu_read_unlock();
1073
}
1074

1075 1076
static inline void memcg_memory_event(struct mem_cgroup *memcg,
				      enum memcg_memory_event event)
1077
{
1078 1079 1080
	bool swap_event = event == MEMCG_SWAP_HIGH || event == MEMCG_SWAP_MAX ||
			  event == MEMCG_SWAP_FAIL;

1081
	atomic_long_inc(&memcg->memory_events_local[event]);
1082 1083
	if (!swap_event)
		cgroup_file_notify(&memcg->events_local_file);
1084

1085 1086
	do {
		atomic_long_inc(&memcg->memory_events[event]);
1087 1088 1089 1090
		if (swap_event)
			cgroup_file_notify(&memcg->swap_events_file);
		else
			cgroup_file_notify(&memcg->events_file);
1091

1092 1093
		if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
			break;
1094 1095 1096 1097
		if (cgrp_dfl_root.flags & CGRP_ROOT_MEMORY_LOCAL_EVENTS)
			break;
	} while ((memcg = parent_mem_cgroup(memcg)) &&
		 !mem_cgroup_is_root(memcg));
1098 1099
}

R
Roman Gushchin 已提交
1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114
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();
}

1115
void split_page_memcg(struct page *head, unsigned int nr);
1116

1117 1118 1119 1120
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
1121
#else /* CONFIG_MEMCG */
1122 1123 1124 1125

#define MEM_CGROUP_ID_SHIFT	0
#define MEM_CGROUP_ID_MAX	0

1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141
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;
}

1142 1143 1144 1145 1146
static inline bool PageMemcgKmem(struct page *page)
{
	return false;
}

1147 1148 1149 1150 1151
static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
{
	return true;
}

1152 1153 1154 1155 1156
static inline bool mem_cgroup_disabled(void)
{
	return true;
}

1157 1158
static inline void memcg_memory_event(struct mem_cgroup *memcg,
				      enum memcg_memory_event event)
1159 1160 1161
{
}

R
Roman Gushchin 已提交
1162 1163 1164 1165 1166
static inline void memcg_memory_event_mm(struct mm_struct *mm,
					 enum memcg_memory_event event)
{
}

1167 1168
static inline unsigned long mem_cgroup_protection(struct mem_cgroup *root,
						  struct mem_cgroup *memcg,
1169
						  bool in_low_reclaim)
1170
{
1171
	return 0;
1172 1173
}

1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184
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)
1185
{
1186
	return false;
1187 1188
}

1189
static inline int mem_cgroup_charge(struct page *page, struct mm_struct *mm,
1190
				    gfp_t gfp_mask)
1191 1192 1193 1194
{
	return 0;
}

1195 1196 1197 1198 1199 1200 1201 1202 1203 1204
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)
{
}

1205
static inline void mem_cgroup_uncharge(struct page *page)
1206 1207 1208
{
}

1209
static inline void mem_cgroup_uncharge_list(struct list_head *page_list)
1210 1211 1212
{
}

1213
static inline void mem_cgroup_migrate(struct page *old, struct page *new)
1214 1215 1216
{
}

1217 1218
static inline struct lruvec *mem_cgroup_lruvec(struct mem_cgroup *memcg,
					       struct pglist_data *pgdat)
K
KAMEZAWA Hiroyuki 已提交
1219
{
1220
	return &pgdat->__lruvec;
K
KAMEZAWA Hiroyuki 已提交
1221 1222
}

1223
static inline struct lruvec *mem_cgroup_page_lruvec(struct page *page,
M
Mel Gorman 已提交
1224
						    struct pglist_data *pgdat)
1225
{
1226
	return &pgdat->__lruvec;
1227 1228
}

1229 1230 1231 1232 1233 1234 1235 1236
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;
}

1237 1238 1239 1240
static inline void lruvec_memcg_debug(struct lruvec *lruvec, struct page *page)
{
}

1241 1242 1243 1244 1245
static inline struct mem_cgroup *parent_mem_cgroup(struct mem_cgroup *memcg)
{
	return NULL;
}

1246
static inline bool mm_match_cgroup(struct mm_struct *mm,
1247
		struct mem_cgroup *memcg)
1248
{
1249
	return true;
1250 1251
}

1252 1253 1254 1255 1256
static inline struct mem_cgroup *get_mem_cgroup_from_mm(struct mm_struct *mm)
{
	return NULL;
}

1257 1258 1259 1260
static inline void mem_cgroup_put(struct mem_cgroup *memcg)
{
}

1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285
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;
}

1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298
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)
{
}

1299 1300 1301 1302 1303 1304
static inline int mem_cgroup_scan_tasks(struct mem_cgroup *memcg,
		int (*fn)(struct task_struct *, void *), void *arg)
{
	return 0;
}

1305
static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
1306
{
1307 1308 1309 1310 1311 1312 1313 1314
	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;
1315
}
1316

1317 1318 1319 1320 1321
static inline struct mem_cgroup *mem_cgroup_from_seq(struct seq_file *m)
{
	return NULL;
}

1322 1323 1324 1325 1326
static inline struct mem_cgroup *lruvec_memcg(struct lruvec *lruvec)
{
	return NULL;
}

1327
static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
1328
{
1329
	return true;
1330 1331
}

1332 1333 1334 1335 1336 1337
static inline
unsigned long mem_cgroup_get_zone_lru_size(struct lruvec *lruvec,
		enum lru_list lru, int zone_idx)
{
	return 0;
}
1338

1339
static inline unsigned long mem_cgroup_get_max(struct mem_cgroup *memcg)
1340 1341 1342 1343
{
	return 0;
}

1344 1345 1346 1347 1348
static inline unsigned long mem_cgroup_size(struct mem_cgroup *memcg)
{
	return 0;
}

1349
static inline void
1350 1351 1352 1353 1354 1355
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)
1356 1357 1358
{
}

1359
static inline void lock_page_memcg(struct page *page)
1360 1361 1362
{
}

J
Johannes Weiner 已提交
1363
static inline void unlock_page_memcg(struct page *page)
1364 1365 1366
{
}

1367 1368 1369 1370
static inline void mem_cgroup_handle_over_high(void)
{
}

1371
static inline void mem_cgroup_enter_user_fault(void)
1372 1373 1374
{
}

1375
static inline void mem_cgroup_exit_user_fault(void)
1376 1377 1378
{
}

1379 1380 1381 1382 1383
static inline bool task_in_memcg_oom(struct task_struct *p)
{
	return false;
}

1384
static inline bool mem_cgroup_oom_synchronize(bool wait)
1385 1386 1387 1388
{
	return false;
}

1389 1390 1391 1392 1393 1394 1395 1396 1397 1398
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)
{
}

1399
static inline void __mod_memcg_state(struct mem_cgroup *memcg,
1400
				     int idx,
1401
				     int nr)
1402 1403 1404
{
}

1405
static inline void mod_memcg_state(struct mem_cgroup *memcg,
1406
				   int idx,
1407
				   int nr)
1408 1409 1410
{
}

1411 1412 1413 1414 1415 1416
static inline unsigned long lruvec_page_state(struct lruvec *lruvec,
					      enum node_stat_item idx)
{
	return node_page_state(lruvec_pgdat(lruvec), idx);
}

1417 1418
static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
						    enum node_stat_item idx)
1419
{
1420
	return node_page_state(lruvec_pgdat(lruvec), idx);
1421 1422
}

1423 1424 1425 1426 1427
static inline void __mod_memcg_lruvec_state(struct lruvec *lruvec,
					    enum node_stat_item idx, int val)
{
}

1428
static inline void __mod_lruvec_kmem_state(void *p, enum node_stat_item idx,
1429 1430 1431 1432 1433 1434 1435
					   int val)
{
	struct page *page = virt_to_head_page(p);

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

1436
static inline void mod_lruvec_kmem_state(void *p, enum node_stat_item idx,
1437 1438 1439 1440 1441 1442 1443
					 int val)
{
	struct page *page = virt_to_head_page(p);

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

1444 1445 1446 1447 1448 1449
static inline void count_memcg_events(struct mem_cgroup *memcg,
				      enum vm_event_item idx,
				      unsigned long count)
{
}

1450 1451 1452 1453 1454 1455
static inline void __count_memcg_events(struct mem_cgroup *memcg,
					enum vm_event_item idx,
					unsigned long count)
{
}

1456
static inline void count_memcg_page_event(struct page *page,
1457
					  int idx)
1458 1459 1460
{
}

1461
static inline
1462
void count_memcg_event_mm(struct mm_struct *mm, enum vm_event_item idx)
1463 1464
{
}
1465

1466 1467 1468 1469 1470 1471 1472 1473
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)
1474
{
1475
	return 0;
1476
}
A
Andrew Morton 已提交
1477
#endif /* CONFIG_MEMCG */
1478

1479
static inline void __inc_lruvec_kmem_state(void *p, enum node_stat_item idx)
1480
{
1481
	__mod_lruvec_kmem_state(p, idx, 1);
1482 1483
}

1484
static inline void __dec_lruvec_kmem_state(void *p, enum node_stat_item idx)
1485
{
1486
	__mod_lruvec_kmem_state(p, idx, -1);
1487 1488
}

1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501
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));
}

1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517
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);
}

1518 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
/* 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);
}

1546
#ifdef CONFIG_CGROUP_WRITEBACK
T
Tejun Heo 已提交
1547 1548

struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb);
1549 1550 1551
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
1552

1553 1554 1555 1556 1557 1558
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)
{
1559 1560 1561
	if (mem_cgroup_disabled())
		return;

1562
	if (unlikely(&page_memcg(page)->css != wb->memcg_css))
1563 1564 1565 1566 1567
		mem_cgroup_track_foreign_dirty_slowpath(page, wb);
}

void mem_cgroup_flush_foreign(struct bdi_writeback *wb);

T
Tejun Heo 已提交
1568 1569 1570 1571 1572 1573 1574
#else	/* CONFIG_CGROUP_WRITEBACK */

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

1575
static inline void mem_cgroup_wb_stats(struct bdi_writeback *wb,
1576 1577
				       unsigned long *pfilepages,
				       unsigned long *pheadroom,
1578 1579 1580 1581 1582
				       unsigned long *pdirty,
				       unsigned long *pwriteback)
{
}

1583 1584 1585 1586 1587 1588 1589 1590 1591
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
1592
#endif	/* CONFIG_CGROUP_WRITEBACK */
1593

G
Glauber Costa 已提交
1594
struct sock;
1595 1596
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);
1597
#ifdef CONFIG_MEMCG
1598 1599
extern struct static_key_false memcg_sockets_enabled_key;
#define mem_cgroup_sockets_enabled static_branch_unlikely(&memcg_sockets_enabled_key)
1600 1601
void mem_cgroup_sk_alloc(struct sock *sk);
void mem_cgroup_sk_free(struct sock *sk);
1602
static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg)
1603
{
1604
	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && memcg->tcpmem_pressure)
1605 1606 1607 1608 1609 1610
		return true;
	do {
		if (time_before(jiffies, memcg->socket_pressure))
			return true;
	} while ((memcg = parent_mem_cgroup(memcg)));
	return false;
1611
}
1612

1613 1614
int alloc_shrinker_info(struct mem_cgroup *memcg);
void free_shrinker_info(struct mem_cgroup *memcg);
1615
void set_shrinker_bit(struct mem_cgroup *memcg, int nid, int shrinker_id);
1616
#else
1617
#define mem_cgroup_sockets_enabled 0
1618 1619
static inline void mem_cgroup_sk_alloc(struct sock *sk) { };
static inline void mem_cgroup_sk_free(struct sock *sk) { };
1620
static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg)
1621 1622 1623
{
	return false;
}
1624

1625 1626
static inline void set_shrinker_bit(struct mem_cgroup *memcg,
				    int nid, int shrinker_id)
1627 1628
{
}
1629
#endif
1630

1631
#ifdef CONFIG_MEMCG_KMEM
1632 1633
int __memcg_kmem_charge_page(struct page *page, gfp_t gfp, int order);
void __memcg_kmem_uncharge_page(struct page *page, int order);
1634

R
Roman Gushchin 已提交
1635 1636 1637 1638 1639
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);

1640
extern struct static_key_false memcg_kmem_enabled_key;
1641

1642
extern int memcg_nr_cache_ids;
1643 1644
void memcg_get_cache_ids(void);
void memcg_put_cache_ids(void);
1645 1646 1647 1648 1649 1650

/*
 * 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
 */
1651
#define for_each_memcg_cache_index(_idx)	\
1652
	for ((_idx) = 0; (_idx) < memcg_nr_cache_ids; (_idx)++)
1653

1654 1655
static inline bool memcg_kmem_enabled(void)
{
1656
	return static_branch_likely(&memcg_kmem_enabled_key);
1657 1658
}

1659 1660
static inline int memcg_kmem_charge_page(struct page *page, gfp_t gfp,
					 int order)
1661 1662
{
	if (memcg_kmem_enabled())
1663
		return __memcg_kmem_charge_page(page, gfp, order);
1664 1665 1666
	return 0;
}

1667
static inline void memcg_kmem_uncharge_page(struct page *page, int order)
1668 1669
{
	if (memcg_kmem_enabled())
1670
		__memcg_kmem_uncharge_page(page, order);
1671 1672
}

M
Michal Hocko 已提交
1673
/*
1674 1675
 * A helper for accessing memcg's kmem_id, used for getting
 * corresponding LRU lists.
M
Michal Hocko 已提交
1676 1677 1678 1679 1680
 */
static inline int memcg_cache_id(struct mem_cgroup *memcg)
{
	return memcg ? memcg->kmemcg_id : -1;
}
1681

1682 1683
struct mem_cgroup *mem_cgroup_from_obj(void *p);

1684
#else
1685

1686 1687
static inline int memcg_kmem_charge_page(struct page *page, gfp_t gfp,
					 int order)
1688 1689 1690 1691
{
	return 0;
}

1692
static inline void memcg_kmem_uncharge_page(struct page *page, int order)
1693 1694 1695
{
}

1696 1697
static inline int __memcg_kmem_charge_page(struct page *page, gfp_t gfp,
					   int order)
1698 1699 1700 1701
{
	return 0;
}

1702
static inline void __memcg_kmem_uncharge_page(struct page *page, int order)
1703 1704 1705
{
}

1706 1707 1708
#define for_each_memcg_cache_index(_idx)	\
	for (; NULL; )

1709 1710 1711 1712 1713
static inline bool memcg_kmem_enabled(void)
{
	return false;
}

1714 1715 1716 1717 1718
static inline int memcg_cache_id(struct mem_cgroup *memcg)
{
	return -1;
}

1719 1720 1721 1722 1723 1724 1725 1726
static inline void memcg_get_cache_ids(void)
{
}

static inline void memcg_put_cache_ids(void)
{
}

1727 1728 1729 1730 1731
static inline struct mem_cgroup *mem_cgroup_from_obj(void *p)
{
       return NULL;
}

1732
#endif /* CONFIG_MEMCG_KMEM */
1733

B
Balbir Singh 已提交
1734
#endif /* _LINUX_MEMCONTROL_H */