[mirror_ubuntu-hirsute-kernel.git] / mm / slab.h

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef MM_SLAB_H
#define MM_SLAB_H
/*
 * Internal slab definitions
 */

#ifdef CONFIG_SLOB
/*
 * Common fields provided in kmem_cache by all slab allocators
 * This struct is either used directly by the allocator (SLOB)
 * or the allocator must include definitions for all fields
 * provided in kmem_cache_common in their definition of kmem_cache.
 *
 * Once we can do anonymous structs (C11 standard) we could put a
 * anonymous struct definition in these allocators so that the
 * separate allocations in the kmem_cache structure of SLAB and
 * SLUB is no longer needed.
 */
struct kmem_cache {
	unsigned int object_size;/* The original size of the object */
	unsigned int size;	/* The aligned/padded/added on size  */
	unsigned int align;	/* Alignment as calculated */
	slab_flags_t flags;	/* Active flags on the slab */
	unsigned int useroffset;/* Usercopy region offset */
	unsigned int usersize;	/* Usercopy region size */
	const char *name;	/* Slab name for sysfs */
	int refcount;		/* Use counter */
	void (*ctor)(void *);	/* Called on object slot creation */
	struct list_head list;	/* List of all slab caches on the system */
};

#endif /* CONFIG_SLOB */

#ifdef CONFIG_SLAB
#include <linux/slab_def.h>
#endif

#ifdef CONFIG_SLUB
#include <linux/slub_def.h>
#endif

#include <linux/memcontrol.h>
#include <linux/fault-inject.h>
#include <linux/kasan.h>
#include <linux/kmemleak.h>
#include <linux/random.h>
#include <linux/sched/mm.h>

/*
 * State of the slab allocator.
 *
 * This is used to describe the states of the allocator during bootup.
 * Allocators use this to gradually bootstrap themselves. Most allocators
 * have the problem that the structures used for managing slab caches are
 * allocated from slab caches themselves.
 */
enum slab_state {
	DOWN,			/* No slab functionality yet */
	PARTIAL,		/* SLUB: kmem_cache_node available */
	PARTIAL_NODE,		/* SLAB: kmalloc size for node struct available */
	UP,			/* Slab caches usable but not all extras yet */
	FULL			/* Everything is working */
};

extern enum slab_state slab_state;

/* The slab cache mutex protects the management structures during changes */
extern struct mutex slab_mutex;

/* The list of all slab caches on the system */
extern struct list_head slab_caches;

/* The slab cache that manages slab cache information */
extern struct kmem_cache *kmem_cache;

/* A table of kmalloc cache names and sizes */
extern const struct kmalloc_info_struct {
	const char *name;
	unsigned int size;
} kmalloc_info[];

#ifndef CONFIG_SLOB
/* Kmalloc array related functions */
void setup_kmalloc_cache_index_table(void);
void create_kmalloc_caches(slab_flags_t);

/* Find the kmalloc slab corresponding for a certain size */
struct kmem_cache *kmalloc_slab(size_t, gfp_t);
#endif


/* Functions provided by the slab allocators */
int __kmem_cache_create(struct kmem_cache *, slab_flags_t flags);

struct kmem_cache *create_kmalloc_cache(const char *name, unsigned int size,
			slab_flags_t flags, unsigned int useroffset,
			unsigned int usersize);
extern void create_boot_cache(struct kmem_cache *, const char *name,
			unsigned int size, slab_flags_t flags,
			unsigned int useroffset, unsigned int usersize);

int slab_unmergeable(struct kmem_cache *s);
struct kmem_cache *find_mergeable(unsigned size, unsigned align,
		slab_flags_t flags, const char *name, void (*ctor)(void *));
#ifndef CONFIG_SLOB
struct kmem_cache *
__kmem_cache_alias(const char *name, unsigned int size, unsigned int align,
		   slab_flags_t flags, void (*ctor)(void *));

slab_flags_t kmem_cache_flags(unsigned int object_size,
	slab_flags_t flags, const char *name,
	void (*ctor)(void *));
#else
static inline struct kmem_cache *
__kmem_cache_alias(const char *name, unsigned int size, unsigned int align,
		   slab_flags_t flags, void (*ctor)(void *))
{ return NULL; }

static inline slab_flags_t kmem_cache_flags(unsigned int object_size,
	slab_flags_t flags, const char *name,
	void (*ctor)(void *))
{
	return flags;
}
#endif


/* Legal flag mask for kmem_cache_create(), for various configurations */
#define SLAB_CORE_FLAGS (SLAB_HWCACHE_ALIGN | SLAB_CACHE_DMA | \
			 SLAB_CACHE_DMA32 | SLAB_PANIC | \
			 SLAB_TYPESAFE_BY_RCU | SLAB_DEBUG_OBJECTS )

#if defined(CONFIG_DEBUG_SLAB)
#define SLAB_DEBUG_FLAGS (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER)
#elif defined(CONFIG_SLUB_DEBUG)
#define SLAB_DEBUG_FLAGS (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER | \
			  SLAB_TRACE | SLAB_CONSISTENCY_CHECKS)
#else
#define SLAB_DEBUG_FLAGS (0)
#endif

#if defined(CONFIG_SLAB)
#define SLAB_CACHE_FLAGS (SLAB_MEM_SPREAD | SLAB_NOLEAKTRACE | \
			  SLAB_RECLAIM_ACCOUNT | SLAB_TEMPORARY | \
			  SLAB_ACCOUNT)
#elif defined(CONFIG_SLUB)
#define SLAB_CACHE_FLAGS (SLAB_NOLEAKTRACE | SLAB_RECLAIM_ACCOUNT | \
			  SLAB_TEMPORARY | SLAB_ACCOUNT)
#else
#define SLAB_CACHE_FLAGS (0)
#endif

/* Common flags available with current configuration */
#define CACHE_CREATE_MASK (SLAB_CORE_FLAGS | SLAB_DEBUG_FLAGS | SLAB_CACHE_FLAGS)

/* Common flags permitted for kmem_cache_create */
#define SLAB_FLAGS_PERMITTED (SLAB_CORE_FLAGS | \
			      SLAB_RED_ZONE | \
			      SLAB_POISON | \
			      SLAB_STORE_USER | \
			      SLAB_TRACE | \
			      SLAB_CONSISTENCY_CHECKS | \
			      SLAB_MEM_SPREAD | \
			      SLAB_NOLEAKTRACE | \
			      SLAB_RECLAIM_ACCOUNT | \
			      SLAB_TEMPORARY | \
			      SLAB_ACCOUNT)

bool __kmem_cache_empty(struct kmem_cache *);
int __kmem_cache_shutdown(struct kmem_cache *);
void __kmem_cache_release(struct kmem_cache *);
int __kmem_cache_shrink(struct kmem_cache *);
void __kmemcg_cache_deactivate(struct kmem_cache *s);
void __kmemcg_cache_deactivate_after_rcu(struct kmem_cache *s);
void slab_kmem_cache_release(struct kmem_cache *);

struct seq_file;
struct file;

struct slabinfo {
	unsigned long active_objs;
	unsigned long num_objs;
	unsigned long active_slabs;
	unsigned long num_slabs;
	unsigned long shared_avail;
	unsigned int limit;
	unsigned int batchcount;
	unsigned int shared;
	unsigned int objects_per_slab;
	unsigned int cache_order;
};

void get_slabinfo(struct kmem_cache *s, struct slabinfo *sinfo);
void slabinfo_show_stats(struct seq_file *m, struct kmem_cache *s);
ssize_t slabinfo_write(struct file *file, const char __user *buffer,
		       size_t count, loff_t *ppos);

/*
 * Generic implementation of bulk operations
 * These are useful for situations in which the allocator cannot
 * perform optimizations. In that case segments of the object listed
 * may be allocated or freed using these operations.
 */
void __kmem_cache_free_bulk(struct kmem_cache *, size_t, void **);
int __kmem_cache_alloc_bulk(struct kmem_cache *, gfp_t, size_t, void **);

static inline int cache_vmstat_idx(struct kmem_cache *s)
{
	return (s->flags & SLAB_RECLAIM_ACCOUNT) ?
		NR_SLAB_RECLAIMABLE : NR_SLAB_UNRECLAIMABLE;
}

#ifdef CONFIG_MEMCG_KMEM

/* List of all root caches. */
extern struct list_head		slab_root_caches;
#define root_caches_node	memcg_params.__root_caches_node

/*
 * Iterate over all memcg caches of the given root cache. The caller must hold
 * slab_mutex.
 */
#define for_each_memcg_cache(iter, root) \
	list_for_each_entry(iter, &(root)->memcg_params.children, \
			    memcg_params.children_node)

static inline bool is_root_cache(struct kmem_cache *s)
{
	return !s->memcg_params.root_cache;
}

static inline bool slab_equal_or_root(struct kmem_cache *s,
				      struct kmem_cache *p)
{
	return p == s || p == s->memcg_params.root_cache;
}

/*
 * We use suffixes to the name in memcg because we can't have caches
 * created in the system with the same name. But when we print them
 * locally, better refer to them with the base name
 */
static inline const char *cache_name(struct kmem_cache *s)
{
	if (!is_root_cache(s))
		s = s->memcg_params.root_cache;
	return s->name;
}

static inline struct kmem_cache *memcg_root_cache(struct kmem_cache *s)
{
	if (is_root_cache(s))
		return s;
	return s->memcg_params.root_cache;
}

static __always_inline int memcg_charge_slab(struct page *page,
					     gfp_t gfp, int order,
					     struct kmem_cache *s)
{
	int ret;

	if (is_root_cache(s))
		return 0;

	ret = memcg_kmem_charge_memcg(page, gfp, order, s->memcg_params.memcg);
	if (ret)
		return ret;

	percpu_ref_get_many(&s->memcg_params.refcnt, 1 << order);

	return 0;
}

static __always_inline void memcg_uncharge_slab(struct page *page, int order,
						struct kmem_cache *s)
{
	if (!is_root_cache(s))
		percpu_ref_put_many(&s->memcg_params.refcnt, 1 << order);
	memcg_kmem_uncharge(page, order);
}

extern void slab_init_memcg_params(struct kmem_cache *);
extern void memcg_link_cache(struct kmem_cache *s, struct mem_cgroup *memcg);

#else /* CONFIG_MEMCG_KMEM */

/* If !memcg, all caches are root. */
#define slab_root_caches	slab_caches
#define root_caches_node	list

#define for_each_memcg_cache(iter, root) \
	for ((void)(iter), (void)(root); 0; )

static inline bool is_root_cache(struct kmem_cache *s)
{
	return true;
}

static inline bool slab_equal_or_root(struct kmem_cache *s,
				      struct kmem_cache *p)
{
	return s == p;
}

static inline const char *cache_name(struct kmem_cache *s)
{
	return s->name;
}

static inline struct kmem_cache *memcg_root_cache(struct kmem_cache *s)
{
	return s;
}

static inline int memcg_charge_slab(struct page *page, gfp_t gfp, int order,
				    struct kmem_cache *s)
{
	return 0;
}

static inline void memcg_uncharge_slab(struct page *page, int order,
				       struct kmem_cache *s)
{
}

static inline void slab_init_memcg_params(struct kmem_cache *s)
{
}

static inline void memcg_link_cache(struct kmem_cache *s,
				    struct mem_cgroup *memcg)
{
}

#endif /* CONFIG_MEMCG_KMEM */

static inline struct kmem_cache *virt_to_cache(const void *obj)
{
	struct page *page;

	page = virt_to_head_page(obj);
	if (WARN_ONCE(!PageSlab(page), "%s: Object is not a Slab page!\n",
					__func__))
		return NULL;
	return page->slab_cache;
}

static __always_inline int charge_slab_page(struct page *page,
					    gfp_t gfp, int order,
					    struct kmem_cache *s)
{
	int ret = memcg_charge_slab(page, gfp, order, s);

	if (!ret)
		mod_lruvec_page_state(page, cache_vmstat_idx(s), 1 << order);

	return ret;
}

static __always_inline void uncharge_slab_page(struct page *page, int order,
					       struct kmem_cache *s)
{
	mod_lruvec_page_state(page, cache_vmstat_idx(s), -(1 << order));
	memcg_uncharge_slab(page, order, s);
}

static inline struct kmem_cache *cache_from_obj(struct kmem_cache *s, void *x)
{
	struct kmem_cache *cachep;

	/*
	 * When kmemcg is not being used, both assignments should return the
	 * same value. but we don't want to pay the assignment price in that
	 * case. If it is not compiled in, the compiler should be smart enough
	 * to not do even the assignment. In that case, slab_equal_or_root
	 * will also be a constant.
	 */
	if (!memcg_kmem_enabled() &&
	    !IS_ENABLED(CONFIG_SLAB_FREELIST_HARDENED) &&
	    !unlikely(s->flags & SLAB_CONSISTENCY_CHECKS))
		return s;

	cachep = virt_to_cache(x);
	WARN_ONCE(cachep && !slab_equal_or_root(cachep, s),
		  "%s: Wrong slab cache. %s but object is from %s\n",
		  __func__, s->name, cachep->name);
	return cachep;
}

static inline size_t slab_ksize(const struct kmem_cache *s)
{
#ifndef CONFIG_SLUB
	return s->object_size;

#else /* CONFIG_SLUB */
# ifdef CONFIG_SLUB_DEBUG
	/*
	 * Debugging requires use of the padding between object
	 * and whatever may come after it.
	 */
	if (s->flags & (SLAB_RED_ZONE | SLAB_POISON))
		return s->object_size;
# endif
	if (s->flags & SLAB_KASAN)
		return s->object_size;
	/*
	 * If we have the need to store the freelist pointer
	 * back there or track user information then we can
	 * only use the space before that information.
	 */
	if (s->flags & (SLAB_TYPESAFE_BY_RCU | SLAB_STORE_USER))
		return s->inuse;
	/*
	 * Else we can use all the padding etc for the allocation
	 */
	return s->size;
#endif
}

static inline struct kmem_cache *slab_pre_alloc_hook(struct kmem_cache *s,
						     gfp_t flags)
{
	flags &= gfp_allowed_mask;

	fs_reclaim_acquire(flags);
	fs_reclaim_release(flags);

	might_sleep_if(gfpflags_allow_blocking(flags));

	if (should_failslab(s, flags))
		return NULL;

	if (memcg_kmem_enabled() &&
	    ((flags & __GFP_ACCOUNT) || (s->flags & SLAB_ACCOUNT)))
		return memcg_kmem_get_cache(s);

	return s;
}

static inline void slab_post_alloc_hook(struct kmem_cache *s, gfp_t flags,
					size_t size, void **p)
{
	size_t i;

	flags &= gfp_allowed_mask;
	for (i = 0; i < size; i++) {
		p[i] = kasan_slab_alloc(s, p[i], flags);
		/* As p[i] might get tagged, call kmemleak hook after KASAN. */
		kmemleak_alloc_recursive(p[i], s->object_size, 1,
					 s->flags, flags);
	}

	if (memcg_kmem_enabled())
		memcg_kmem_put_cache(s);
}

#ifndef CONFIG_SLOB
/*
 * The slab lists for all objects.
 */
struct kmem_cache_node {
	spinlock_t list_lock;

#ifdef CONFIG_SLAB
	struct list_head slabs_partial;	/* partial list first, better asm code */
	struct list_head slabs_full;
	struct list_head slabs_free;
	unsigned long total_slabs;	/* length of all slab lists */
	unsigned long free_slabs;	/* length of free slab list only */
	unsigned long free_objects;
	unsigned int free_limit;
	unsigned int colour_next;	/* Per-node cache coloring */
	struct array_cache *shared;	/* shared per node */
	struct alien_cache **alien;	/* on other nodes */
	unsigned long next_reap;	/* updated without locking */
	int free_touched;		/* updated without locking */
#endif

#ifdef CONFIG_SLUB
	unsigned long nr_partial;
	struct list_head partial;
#ifdef CONFIG_SLUB_DEBUG
	atomic_long_t nr_slabs;
	atomic_long_t total_objects;
	struct list_head full;
#endif
#endif

};

static inline struct kmem_cache_node *get_node(struct kmem_cache *s, int node)
{
	return s->node[node];
}

/*
 * Iterator over all nodes. The body will be executed for each node that has
 * a kmem_cache_node structure allocated (which is true for all online nodes)
 */
#define for_each_kmem_cache_node(__s, __node, __n) \
	for (__node = 0; __node < nr_node_ids; __node++) \
		 if ((__n = get_node(__s, __node)))

#endif

void *slab_start(struct seq_file *m, loff_t *pos);
void *slab_next(struct seq_file *m, void *p, loff_t *pos);
void slab_stop(struct seq_file *m, void *p);
void *memcg_slab_start(struct seq_file *m, loff_t *pos);
void *memcg_slab_next(struct seq_file *m, void *p, loff_t *pos);
void memcg_slab_stop(struct seq_file *m, void *p);
int memcg_slab_show(struct seq_file *m, void *p);

#if defined(CONFIG_SLAB) || defined(CONFIG_SLUB_DEBUG)
void dump_unreclaimable_slab(void);
#else
static inline void dump_unreclaimable_slab(void)
{
}
#endif

void ___cache_free(struct kmem_cache *cache, void *x, unsigned long addr);

#ifdef CONFIG_SLAB_FREELIST_RANDOM
int cache_random_seq_create(struct kmem_cache *cachep, unsigned int count,
			gfp_t gfp);
void cache_random_seq_destroy(struct kmem_cache *cachep);
#else
static inline int cache_random_seq_create(struct kmem_cache *cachep,
					unsigned int count, gfp_t gfp)
{
	return 0;
}
static inline void cache_random_seq_destroy(struct kmem_cache *cachep) { }
#endif /* CONFIG_SLAB_FREELIST_RANDOM */

#endif /* MM_SLAB_H */
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
97d06609 CL	2	#ifndef MM_SLAB_H
	3	#define MM_SLAB_H
	4	/*
	5	* Internal slab definitions
	6	*/
	7
07f361b2 JK	8	#ifdef CONFIG_SLOB
	9	/*
	10	* Common fields provided in kmem_cache by all slab allocators
	11	* This struct is either used directly by the allocator (SLOB)
	12	* or the allocator must include definitions for all fields
	13	* provided in kmem_cache_common in their definition of kmem_cache.
	14	*
	15	* Once we can do anonymous structs (C11 standard) we could put a
	16	* anonymous struct definition in these allocators so that the
	17	* separate allocations in the kmem_cache structure of SLAB and
	18	* SLUB is no longer needed.
	19	*/
	20	struct kmem_cache {
	21	unsigned int object_size;/* The original size of the object */
	22	unsigned int size; /* The aligned/padded/added on size */
	23	unsigned int align; /* Alignment as calculated */
d50112ed	24	slab_flags_t flags; /* Active flags on the slab */
7bbdb81e AD	25	unsigned int useroffset;/* Usercopy region offset */
7bbdb81e AD	26	unsigned int usersize; /* Usercopy region size */
07f361b2 JK	27	const char name; / Slab name for sysfs */
	28	int refcount; /* Use counter */
	29	void (ctor)(void ); /* Called on object slot creation */
	30	struct list_head list; /* List of all slab caches on the system */
	31	};
	32
	33	#endif /* CONFIG_SLOB */
	34
	35	#ifdef CONFIG_SLAB
	36	#include <linux/slab_def.h>
	37	#endif
	38
	39	#ifdef CONFIG_SLUB
	40	#include <linux/slub_def.h>
	41	#endif
	42
	43	#include <linux/memcontrol.h>
11c7aec2	44	#include <linux/fault-inject.h>
11c7aec2 JDB	45	#include <linux/kasan.h>
11c7aec2 JDB	46	#include <linux/kmemleak.h>
7c00fce9	47	#include <linux/random.h>
d92a8cfc	48	#include <linux/sched/mm.h>
07f361b2	49
97d06609 CL	50	/*
	51	* State of the slab allocator.
	52	*
	53	* This is used to describe the states of the allocator during bootup.
	54	* Allocators use this to gradually bootstrap themselves. Most allocators
	55	* have the problem that the structures used for managing slab caches are
	56	* allocated from slab caches themselves.
	57	*/
	58	enum slab_state {
	59	DOWN, /* No slab functionality yet */
	60	PARTIAL, /* SLUB: kmem_cache_node available */
ce8eb6c4	61	PARTIAL_NODE, /* SLAB: kmalloc size for node struct available */
97d06609 CL	62	UP, /* Slab caches usable but not all extras yet */
	63	FULL /* Everything is working */
	64	};
	65
	66	extern enum slab_state slab_state;
	67
18004c5d CL	68	/* The slab cache mutex protects the management structures during changes */
18004c5d CL	69	extern struct mutex slab_mutex;
9b030cb8 CL	70
9b030cb8 CL	71	/* The list of all slab caches on the system */
18004c5d CL	72	extern struct list_head slab_caches;
18004c5d CL	73
9b030cb8 CL	74	/* The slab cache that manages slab cache information */
	75	extern struct kmem_cache *kmem_cache;
	76
af3b5f87 VB	77	/* A table of kmalloc cache names and sizes */
	78	extern const struct kmalloc_info_struct {
	79	const char *name;
55de8b9c	80	unsigned int size;
af3b5f87 VB	81	} kmalloc_info[];
af3b5f87 VB	82
f97d5f63 CL	83	#ifndef CONFIG_SLOB
f97d5f63 CL	84	/* Kmalloc array related functions */
34cc6990	85	void setup_kmalloc_cache_index_table(void);
d50112ed	86	void create_kmalloc_caches(slab_flags_t);
2c59dd65 CL	87
	88	/* Find the kmalloc slab corresponding for a certain size */
	89	struct kmem_cache *kmalloc_slab(size_t, gfp_t);
f97d5f63 CL	90	#endif
	91
	92
9b030cb8	93	/* Functions provided by the slab allocators */
d50112ed	94	int __kmem_cache_create(struct kmem_cache *, slab_flags_t flags);
97d06609	95
55de8b9c AD	96	struct kmem_cache create_kmalloc_cache(const char name, unsigned int size,
	97	slab_flags_t flags, unsigned int useroffset,
	98	unsigned int usersize);
45530c44	99	extern void create_boot_cache(struct kmem_cache , const char name,
361d575e AD	100	unsigned int size, slab_flags_t flags,
361d575e AD	101	unsigned int useroffset, unsigned int usersize);
45530c44	102
423c929c	103	int slab_unmergeable(struct kmem_cache *s);
f4957d5b	104	struct kmem_cache *find_mergeable(unsigned size, unsigned align,
d50112ed	105	slab_flags_t flags, const char name, void (ctor)(void *));
12220dea	106	#ifndef CONFIG_SLOB
2633d7a0	107	struct kmem_cache *
f4957d5b	108	__kmem_cache_alias(const char *name, unsigned int size, unsigned int align,
d50112ed	109	slab_flags_t flags, void (ctor)(void ));
423c929c	110
0293d1fd	111	slab_flags_t kmem_cache_flags(unsigned int object_size,
d50112ed	112	slab_flags_t flags, const char *name,
423c929c	113	void (ctor)(void ));
cbb79694	114	#else
2633d7a0	115	static inline struct kmem_cache *
f4957d5b	116	__kmem_cache_alias(const char *name, unsigned int size, unsigned int align,
d50112ed	117	slab_flags_t flags, void (ctor)(void ))
cbb79694	118	{ return NULL; }
423c929c	119
0293d1fd	120	static inline slab_flags_t kmem_cache_flags(unsigned int object_size,
d50112ed	121	slab_flags_t flags, const char *name,
423c929c JK	122	void (ctor)(void ))
	123	{
	124	return flags;
	125	}
cbb79694 CL	126	#endif
	127
	128
d8843922	129	/* Legal flag mask for kmem_cache_create(), for various configurations */
6d6ea1e9 NB	130	#define SLAB_CORE_FLAGS (SLAB_HWCACHE_ALIGN \| SLAB_CACHE_DMA \| \
6d6ea1e9 NB	131	SLAB_CACHE_DMA32 \| SLAB_PANIC \| \
5f0d5a3a	132	SLAB_TYPESAFE_BY_RCU \| SLAB_DEBUG_OBJECTS )
d8843922 GC	133
	134	#if defined(CONFIG_DEBUG_SLAB)
	135	#define SLAB_DEBUG_FLAGS (SLAB_RED_ZONE \| SLAB_POISON \| SLAB_STORE_USER)
	136	#elif defined(CONFIG_SLUB_DEBUG)
	137	#define SLAB_DEBUG_FLAGS (SLAB_RED_ZONE \| SLAB_POISON \| SLAB_STORE_USER \| \
becfda68	138	SLAB_TRACE \| SLAB_CONSISTENCY_CHECKS)
d8843922 GC	139	#else
	140	#define SLAB_DEBUG_FLAGS (0)
	141	#endif
	142
	143	#if defined(CONFIG_SLAB)
	144	#define SLAB_CACHE_FLAGS (SLAB_MEM_SPREAD \| SLAB_NOLEAKTRACE \| \
230e9fc2	145	SLAB_RECLAIM_ACCOUNT \| SLAB_TEMPORARY \| \
75f296d9	146	SLAB_ACCOUNT)
d8843922 GC	147	#elif defined(CONFIG_SLUB)
d8843922 GC	148	#define SLAB_CACHE_FLAGS (SLAB_NOLEAKTRACE \| SLAB_RECLAIM_ACCOUNT \| \
75f296d9	149	SLAB_TEMPORARY \| SLAB_ACCOUNT)
d8843922 GC	150	#else
	151	#define SLAB_CACHE_FLAGS (0)
	152	#endif
	153
e70954fd	154	/* Common flags available with current configuration */
d8843922 GC	155	#define CACHE_CREATE_MASK (SLAB_CORE_FLAGS \| SLAB_DEBUG_FLAGS \| SLAB_CACHE_FLAGS)
d8843922 GC	156
e70954fd TG	157	/* Common flags permitted for kmem_cache_create */
	158	#define SLAB_FLAGS_PERMITTED (SLAB_CORE_FLAGS \| \
	159	SLAB_RED_ZONE \| \
	160	SLAB_POISON \| \
	161	SLAB_STORE_USER \| \
	162	SLAB_TRACE \| \
	163	SLAB_CONSISTENCY_CHECKS \| \
	164	SLAB_MEM_SPREAD \| \
	165	SLAB_NOLEAKTRACE \| \
	166	SLAB_RECLAIM_ACCOUNT \| \
	167	SLAB_TEMPORARY \| \
e70954fd TG	168	SLAB_ACCOUNT)
e70954fd TG	169
f9e13c0a	170	bool __kmem_cache_empty(struct kmem_cache *);
945cf2b6	171	int __kmem_cache_shutdown(struct kmem_cache *);
52b4b950	172	void __kmem_cache_release(struct kmem_cache *);
c9fc5864 TH	173	int __kmem_cache_shrink(struct kmem_cache *);
c9fc5864 TH	174	void __kmemcg_cache_deactivate(struct kmem_cache *s);
43486694	175	void __kmemcg_cache_deactivate_after_rcu(struct kmem_cache *s);
41a21285	176	void slab_kmem_cache_release(struct kmem_cache *);
945cf2b6	177
b7454ad3 GC	178	struct seq_file;
b7454ad3 GC	179	struct file;
b7454ad3	180
0d7561c6 GC	181	struct slabinfo {
	182	unsigned long active_objs;
	183	unsigned long num_objs;
	184	unsigned long active_slabs;
	185	unsigned long num_slabs;
	186	unsigned long shared_avail;
	187	unsigned int limit;
	188	unsigned int batchcount;
	189	unsigned int shared;
	190	unsigned int objects_per_slab;
	191	unsigned int cache_order;
	192	};
	193
	194	void get_slabinfo(struct kmem_cache s, struct slabinfo sinfo);
	195	void slabinfo_show_stats(struct seq_file m, struct kmem_cache s);
b7454ad3 GC	196	ssize_t slabinfo_write(struct file file, const char __user buffer,
b7454ad3 GC	197	size_t count, loff_t *ppos);
ba6c496e	198
484748f0 CL	199	/*
	200	* Generic implementation of bulk operations
	201	* These are useful for situations in which the allocator cannot
9f706d68	202	* perform optimizations. In that case segments of the object listed
484748f0 CL	203	* may be allocated or freed using these operations.
	204	*/
	205	void __kmem_cache_free_bulk(struct kmem_cache , size_t, void *);
865762a8	206	int __kmem_cache_alloc_bulk(struct kmem_cache , gfp_t, size_t, void *);
484748f0	207
6cea1d56 RG	208	static inline int cache_vmstat_idx(struct kmem_cache *s)
	209	{
	210	return (s->flags & SLAB_RECLAIM_ACCOUNT) ?
	211	NR_SLAB_RECLAIMABLE : NR_SLAB_UNRECLAIMABLE;
	212	}
	213
84c07d11	214	#ifdef CONFIG_MEMCG_KMEM
510ded33 TH	215
	216	/* List of all root caches. */
	217	extern struct list_head slab_root_caches;
	218	#define root_caches_node memcg_params.__root_caches_node
	219
426589f5 VD	220	/*
	221	* Iterate over all memcg caches of the given root cache. The caller must hold
	222	* slab_mutex.
	223	*/
	224	#define for_each_memcg_cache(iter, root) \
9eeadc8b TH	225	list_for_each_entry(iter, &(root)->memcg_params.children, \
9eeadc8b TH	226	memcg_params.children_node)
426589f5	227
ba6c496e GC	228	static inline bool is_root_cache(struct kmem_cache *s)
ba6c496e GC	229	{
9eeadc8b	230	return !s->memcg_params.root_cache;
ba6c496e	231	}
2633d7a0	232
b9ce5ef4	233	static inline bool slab_equal_or_root(struct kmem_cache *s,
f7ce3190	234	struct kmem_cache *p)
b9ce5ef4	235	{
f7ce3190	236	return p == s \|\| p == s->memcg_params.root_cache;
b9ce5ef4	237	}
749c5415 GC	238
	239	/*
	240	* We use suffixes to the name in memcg because we can't have caches
	241	* created in the system with the same name. But when we print them
	242	* locally, better refer to them with the base name
	243	*/
	244	static inline const char cache_name(struct kmem_cache s)
	245	{
	246	if (!is_root_cache(s))
f7ce3190	247	s = s->memcg_params.root_cache;
749c5415 GC	248	return s->name;
	249	}
	250
943a451a GC	251	static inline struct kmem_cache memcg_root_cache(struct kmem_cache s)
	252	{
	253	if (is_root_cache(s))
	254	return s;
f7ce3190	255	return s->memcg_params.root_cache;
943a451a	256	}
5dfb4175	257
f3ccb2c4 VD	258	static __always_inline int memcg_charge_slab(struct page *page,
	259	gfp_t gfp, int order,
	260	struct kmem_cache *s)
5dfb4175	261	{
f0a3a24b RG	262	int ret;
f0a3a24b RG	263
5dfb4175 VD	264	if (is_root_cache(s))
5dfb4175 VD	265	return 0;
f0a3a24b RG	266
	267	ret = memcg_kmem_charge_memcg(page, gfp, order, s->memcg_params.memcg);
	268	if (ret)
	269	return ret;
	270
	271	percpu_ref_get_many(&s->memcg_params.refcnt, 1 << order);
	272
	273	return 0;
27ee57c9 VD	274	}
	275
	276	static __always_inline void memcg_uncharge_slab(struct page *page, int order,
	277	struct kmem_cache *s)
	278	{
f0a3a24b RG	279	if (!is_root_cache(s))
f0a3a24b RG	280	percpu_ref_put_many(&s->memcg_params.refcnt, 1 << order);
27ee57c9	281	memcg_kmem_uncharge(page, order);
5dfb4175	282	}
f7ce3190 VD	283
f7ce3190 VD	284	extern void slab_init_memcg_params(struct kmem_cache *);
c03914b7	285	extern void memcg_link_cache(struct kmem_cache s, struct mem_cgroup memcg);
f7ce3190	286
84c07d11	287	#else /* CONFIG_MEMCG_KMEM */
f7ce3190	288
510ded33 TH	289	/* If !memcg, all caches are root. */
	290	#define slab_root_caches slab_caches
	291	#define root_caches_node list
	292
426589f5 VD	293	#define for_each_memcg_cache(iter, root) \
426589f5 VD	294	for ((void)(iter), (void)(root); 0; )
426589f5	295
ba6c496e GC	296	static inline bool is_root_cache(struct kmem_cache *s)
	297	{
	298	return true;
	299	}
	300
b9ce5ef4 GC	301	static inline bool slab_equal_or_root(struct kmem_cache *s,
	302	struct kmem_cache *p)
	303	{
598a0717	304	return s == p;
b9ce5ef4	305	}
749c5415 GC	306
	307	static inline const char cache_name(struct kmem_cache s)
	308	{
	309	return s->name;
	310	}
	311
943a451a GC	312	static inline struct kmem_cache memcg_root_cache(struct kmem_cache s)
	313	{
	314	return s;
	315	}
5dfb4175	316
f3ccb2c4 VD	317	static inline int memcg_charge_slab(struct page *page, gfp_t gfp, int order,
f3ccb2c4 VD	318	struct kmem_cache *s)
5dfb4175 VD	319	{
	320	return 0;
	321	}
	322
27ee57c9 VD	323	static inline void memcg_uncharge_slab(struct page *page, int order,
	324	struct kmem_cache *s)
	325	{
	326	}
	327
f7ce3190 VD	328	static inline void slab_init_memcg_params(struct kmem_cache *s)
	329	{
	330	}
510ded33	331
c03914b7 RG	332	static inline void memcg_link_cache(struct kmem_cache *s,
c03914b7 RG	333	struct mem_cgroup *memcg)
510ded33 TH	334	{
	335	}
	336
84c07d11	337	#endif /* CONFIG_MEMCG_KMEM */
b9ce5ef4	338
a64b5378 KC	339	static inline struct kmem_cache virt_to_cache(const void obj)
	340	{
	341	struct page *page;
	342
	343	page = virt_to_head_page(obj);
	344	if (WARN_ONCE(!PageSlab(page), "%s: Object is not a Slab page!\n",
	345	__func__))
	346	return NULL;
	347	return page->slab_cache;
	348	}
	349
6cea1d56 RG	350	static __always_inline int charge_slab_page(struct page *page,
	351	gfp_t gfp, int order,
	352	struct kmem_cache *s)
	353	{
	354	int ret = memcg_charge_slab(page, gfp, order, s);
	355
	356	if (!ret)
	357	mod_lruvec_page_state(page, cache_vmstat_idx(s), 1 << order);
	358
	359	return ret;
	360	}
	361
	362	static __always_inline void uncharge_slab_page(struct page *page, int order,
	363	struct kmem_cache *s)
	364	{
	365	mod_lruvec_page_state(page, cache_vmstat_idx(s), -(1 << order));
	366	memcg_uncharge_slab(page, order, s);
	367	}
	368
b9ce5ef4 GC	369	static inline struct kmem_cache cache_from_obj(struct kmem_cache s, void *x)
	370	{
	371	struct kmem_cache *cachep;
b9ce5ef4 GC	372
	373	/*
	374	* When kmemcg is not being used, both assignments should return the
	375	* same value. but we don't want to pay the assignment price in that
	376	* case. If it is not compiled in, the compiler should be smart enough
	377	* to not do even the assignment. In that case, slab_equal_or_root
	378	* will also be a constant.
	379	*/
becfda68	380	if (!memcg_kmem_enabled() &&
598a0717	381	!IS_ENABLED(CONFIG_SLAB_FREELIST_HARDENED) &&
becfda68	382	!unlikely(s->flags & SLAB_CONSISTENCY_CHECKS))
b9ce5ef4 GC	383	return s;
b9ce5ef4 GC	384
a64b5378 KC	385	cachep = virt_to_cache(x);
a64b5378 KC	386	WARN_ONCE(cachep && !slab_equal_or_root(cachep, s),
598a0717 KC	387	"%s: Wrong slab cache. %s but object is from %s\n",
	388	__func__, s->name, cachep->name);
	389	return cachep;
b9ce5ef4	390	}
ca34956b	391
11c7aec2 JDB	392	static inline size_t slab_ksize(const struct kmem_cache *s)
	393	{
	394	#ifndef CONFIG_SLUB
	395	return s->object_size;
	396
	397	#else /* CONFIG_SLUB */
	398	# ifdef CONFIG_SLUB_DEBUG
	399	/*
	400	* Debugging requires use of the padding between object
	401	* and whatever may come after it.
	402	*/
	403	if (s->flags & (SLAB_RED_ZONE \| SLAB_POISON))
	404	return s->object_size;
	405	# endif
80a9201a AP	406	if (s->flags & SLAB_KASAN)
80a9201a AP	407	return s->object_size;
11c7aec2 JDB	408	/*
	409	* If we have the need to store the freelist pointer
	410	* back there or track user information then we can
	411	* only use the space before that information.
	412	*/
5f0d5a3a	413	if (s->flags & (SLAB_TYPESAFE_BY_RCU \| SLAB_STORE_USER))
11c7aec2 JDB	414	return s->inuse;
	415	/*
	416	* Else we can use all the padding etc for the allocation
	417	*/
	418	return s->size;
	419	#endif
	420	}
	421
	422	static inline struct kmem_cache slab_pre_alloc_hook(struct kmem_cache s,
	423	gfp_t flags)
	424	{
	425	flags &= gfp_allowed_mask;
d92a8cfc PZ	426
	427	fs_reclaim_acquire(flags);
	428	fs_reclaim_release(flags);
	429
11c7aec2 JDB	430	might_sleep_if(gfpflags_allow_blocking(flags));
11c7aec2 JDB	431
fab9963a	432	if (should_failslab(s, flags))
11c7aec2 JDB	433	return NULL;
11c7aec2 JDB	434
45264778 VD	435	if (memcg_kmem_enabled() &&
	436	((flags & __GFP_ACCOUNT) \|\| (s->flags & SLAB_ACCOUNT)))
	437	return memcg_kmem_get_cache(s);
	438
	439	return s;
11c7aec2 JDB	440	}
	441
	442	static inline void slab_post_alloc_hook(struct kmem_cache *s, gfp_t flags,
	443	size_t size, void **p)
	444	{
	445	size_t i;
	446
	447	flags &= gfp_allowed_mask;
	448	for (i = 0; i < size; i++) {
53128245	449	p[i] = kasan_slab_alloc(s, p[i], flags);
a2f77575	450	/* As p[i] might get tagged, call kmemleak hook after KASAN. */
53128245	451	kmemleak_alloc_recursive(p[i], s->object_size, 1,
11c7aec2	452	s->flags, flags);
11c7aec2	453	}
45264778 VD	454
	455	if (memcg_kmem_enabled())
	456	memcg_kmem_put_cache(s);
11c7aec2 JDB	457	}
11c7aec2 JDB	458
44c5356f	459	#ifndef CONFIG_SLOB
ca34956b CL	460	/*
	461	* The slab lists for all objects.
	462	*/
	463	struct kmem_cache_node {
	464	spinlock_t list_lock;
	465
	466	#ifdef CONFIG_SLAB
	467	struct list_head slabs_partial; /* partial list first, better asm code */
	468	struct list_head slabs_full;
	469	struct list_head slabs_free;
bf00bd34 DR	470	unsigned long total_slabs; /* length of all slab lists */
bf00bd34 DR	471	unsigned long free_slabs; /* length of free slab list only */
ca34956b CL	472	unsigned long free_objects;
	473	unsigned int free_limit;
	474	unsigned int colour_next; /* Per-node cache coloring */
	475	struct array_cache shared; / shared per node */
c8522a3a	476	struct alien_cache *alien; / on other nodes */
ca34956b CL	477	unsigned long next_reap; /* updated without locking */
	478	int free_touched; /* updated without locking */
	479	#endif
	480
	481	#ifdef CONFIG_SLUB
	482	unsigned long nr_partial;
	483	struct list_head partial;
	484	#ifdef CONFIG_SLUB_DEBUG
	485	atomic_long_t nr_slabs;
	486	atomic_long_t total_objects;
	487	struct list_head full;
	488	#endif
	489	#endif
	490
	491	};
e25839f6	492
44c5356f CL	493	static inline struct kmem_cache_node get_node(struct kmem_cache s, int node)
	494	{
	495	return s->node[node];
	496	}
	497
	498	/*
	499	* Iterator over all nodes. The body will be executed for each node that has
	500	* a kmem_cache_node structure allocated (which is true for all online nodes)
	501	*/
	502	#define for_each_kmem_cache_node(__s, __node, __n) \
9163582c MP	503	for (__node = 0; __node < nr_node_ids; __node++) \
9163582c MP	504	if ((__n = get_node(__s, __node)))
44c5356f CL	505
	506	#endif
	507
1df3b26f	508	void slab_start(struct seq_file m, loff_t *pos);
276a2439 WL	509	void slab_next(struct seq_file m, void p, loff_t pos);
276a2439 WL	510	void slab_stop(struct seq_file m, void p);
bc2791f8 TH	511	void memcg_slab_start(struct seq_file m, loff_t *pos);
	512	void memcg_slab_next(struct seq_file m, void p, loff_t pos);
	513	void memcg_slab_stop(struct seq_file m, void p);
b047501c	514	int memcg_slab_show(struct seq_file m, void p);
5240ab40	515
852d8be0 YS	516	#if defined(CONFIG_SLAB) \|\| defined(CONFIG_SLUB_DEBUG)
	517	void dump_unreclaimable_slab(void);
	518	#else
	519	static inline void dump_unreclaimable_slab(void)
	520	{
	521	}
	522	#endif
	523
55834c59 AP	524	void ___cache_free(struct kmem_cache cache, void x, unsigned long addr);
55834c59 AP	525
7c00fce9 TG	526	#ifdef CONFIG_SLAB_FREELIST_RANDOM
	527	int cache_random_seq_create(struct kmem_cache *cachep, unsigned int count,
	528	gfp_t gfp);
	529	void cache_random_seq_destroy(struct kmem_cache *cachep);
	530	#else
	531	static inline int cache_random_seq_create(struct kmem_cache *cachep,
	532	unsigned int count, gfp_t gfp)
	533	{
	534	return 0;
	535	}
	536	static inline void cache_random_seq_destroy(struct kmem_cache *cachep) { }
	537	#endif /* CONFIG_SLAB_FREELIST_RANDOM */
	538
5240ab40	539	#endif /* MM_SLAB_H */